"
+ ],
+ "image/png": "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\n"
+ },
+ "metadata": {}
+ },
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "---------------------------------------------------\n",
+ "Number of Classes: 3\n"
+ ]
+ }
+ ]
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "# After train_ds is defined\n",
+ "for image_batch, label_batch in train_ds.take(1):\n",
+ " print(\"Image batch shape:\", image_batch.shape) # full batch shape\n",
+ " print(\"Label batch shape:\", label_batch.shape) # labels shape\n",
+ "\n",
+ " input_shape = image_batch.shape[1:] # shape of a single image\n",
+ " print(\"Single image shape:\", input_shape)\n",
+ " break"
+ ],
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "id": "ujmm5rl6zDX3",
+ "outputId": "be9ebfc7-7642-4874-b8e4-126177b7a558"
+ },
+ "execution_count": 16,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "Image batch shape: (32, 256, 256, 3)\n",
+ "Label batch shape: (32,)\n",
+ "Single image shape: (256, 256, 3)\n"
+ ]
+ }
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "source": [
+ "# **Model Loading**"
+ ],
+ "metadata": {
+ "id": "IDQlN3IGwR7J"
+ }
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "inception = InceptionV3(input_shape=input_shape, weights='imagenet', include_top=False)\n",
+ "\n",
+ "# don't train existing weights\n",
+ "for layer in inception.layers:\n",
+ " layer.trainable = False\n",
+ "\n",
+ "# Number of classes\n",
+ "print(\"Number of Classes: \", len(le.classes_))"
+ ],
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "id": "zWI-WJAziUUO",
+ "outputId": "05f06abf-78f2-4698-bf82-92f35283aaaa"
+ },
+ "execution_count": 17,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/inception_v3/inception_v3_weights_tf_dim_ordering_tf_kernels_notop.h5\n",
+ "\u001b[1m87910968/87910968\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 0us/step\n",
+ "Number of Classes: 3\n"
+ ]
+ }
+ ]
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "x = GlobalAveragePooling2D()(inception.output)\n",
+ "x = Dense(512, activation='relu')(x)\n",
+ "x = Dropout(0.5)(x)\n",
+ "prediction = Dense(len(le.classes_), activation='softmax')(x)\n",
+ "\n",
+ "# create a model object\n",
+ "model = Model(inputs=inception.input, outputs=prediction)\n",
+ "\n",
+ "# view the structure of the model\n",
+ "model.summary()"
+ ],
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/",
+ "height": 1000
+ },
+ "id": "hKHB7kzkiUhB",
+ "outputId": "4fb8d372-df67-4f8a-e2fb-13ce852a9b2f"
+ },
+ "execution_count": 18,
+ "outputs": [
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ "\u001b[1mModel: \"functional\"\u001b[0m\n"
+ ],
+ "text/html": [
+ "Model: \"functional\"\n",
+ "
\n"
+ ]
+ },
+ "metadata": {}
+ },
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ "┏━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┓\n",
+ "┃\u001b[1m \u001b[0m\u001b[1mLayer (type) \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m Param #\u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mConnected to \u001b[0m\u001b[1m \u001b[0m┃\n",
+ "┡━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━┩\n",
+ "│ input_layer │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m, \u001b[38;5;34m256\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ - │\n",
+ "│ (\u001b[38;5;33mInputLayer\u001b[0m) │ \u001b[38;5;34m3\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m127\u001b[0m, \u001b[38;5;34m127\u001b[0m, │ \u001b[38;5;34m864\u001b[0m │ input_layer[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m32\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalization │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m127\u001b[0m, \u001b[38;5;34m127\u001b[0m, │ \u001b[38;5;34m96\u001b[0m │ conv2d[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m32\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m127\u001b[0m, \u001b[38;5;34m127\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m32\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_1 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m125\u001b[0m, \u001b[38;5;34m125\u001b[0m, │ \u001b[38;5;34m9,216\u001b[0m │ activation[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m32\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m125\u001b[0m, \u001b[38;5;34m125\u001b[0m, │ \u001b[38;5;34m96\u001b[0m │ conv2d_1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m32\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_1 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m125\u001b[0m, \u001b[38;5;34m125\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m32\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_2 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m125\u001b[0m, \u001b[38;5;34m125\u001b[0m, │ \u001b[38;5;34m18,432\u001b[0m │ activation_1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
+ "│ │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m125\u001b[0m, \u001b[38;5;34m125\u001b[0m, │ \u001b[38;5;34m192\u001b[0m │ conv2d_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_2 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m125\u001b[0m, \u001b[38;5;34m125\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ max_pooling2d │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m62\u001b[0m, \u001b[38;5;34m62\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ activation_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
+ "│ (\u001b[38;5;33mMaxPooling2D\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_3 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m62\u001b[0m, \u001b[38;5;34m62\u001b[0m, │ \u001b[38;5;34m5,120\u001b[0m │ max_pooling2d[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m80\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m62\u001b[0m, \u001b[38;5;34m62\u001b[0m, │ \u001b[38;5;34m240\u001b[0m │ conv2d_3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m80\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_3 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m62\u001b[0m, \u001b[38;5;34m62\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m80\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_4 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m60\u001b[0m, \u001b[38;5;34m60\u001b[0m, │ \u001b[38;5;34m138,240\u001b[0m │ activation_3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m60\u001b[0m, \u001b[38;5;34m60\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_4[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_4 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m60\u001b[0m, \u001b[38;5;34m60\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ max_pooling2d_1 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ activation_4[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
+ "│ (\u001b[38;5;33mMaxPooling2D\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_8 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m12,288\u001b[0m │ max_pooling2d_1[\u001b[38;5;34m…\u001b[0m │\n",
+ "│ │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m192\u001b[0m │ conv2d_8[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_8 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_6 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m9,216\u001b[0m │ max_pooling2d_1[\u001b[38;5;34m…\u001b[0m │\n",
+ "│ │ \u001b[38;5;34m48\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_9 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m55,296\u001b[0m │ activation_8[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
+ "│ │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m144\u001b[0m │ conv2d_6[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m48\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m288\u001b[0m │ conv2d_9[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_6 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m48\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_9 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ average_pooling2d │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ max_pooling2d_1[\u001b[38;5;34m…\u001b[0m │\n",
+ "│ (\u001b[38;5;33mAveragePooling2D\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_5 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m12,288\u001b[0m │ max_pooling2d_1[\u001b[38;5;34m…\u001b[0m │\n",
+ "│ │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_7 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m76,800\u001b[0m │ activation_6[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
+ "│ │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_10 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m82,944\u001b[0m │ activation_9[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
+ "│ │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_11 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m6,144\u001b[0m │ average_pooling2… │\n",
+ "│ │ \u001b[38;5;34m32\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m192\u001b[0m │ conv2d_5[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m192\u001b[0m │ conv2d_7[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m288\u001b[0m │ conv2d_10[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m96\u001b[0m │ conv2d_11[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m32\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_5 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_7 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_10 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_11 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m32\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed0 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ activation_5[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
+ "│ (\u001b[38;5;33mConcatenate\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ activation_7[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
+ "│ │ │ │ activation_10[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ │ │ activation_11[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_15 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m16,384\u001b[0m │ mixed0[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m192\u001b[0m │ conv2d_15[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_15 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼──────────────────���┼────────────┼───────────────────┤\n",
+ "│ conv2d_13 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m12,288\u001b[0m │ mixed0[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m48\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_16 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m55,296\u001b[0m │ activation_15[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m144\u001b[0m │ conv2d_13[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m48\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m288\u001b[0m │ conv2d_16[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_13 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m48\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_16 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ average_pooling2d_1 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ mixed0[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mAveragePooling2D\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_12 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m16,384\u001b[0m │ mixed0[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼─────────────��─────┼────────────┼───────────────────┤\n",
+ "│ conv2d_14 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m76,800\u001b[0m │ activation_13[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_17 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m82,944\u001b[0m │ activation_16[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_18 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m16,384\u001b[0m │ average_pooling2… │\n",
+ "│ │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m192\u001b[0m │ conv2d_12[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m192\u001b[0m │ conv2d_14[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m288\u001b[0m │ conv2d_17[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m192\u001b[0m │ conv2d_18[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_12 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼─────────────��─────┼────────────┼───────────────────┤\n",
+ "│ activation_14 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_17 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_18 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed1 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ activation_12[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ (\u001b[38;5;33mConcatenate\u001b[0m) │ \u001b[38;5;34m288\u001b[0m) │ │ activation_14[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ │ │ activation_17[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ │ │ activation_18[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_22 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m18,432\u001b[0m │ mixed1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m192\u001b[0m │ conv2d_22[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_22 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_20 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m13,824\u001b[0m │ mixed1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m48\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_23 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m55,296\u001b[0m │ activation_22[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m144\u001b[0m │ conv2d_20[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m48\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m288\u001b[0m │ conv2d_23[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_20 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m48\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_23 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ average_pooling2d_2 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ mixed1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mAveragePooling2D\u001b[0m) │ \u001b[38;5;34m288\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_19 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m18,432\u001b[0m │ mixed1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_21 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m76,800\u001b[0m │ activation_20[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_24 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m82,944\u001b[0m │ activation_23[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_25 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m18,432\u001b[0m │ average_pooling2… │\n",
+ "│ │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m192\u001b[0m │ conv2d_19[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m192\u001b[0m │ conv2d_21[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m288\u001b[0m │ conv2d_24[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m192\u001b[0m │ conv2d_25[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_19 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_21 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_24 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_25 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed2 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ activation_19[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ (\u001b[38;5;33mConcatenate\u001b[0m) │ \u001b[38;5;34m288\u001b[0m) │ │ activation_21[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ │ │ activation_24[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ │ │ activation_25[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_27 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m18,432\u001b[0m │ mixed2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m192\u001b[0m │ conv2d_27[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_27 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_28 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m55,296\u001b[0m │ activation_27[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m288\u001b[0m │ conv2d_28[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_28 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m29\u001b[0m, \u001b[38;5;34m29\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_26 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m995,328\u001b[0m │ mixed2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m384\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_29 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m82,944\u001b[0m │ activation_28[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m1,152\u001b[0m │ conv2d_26[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m384\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m288\u001b[0m │ conv2d_29[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_26 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m384\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_29 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m96\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ max_pooling2d_2 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ mixed2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mMaxPooling2D\u001b[0m) │ \u001b[38;5;34m288\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed3 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ activation_26[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ (\u001b[38;5;33mConcatenate\u001b[0m) │ \u001b[38;5;34m768\u001b[0m) │ │ activation_29[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ │ │ max_pooling2d_2[\u001b[38;5;34m…\u001b[0m │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_34 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m98,304\u001b[0m │ mixed3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m128\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m384\u001b[0m │ conv2d_34[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m128\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_34 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_35 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m114,688\u001b[0m │ activation_34[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m128\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m384\u001b[0m │ conv2d_35[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m128\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_35 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_31 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m98,304\u001b[0m │ mixed3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m128\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_36 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m114,688\u001b[0m │ activation_35[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m128\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m384\u001b[0m │ conv2d_31[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m128\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m384\u001b[0m │ conv2d_36[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m128\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_31 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_36 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_32 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m114,688\u001b[0m │ activation_31[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m128\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_37 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m114,688\u001b[0m │ activation_36[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m128\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m384\u001b[0m │ conv2d_32[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m128\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m384\u001b[0m │ conv2d_37[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m128\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_32 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_37 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ average_pooling2d_3 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ mixed3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mAveragePooling2D\u001b[0m) │ \u001b[38;5;34m768\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_30 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m147,456\u001b[0m │ mixed3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_33 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m172,032\u001b[0m │ activation_32[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_38 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m172,032\u001b[0m │ activation_37[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_39 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m147,456\u001b[0m │ average_pooling2… │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_30[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_33[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_38[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_39[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_30 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_33 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_38 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_39 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed4 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ activation_30[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ (\u001b[38;5;33mConcatenate\u001b[0m) │ \u001b[38;5;34m768\u001b[0m) │ │ activation_33[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ │ │ activation_38[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ │ │ activation_39[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_44 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m122,880\u001b[0m │ mixed4[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m480\u001b[0m │ conv2d_44[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_44 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_45 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m179,200\u001b[0m │ activation_44[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m480\u001b[0m │ conv2d_45[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_45 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_41 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m122,880\u001b[0m │ mixed4[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_46 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m179,200\u001b[0m │ activation_45[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m480\u001b[0m │ conv2d_41[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m480\u001b[0m │ conv2d_46[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_41 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_46 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_42 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m179,200\u001b[0m │ activation_41[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_47 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m179,200\u001b[0m │ activation_46[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m480\u001b[0m │ conv2d_42[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m480\u001b[0m │ conv2d_47[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_42 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_47 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ average_pooling2d_4 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ mixed4[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mAveragePooling2D\u001b[0m) │ \u001b[38;5;34m768\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_40 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m147,456\u001b[0m │ mixed4[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_43 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m215,040\u001b[0m │ activation_42[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_48 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m215,040\u001b[0m │ activation_47[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────��───┼────────────┼───────────────────┤\n",
+ "│ conv2d_49 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m147,456\u001b[0m │ average_pooling2… │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_40[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_43[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_48[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_49[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_40 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_43 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_48 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼───────────��┼───────────────────┤\n",
+ "│ activation_49 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed5 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ activation_40[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ (\u001b[38;5;33mConcatenate\u001b[0m) │ \u001b[38;5;34m768\u001b[0m) │ │ activation_43[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ │ │ activation_48[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ │ │ activation_49[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_54 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m122,880\u001b[0m │ mixed5[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m480\u001b[0m │ conv2d_54[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_54 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_55 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m179,200\u001b[0m │ activation_54[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m480\u001b[0m │ conv2d_55[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_55 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_51 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m122,880\u001b[0m │ mixed5[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_56 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m179,200\u001b[0m │ activation_55[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m480\u001b[0m │ conv2d_51[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m480\u001b[0m │ conv2d_56[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_51 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_56 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_52 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m179,200\u001b[0m │ activation_51[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_57 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m179,200\u001b[0m │ activation_56[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m160\u001b[0m) │ �� │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m480\u001b[0m │ conv2d_52[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m480\u001b[0m │ conv2d_57[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_52 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_57 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m160\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ average_pooling2d_5 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ mixed5[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mAveragePooling2D\u001b[0m) │ \u001b[38;5;34m768\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_50 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m147,456\u001b[0m │ mixed5[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_53 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m215,040\u001b[0m │ activation_52[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_58 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m215,040\u001b[0m │ activation_57[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_59 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m147,456\u001b[0m │ average_pooling2… │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_50[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_53[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_58[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_59[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_50 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_53 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_58 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_59 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed6 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ activation_50[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ (\u001b[38;5;33mConcatenate\u001b[0m) │ \u001b[38;5;34m768\u001b[0m) │ │ activation_53[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ │ │ activation_58[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ │ │ activation_59[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_64 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m147,456\u001b[0m │ mixed6[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_64[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_64 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_65 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m258,048\u001b[0m │ activation_64[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_65[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_65 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_61 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m147,456\u001b[0m │ mixed6[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_66 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m258,048\u001b[0m │ activation_65[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_61[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_66[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_61 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_66 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_62 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m258,048\u001b[0m │ activation_61[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_67 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m258,048\u001b[0m │ activation_66[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_62[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_67[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_62 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_67 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ average_pooling2d_6 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ mixed6[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mAveragePooling2D\u001b[0m) │ \u001b[38;5;34m768\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_60 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m147,456\u001b[0m │ mixed6[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_63 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m258,048\u001b[0m │ activation_62[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_68 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m258,048\u001b[0m │ activation_67[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_69 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m147,456\u001b[0m │ average_pooling2… │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_60[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_63[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_68[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_69[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_60 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_63 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_68 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_69 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed7 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ activation_60[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ (\u001b[38;5;33mConcatenate\u001b[0m) │ \u001b[38;5;34m768\u001b[0m) │ │ activation_63[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ │ │ activation_68[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ │ │ activation_69[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_72 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m147,456\u001b[0m │ mixed7[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_72[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_72 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_73 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m258,048\u001b[0m │ activation_72[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_73[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_73 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_70 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m147,456\u001b[0m │ mixed7[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_74 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m258,048\u001b[0m │ activation_73[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_70[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m576\u001b[0m │ conv2d_74[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_70 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_74 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m192\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_71 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m320\u001b[0m) │ \u001b[38;5;34m552,960\u001b[0m │ activation_70[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_75 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m192\u001b[0m) │ \u001b[38;5;34m331,776\u001b[0m │ activation_74[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m320\u001b[0m) │ \u001b[38;5;34m960\u001b[0m │ conv2d_71[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m192\u001b[0m) │ \u001b[38;5;34m576\u001b[0m │ conv2d_75[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_71 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m320\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_75 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m192\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ max_pooling2d_3 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m768\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ mixed7[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mMaxPooling2D\u001b[0m) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed8 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ activation_71[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ (\u001b[38;5;33mConcatenate\u001b[0m) │ \u001b[38;5;34m1280\u001b[0m) │ │ activation_75[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ │ │ max_pooling2d_3[\u001b[38;5;34m…\u001b[0m │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_80 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m448\u001b[0m) │ \u001b[38;5;34m573,440\u001b[0m │ mixed8[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m448\u001b[0m) │ \u001b[38;5;34m1,344\u001b[0m │ conv2d_80[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_80 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m448\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_77 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m491,520\u001b[0m │ mixed8[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_81 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m1,548,288\u001b[0m │ activation_80[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m1,152\u001b[0m │ conv2d_77[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m1,152\u001b[0m │ conv2d_81[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_77 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_81 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_78 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m442,368\u001b[0m │ activation_77[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_79 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m442,368\u001b[0m │ activation_77[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_82 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m442,368\u001b[0m │ activation_81[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_83 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m442,368\u001b[0m │ activation_81[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ average_pooling2d_7 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ mixed8[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mAveragePooling2D\u001b[0m) │ \u001b[38;5;34m1280\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_76 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m320\u001b[0m) │ \u001b[38;5;34m409,600\u001b[0m │ mixed8[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m1,152\u001b[0m │ conv2d_78[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m1,152\u001b[0m │ conv2d_79[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m1,152\u001b[0m │ conv2d_82[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m1,152\u001b[0m │ conv2d_83[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_84 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m192\u001b[0m) │ \u001b[38;5;34m245,760\u001b[0m │ average_pooling2… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m320\u001b[0m) │ \u001b[38;5;34m960\u001b[0m │ conv2d_76[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_78 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_79 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_82 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_83 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m192\u001b[0m) │ \u001b[38;5;34m576\u001b[0m │ conv2d_84[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_76 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m320\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed9_0 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m768\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ activation_78[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ (\u001b[38;5;33mConcatenate\u001b[0m) │ │ │ activation_79[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ concatenate │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m768\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ activation_82[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ (\u001b[38;5;33mConcatenate\u001b[0m) │ │ │ activation_83[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_84 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m192\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed9 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ activation_76[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ (\u001b[38;5;33mConcatenate\u001b[0m) │ \u001b[38;5;34m2048\u001b[0m) │ │ mixed9_0[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m], │\n",
+ "│ │ │ │ concatenate[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m… │\n",
+ "│ │ │ │ activation_84[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_89 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m448\u001b[0m) │ \u001b[38;5;34m917,504\u001b[0m │ mixed9[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m448\u001b[0m) │ \u001b[38;5;34m1,344\u001b[0m │ conv2d_89[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_89 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m448\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_86 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m786,432\u001b[0m │ mixed9[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_90 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m1,548,288\u001b[0m │ activation_89[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m1,152\u001b[0m │ conv2d_86[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m1,152\u001b[0m │ conv2d_90[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_86 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_90 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_87 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m442,368\u001b[0m │ activation_86[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_88 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m442,368\u001b[0m │ activation_86[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_91 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m442,368\u001b[0m │ activation_90[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_92 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m442,368\u001b[0m │ activation_90[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ average_pooling2d_8 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ mixed9[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mAveragePooling2D\u001b[0m) │ \u001b[38;5;34m2048\u001b[0m) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_85 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m320\u001b[0m) │ \u001b[38;5;34m655,360\u001b[0m │ mixed9[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m1,152\u001b[0m │ conv2d_87[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m1,152\u001b[0m │ conv2d_88[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m1,152\u001b[0m │ conv2d_91[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m1,152\u001b[0m │ conv2d_92[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_93 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m192\u001b[0m) │ \u001b[38;5;34m393,216\u001b[0m │ average_pooling2… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m320\u001b[0m) │ \u001b[38;5;34m960\u001b[0m │ conv2d_85[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n",
+ "├───────────────────���─┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_87 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_88 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_91 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_92 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m384\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m192\u001b[0m) │ \u001b[38;5;34m576\u001b[0m │ conv2d_93[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_85 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m320\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed9_1 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m768\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ activation_87[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ (\u001b[38;5;33mConcatenate\u001b[0m) │ │ │ activation_88[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ concatenate_1 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m768\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ activation_91[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ (\u001b[38;5;33mConcatenate\u001b[0m) │ │ │ activation_92[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_93 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m192\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
+ "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed10 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ activation_85[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ (\u001b[38;5;33mConcatenate\u001b[0m) │ \u001b[38;5;34m2048\u001b[0m) │ │ mixed9_1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m], │\n",
+ "│ │ │ │ concatenate_1[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "│ │ │ │ activation_93[\u001b[38;5;34m0\u001b[0m]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ global_average_poo… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2048\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ mixed10[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "│ (\u001b[38;5;33mGlobalAveragePool…\u001b[0m │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ dense (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m1,049,088\u001b[0m │ global_average_p… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ dropout (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ dense[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ dense_1 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m3\u001b[0m) │ \u001b[38;5;34m1,539\u001b[0m │ dropout[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
+ "└─────────────────────┴───────────────────┴────────────┴───────────────────┘\n"
+ ],
+ "text/html": [
+ "┏━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┓\n",
+ "┃ Layer (type) ┃ Output Shape ┃ Param # ┃ Connected to ┃\n",
+ "┡━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━┩\n",
+ "│ input_layer │ (None, 256, 256, │ 0 │ - │\n",
+ "│ (InputLayer) │ 3) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d (Conv2D) │ (None, 127, 127, │ 864 │ input_layer[0][0] │\n",
+ "│ │ 32) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalization │ (None, 127, 127, │ 96 │ conv2d[0][0] │\n",
+ "│ (BatchNormalizatio… │ 32) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation │ (None, 127, 127, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 32) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_1 (Conv2D) │ (None, 125, 125, │ 9,216 │ activation[0][0] │\n",
+ "│ │ 32) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 125, 125, │ 96 │ conv2d_1[0][0] │\n",
+ "│ (BatchNormalizatio… │ 32) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_1 │ (None, 125, 125, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 32) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_2 (Conv2D) │ (None, 125, 125, │ 18,432 │ activation_1[0][… │\n",
+ "│ │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 125, 125, │ 192 │ conv2d_2[0][0] │\n",
+ "│ (BatchNormalizatio… │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_2 │ (None, 125, 125, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ max_pooling2d │ (None, 62, 62, │ 0 │ activation_2[0][… │\n",
+ "│ (MaxPooling2D) │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_3 (Conv2D) │ (None, 62, 62, │ 5,120 │ max_pooling2d[0]… │\n",
+ "│ │ 80) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 62, 62, │ 240 │ conv2d_3[0][0] │\n",
+ "│ (BatchNormalizatio… │ 80) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_3 │ (None, 62, 62, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 80) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_4 (Conv2D) │ (None, 60, 60, │ 138,240 │ activation_3[0][… │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 60, 60, │ 576 │ conv2d_4[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_4 │ (None, 60, 60, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ max_pooling2d_1 │ (None, 29, 29, │ 0 │ activation_4[0][… │\n",
+ "│ (MaxPooling2D) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_8 (Conv2D) │ (None, 29, 29, │ 12,288 │ max_pooling2d_1[… │\n",
+ "│ │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 192 │ conv2d_8[0][0] │\n",
+ "│ (BatchNormalizatio… │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_8 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_6 (Conv2D) │ (None, 29, 29, │ 9,216 │ max_pooling2d_1[… │\n",
+ "│ │ 48) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_9 (Conv2D) │ (None, 29, 29, │ 55,296 │ activation_8[0][… │\n",
+ "│ │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 144 │ conv2d_6[0][0] │\n",
+ "│ (BatchNormalizatio… │ 48) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 288 │ conv2d_9[0][0] │\n",
+ "│ (BatchNormalizatio… │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_6 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 48) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_9 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 96) │ │ │\n",
+ "├─────────────────────┼─────────���─────────┼────────────┼───────────────────┤\n",
+ "│ average_pooling2d │ (None, 29, 29, │ 0 │ max_pooling2d_1[… │\n",
+ "│ (AveragePooling2D) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_5 (Conv2D) │ (None, 29, 29, │ 12,288 │ max_pooling2d_1[… │\n",
+ "│ │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_7 (Conv2D) │ (None, 29, 29, │ 76,800 │ activation_6[0][… │\n",
+ "│ │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_10 (Conv2D) │ (None, 29, 29, │ 82,944 │ activation_9[0][… │\n",
+ "│ │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_11 (Conv2D) │ (None, 29, 29, │ 6,144 │ average_pooling2… │\n",
+ "│ │ 32) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 192 │ conv2d_5[0][0] │\n",
+ "│ (BatchNormalizatio… │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 192 │ conv2d_7[0][0] │\n",
+ "│ (BatchNormalizatio… │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 288 │ conv2d_10[0][0] │\n",
+ "│ (BatchNormalizatio… │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 96 │ conv2d_11[0][0] │\n",
+ "│ (BatchNormalizatio… │ 32) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_5 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_7 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_10 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_11 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 32) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed0 │ (None, 29, 29, │ 0 │ activation_5[0][… │\n",
+ "│ (Concatenate) │ 256) │ │ activation_7[0][… │\n",
+ "│ │ │ │ activation_10[0]… │\n",
+ "│ │ │ │ activation_11[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_15 (Conv2D) │ (None, 29, 29, │ 16,384 │ mixed0[0][0] │\n",
+ "│ │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 192 │ conv2d_15[0][0] │\n",
+ "│ (BatchNormalizatio… │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_15 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_13 (Conv2D) │ (None, 29, 29, │ 12,288 │ mixed0[0][0] │\n",
+ "│ │ 48) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_16 (Conv2D) │ (None, 29, 29, │ 55,296 │ activation_15[0]… │\n",
+ "│ │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 144 │ conv2d_13[0][0] │\n",
+ "│ (BatchNormalizatio… │ 48) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 288 │ conv2d_16[0][0] │\n",
+ "│ (BatchNormalizatio… │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_13 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 48) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_16 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ average_pooling2d_1 │ (None, 29, 29, │ 0 │ mixed0[0][0] │\n",
+ "│ (AveragePooling2D) │ 256) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_12 (Conv2D) │ (None, 29, 29, │ 16,384 │ mixed0[0][0] │\n",
+ "│ │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_14 (Conv2D) │ (None, 29, 29, │ 76,800 │ activation_13[0]… │\n",
+ "│ │ 64) �� │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_17 (Conv2D) │ (None, 29, 29, │ 82,944 │ activation_16[0]… │\n",
+ "│ │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_18 (Conv2D) │ (None, 29, 29, │ 16,384 │ average_pooling2… │\n",
+ "│ │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 192 │ conv2d_12[0][0] │\n",
+ "│ (BatchNormalizatio… │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 192 │ conv2d_14[0][0] │\n",
+ "│ (BatchNormalizatio… │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 288 │ conv2d_17[0][0] │\n",
+ "│ (BatchNormalizatio… │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼───��────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 192 │ conv2d_18[0][0] │\n",
+ "│ (BatchNormalizatio… │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_12 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_14 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_17 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_18 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed1 │ (None, 29, 29, │ 0 │ activation_12[0]… │\n",
+ "│ (Concatenate) │ 288) │ │ activation_14[0]… │\n",
+ "│ │ │ │ activation_17[0]… │\n",
+ "│ │ │ │ activation_18[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_22 (Conv2D) │ (None, 29, 29, │ 18,432 │ mixed1[0][0] │\n",
+ "│ │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 192 │ conv2d_22[0][0] │\n",
+ "│ (BatchNormalizatio… │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_22 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_20 (Conv2D) │ (None, 29, 29, │ 13,824 │ mixed1[0][0] │\n",
+ "│ │ 48) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_23 (Conv2D) │ (None, 29, 29, │ 55,296 │ activation_22[0]… │\n",
+ "│ │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 144 │ conv2d_20[0][0] │\n",
+ "│ (BatchNormalizatio… │ 48) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 288 │ conv2d_23[0][0] │\n",
+ "│ (BatchNormalizatio… │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_20 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 48) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_23 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ average_pooling2d_2 │ (None, 29, 29, │ 0 │ mixed1[0][0] │\n",
+ "│ (AveragePooling2D) │ 288) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_19 (Conv2D) │ (None, 29, 29, │ 18,432 │ mixed1[0][0] │\n",
+ "│ │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_21 (Conv2D) │ (None, 29, 29, │ 76,800 │ activation_20[0]… │\n",
+ "│ │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_24 (Conv2D) │ (None, 29, 29, │ 82,944 │ activation_23[0]… │\n",
+ "│ │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_25 (Conv2D) │ (None, 29, 29, │ 18,432 │ average_pooling2… │\n",
+ "│ │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 192 │ conv2d_19[0][0] │\n",
+ "│ (BatchNormalizatio… │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 192 │ conv2d_21[0][0] │\n",
+ "│ (BatchNormalizatio… │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 288 │ conv2d_24[0][0] │\n",
+ "│ (BatchNormalizatio… │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 192 │ conv2d_25[0][0] │\n",
+ "│ (BatchNormalizatio… │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_19 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_21 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_24 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_25 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed2 │ (None, 29, 29, │ 0 │ activation_19[0]… │\n",
+ "│ (Concatenate) │ 288) │ │ activation_21[0]… │\n",
+ "│ │ │ │ activation_24[0]… │\n",
+ "│ │ │ │ activation_25[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_27 (Conv2D) │ (None, 29, 29, │ 18,432 │ mixed2[0][0] │\n",
+ "│ │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 192 │ conv2d_27[0][0] │\n",
+ "│ (BatchNormalizatio… │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_27 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 64) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_28 (Conv2D) │ (None, 29, 29, │ 55,296 │ activation_27[0]… │\n",
+ "│ │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 29, 29, │ 288 │ conv2d_28[0][0] │\n",
+ "│ (BatchNormalizatio… │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_28 │ (None, 29, 29, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_26 (Conv2D) │ (None, 14, 14, │ 995,328 │ mixed2[0][0] │\n",
+ "│ │ 384) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_29 (Conv2D) │ (None, 14, 14, │ 82,944 │ activation_28[0]… │\n",
+ "│ │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 1,152 │ conv2d_26[0][0] │\n",
+ "│ (BatchNormalizatio… │ 384) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 288 │ conv2d_29[0][0] │\n",
+ "│ (BatchNormalizatio… │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_26 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 384) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_29 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 96) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ max_pooling2d_2 │ (None, 14, 14, │ 0 │ mixed2[0][0] │\n",
+ "│ (MaxPooling2D) │ 288) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed3 │ (None, 14, 14, │ 0 │ activation_26[0]… │\n",
+ "│ (Concatenate) │ 768) │ │ activation_29[0]… │\n",
+ "│ │ │ │ max_pooling2d_2[… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_34 (Conv2D) │ (None, 14, 14, │ 98,304 │ mixed3[0][0] │\n",
+ "│ │ 128) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 384 │ conv2d_34[0][0] │\n",
+ "│ (BatchNormalizatio… │ 128) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_34 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 128) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_35 (Conv2D) │ (None, 14, 14, │ 114,688 │ activation_34[0]… │\n",
+ "│ │ 128) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 384 │ conv2d_35[0][0] │\n",
+ "│ (BatchNormalizatio… │ 128) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_35 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 128) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_31 (Conv2D) │ (None, 14, 14, │ 98,304 │ mixed3[0][0] │\n",
+ "│ │ 128) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_36 (Conv2D) │ (None, 14, 14, │ 114,688 │ activation_35[0]… │\n",
+ "│ │ 128) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 384 │ conv2d_31[0][0] │\n",
+ "│ (BatchNormalizatio… │ 128) │ │ │\n",
+ "├─────────────────────┼───────────────────┼──��─────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 384 │ conv2d_36[0][0] │\n",
+ "│ (BatchNormalizatio… │ 128) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_31 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 128) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_36 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 128) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_32 (Conv2D) │ (None, 14, 14, │ 114,688 │ activation_31[0]… │\n",
+ "│ │ 128) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_37 (Conv2D) │ (None, 14, 14, │ 114,688 │ activation_36[0]… │\n",
+ "│ │ 128) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 384 │ conv2d_32[0][0] │\n",
+ "│ (BatchNormalizatio… │ 128) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 384 │ conv2d_37[0][0] │\n",
+ "│ (BatchNormalizatio… │ 128) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_32 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 128) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_37 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 128) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ average_pooling2d_3 │ (None, 14, 14, │ 0 │ mixed3[0][0] │\n",
+ "│ (AveragePooling2D) │ 768) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_30 (Conv2D) │ (None, 14, 14, │ 147,456 │ mixed3[0][0] ���\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_33 (Conv2D) │ (None, 14, 14, │ 172,032 │ activation_32[0]… │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_38 (Conv2D) │ (None, 14, 14, │ 172,032 │ activation_37[0]… │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_39 (Conv2D) │ (None, 14, 14, │ 147,456 │ average_pooling2… │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_30[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_33[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼��──────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_38[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_39[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_30 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_33 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_38 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_39 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed4 │ (None, 14, 14, │ 0 │ activation_30[0]… │\n",
+ "│ (Concatenate) │ 768) │ │ activation_33[0]… │\n",
+ "│ │ │ │ activation_38[0]… │\n",
+ "│ │ │ │ activation_39[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_44 (Conv2D) │ (None, 14, 14, │ 122,880 │ mixed4[0][0] │\n",
+ "│ │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 480 │ conv2d_44[0][0] │\n",
+ "│ (BatchNormalizatio… │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_44 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_45 (Conv2D) │ (None, 14, 14, │ 179,200 │ activation_44[0]… │\n",
+ "│ │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 480 │ conv2d_45[0][0] │\n",
+ "│ (BatchNormalizatio… │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_45 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_41 (Conv2D) │ (None, 14, 14, │ 122,880 │ mixed4[0][0] │\n",
+ "│ │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_46 (Conv2D) │ (None, 14, 14, │ 179,200 │ activation_45[0]… │\n",
+ "│ │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 480 │ conv2d_41[0][0] │\n",
+ "│ (BatchNormalizatio… │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 480 │ conv2d_46[0][0] │\n",
+ "│ (BatchNormalizatio… │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_41 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_46 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_42 (Conv2D) │ (None, 14, 14, │ 179,200 │ activation_41[0]… │\n",
+ "│ │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_47 (Conv2D) │ (None, 14, 14, │ 179,200 │ activation_46[0]… │\n",
+ "│ │ 160) │ │ │\n",
+ "├──────────────────���──┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 480 │ conv2d_42[0][0] │\n",
+ "│ (BatchNormalizatio… │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 480 │ conv2d_47[0][0] │\n",
+ "│ (BatchNormalizatio… │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_42 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_47 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ average_pooling2d_4 │ (None, 14, 14, │ 0 │ mixed4[0][0] │\n",
+ "│ (AveragePooling2D) │ 768) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_40 (Conv2D) │ (None, 14, 14, │ 147,456 │ mixed4[0][0] │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_43 (Conv2D) │ (None, 14, 14, │ 215,040 │ activation_42[0]… │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_48 (Conv2D) │ (None, 14, 14, │ 215,040 │ activation_47[0]… │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_49 (Conv2D) │ (None, 14, 14, │ 147,456 │ average_pooling2… │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_40[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_43[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_48[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_49[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_40 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_43 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_48 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_49 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed5 │ (None, 14, 14, │ 0 │ activation_40[0]… │\n",
+ "│ (Concatenate) │ 768) │ │ activation_43[0]… │\n",
+ "│ │ │ │ activation_48[0]… │\n",
+ "│ │ │ │ activation_49[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_54 (Conv2D) │ (None, 14, 14, │ 122,880 │ mixed5[0][0] │\n",
+ "│ │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 480 │ conv2d_54[0][0] │\n",
+ "│ (BatchNormalizatio… │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_54 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_55 (Conv2D) │ (None, 14, 14, │ 179,200 │ activation_54[0]… │\n",
+ "│ │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 480 │ conv2d_55[0][0] │\n",
+ "│ (BatchNormalizatio… │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_55 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_51 (Conv2D) │ (None, 14, 14, │ 122,880 │ mixed5[0][0] │\n",
+ "│ │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_56 (Conv2D) │ (None, 14, 14, │ 179,200 │ activation_55[0]… │\n",
+ "│ │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 480 │ conv2d_51[0][0] │\n",
+ "│ (BatchNormalizatio… │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 480 │ conv2d_56[0][0] │\n",
+ "│ (BatchNormalizatio… │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_51 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_56 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_52 (Conv2D) │ (None, 14, 14, │ 179,200 │ activation_51[0]… │\n",
+ "│ │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_57 (Conv2D) │ (None, 14, 14, │ 179,200 │ activation_56[0]… │\n",
+ "│ │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 480 │ conv2d_52[0][0] │\n",
+ "│ (BatchNormalizatio… │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 480 │ conv2d_57[0][0] │\n",
+ "│ (BatchNormalizatio… │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_52 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_57 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 160) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ average_pooling2d_5 │ (None, 14, 14, │ 0 │ mixed5[0][0] │\n",
+ "│ (AveragePooling2D) │ 768) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_50 (Conv2D) │ (None, 14, 14, │ 147,456 │ mixed5[0][0] │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_53 (Conv2D) │ (None, 14, 14, │ 215,040 │ activation_52[0]… │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_58 (Conv2D) │ (None, 14, 14, │ 215,040 │ activation_57[0]… │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_59 (Conv2D) │ (None, 14, 14, │ 147,456 │ average_pooling2… │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_50[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_53[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_58[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_59[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_50 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_53 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_58 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_59 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed6 │ (None, 14, 14, │ 0 │ activation_50[0]… │\n",
+ "│ (Concatenate) │ 768) │ │ activation_53[0]… │\n",
+ "│ │ │ │ activation_58[0]… │\n",
+ "│ │ │ │ activation_59[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_64 (Conv2D) │ (None, 14, 14, │ 147,456 │ mixed6[0][0] │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_64[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_64 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_65 (Conv2D) │ (None, 14, 14, │ 258,048 │ activation_64[0]… │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_65[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_65 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_61 (Conv2D) │ (None, 14, 14, │ 147,456 │ mixed6[0][0] │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_66 (Conv2D) │ (None, 14, 14, │ 258,048 │ activation_65[0]… │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_61[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_66[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_61 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_66 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_62 (Conv2D) │ (None, 14, 14, │ 258,048 │ activation_61[0]… │\n",
+ "│ │ 192) │ │ │\n",
+ "├────────��────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_67 (Conv2D) │ (None, 14, 14, │ 258,048 │ activation_66[0]… │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_62[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_67[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_62 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_67 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ average_pooling2d_6 │ (None, 14, 14, │ 0 │ mixed6[0][0] │\n",
+ "│ (AveragePooling2D) │ 768) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_60 (Conv2D) │ (None, 14, 14, │ 147,456 │ mixed6[0][0] │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_63 (Conv2D) │ (None, 14, 14, │ 258,048 │ activation_62[0]… │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_68 (Conv2D) │ (None, 14, 14, │ 258,048 │ activation_67[0]… │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_69 (Conv2D) │ (None, 14, 14, │ 147,456 │ average_pooling2… │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_60[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_63[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_68[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_69[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_60 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_63 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_68 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_69 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed7 │ (None, 14, 14, │ 0 │ activation_60[0]… │\n",
+ "│ (Concatenate) │ 768) │ │ activation_63[0]… │\n",
+ "│ │ │ │ activation_68[0]… │\n",
+ "│ │ │ │ activation_69[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_72 (Conv2D) │ (None, 14, 14, │ 147,456 │ mixed7[0][0] │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_72[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_72 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_73 (Conv2D) │ (None, 14, 14, │ 258,048 │ activation_72[0]… │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_73[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_73 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_70 (Conv2D) │ (None, 14, 14, │ 147,456 │ mixed7[0][0] │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼───────���────┼───────────────────┤\n",
+ "│ conv2d_74 (Conv2D) │ (None, 14, 14, │ 258,048 │ activation_73[0]… │\n",
+ "│ │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_70[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 14, 14, │ 576 │ conv2d_74[0][0] │\n",
+ "│ (BatchNormalizatio… │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_70 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_74 │ (None, 14, 14, │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ 192) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_71 (Conv2D) │ (None, 6, 6, 320) │ 552,960 │ activation_70[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_75 (Conv2D) │ (None, 6, 6, 192) │ 331,776 │ activation_74[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 6, 6, 320) │ 960 │ conv2d_71[0][0] │\n",
+ "│ (BatchNormalizatio… │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 6, 6, 192) │ 576 │ conv2d_75[0][0] │\n",
+ "│ (BatchNormalizatio… │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_71 │ (None, 6, 6, 320) │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_75 │ (None, 6, 6, 192) │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ │ │ │\n",
+ "├─────────────────────┼───────��───────────┼────────────┼───────────────────┤\n",
+ "│ max_pooling2d_3 │ (None, 6, 6, 768) │ 0 │ mixed7[0][0] │\n",
+ "│ (MaxPooling2D) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed8 │ (None, 6, 6, │ 0 │ activation_71[0]… │\n",
+ "│ (Concatenate) │ 1280) │ │ activation_75[0]… │\n",
+ "│ │ │ │ max_pooling2d_3[… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_80 (Conv2D) │ (None, 6, 6, 448) │ 573,440 │ mixed8[0][0] │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 6, 6, 448) │ 1,344 │ conv2d_80[0][0] │\n",
+ "│ (BatchNormalizatio… │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_80 │ (None, 6, 6, 448) │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────��───┼───────────────────┤\n",
+ "│ conv2d_77 (Conv2D) │ (None, 6, 6, 384) │ 491,520 │ mixed8[0][0] │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_81 (Conv2D) │ (None, 6, 6, 384) │ 1,548,288 │ activation_80[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 6, 6, 384) │ 1,152 │ conv2d_77[0][0] │\n",
+ "│ (BatchNormalizatio… │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 6, 6, 384) │ 1,152 │ conv2d_81[0][0] │\n",
+ "│ (BatchNormalizatio… │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_77 │ (None, 6, 6, 384) │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_81 │ (None, 6, 6, 384) │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_78 (Conv2D) │ (None, 6, 6, 384) │ 442,368 │ activation_77[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_79 (Conv2D) │ (None, 6, 6, 384) │ 442,368 │ activation_77[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_82 (Conv2D) │ (None, 6, 6, 384) │ 442,368 │ activation_81[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_83 (Conv2D) │ (None, 6, 6, 384) │ 442,368 │ activation_81[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ average_pooling2d_7 │ (None, 6, 6, │ 0 │ mixed8[0][0] │\n",
+ "│ (AveragePooling2D) │ 1280) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_76 (Conv2D) │ (None, 6, 6, 320) │ 409,600 │ mixed8[0][0] │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 6, 6, 384) │ 1,152 │ conv2d_78[0][0] │\n",
+ "│ (BatchNormalizatio… │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 6, 6, 384) │ 1,152 │ conv2d_79[0][0] │\n",
+ "│ (BatchNormalizatio… │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 6, 6, 384) │ 1,152 │ conv2d_82[0][0] │\n",
+ "│ (BatchNormalizatio… │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 6, 6, 384) │ 1,152 │ conv2d_83[0][0] │\n",
+ "│ (BatchNormalizatio… │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_84 (Conv2D) │ (None, 6, 6, 192) │ 245,760 │ average_pooling2… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 6, 6, 320) │ 960 │ conv2d_76[0][0] │\n",
+ "│ (BatchNormalizatio… │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_78 │ (None, 6, 6, 384) │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_79 │ (None, 6, 6, 384) │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_82 │ (None, 6, 6, 384) │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_83 │ (None, 6, 6, 384) │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 6, 6, 192) │ 576 │ conv2d_84[0][0] │\n",
+ "│ (BatchNormalizatio… │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_76 │ (None, 6, 6, 320) │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed9_0 │ (None, 6, 6, 768) │ 0 │ activation_78[0]… │\n",
+ "│ (Concatenate) │ │ │ activation_79[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ concatenate │ (None, 6, 6, 768) │ 0 │ activation_82[0]… │\n",
+ "│ (Concatenate) │ │ │ activation_83[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_84 │ (None, 6, 6, 192) │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed9 │ (None, 6, 6, │ 0 │ activation_76[0]… │\n",
+ "│ (Concatenate) │ 2048) │ │ mixed9_0[0][0], │\n",
+ "│ │ │ │ concatenate[0][0… │\n",
+ "│ │ │ │ activation_84[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_89 (Conv2D) │ (None, 6, 6, 448) │ 917,504 │ mixed9[0][0] │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 6, 6, 448) │ 1,344 │ conv2d_89[0][0] │\n",
+ "│ (BatchNormalizatio… │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_89 │ (None, 6, 6, 448) │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_86 (Conv2D) │ (None, 6, 6, 384) │ 786,432 │ mixed9[0][0] │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_90 (Conv2D) │ (None, 6, 6, 384) │ 1,548,288 │ activation_89[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 6, 6, 384) │ 1,152 │ conv2d_86[0][0] │\n",
+ "│ (BatchNormalizatio… │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 6, 6, 384) │ 1,152 │ conv2d_90[0][0] │\n",
+ "│ (BatchNormalizatio… │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_86 │ (None, 6, 6, 384) │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_90 │ (None, 6, 6, 384) │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_87 (Conv2D) │ (None, 6, 6, 384) │ 442,368 │ activation_86[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_88 (Conv2D) │ (None, 6, 6, 384) │ 442,368 │ activation_86[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_91 (Conv2D) │ (None, 6, 6, 384) │ 442,368 │ activation_90[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_92 (Conv2D) │ (None, 6, 6, 384) │ 442,368 │ activation_90[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ average_pooling2d_8 │ (None, 6, 6, │ 0 │ mixed9[0][0] │\n",
+ "│ (AveragePooling2D) │ 2048) │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_85 (Conv2D) │ (None, 6, 6, 320) │ 655,360 │ mixed9[0][0] │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 6, 6, 384) │ 1,152 │ conv2d_87[0][0] │\n",
+ "│ (BatchNormalizatio… │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 6, 6, 384) │ 1,152 │ conv2d_88[0][0] │\n",
+ "│ (BatchNormalizatio… │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 6, 6, 384) │ 1,152 │ conv2d_91[0][0] │\n",
+ "│ (BatchNormalizatio… │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 6, 6, 384) │ 1,152 │ conv2d_92[0][0] │\n",
+ "│ (BatchNormalizatio… │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ conv2d_93 (Conv2D) │ (None, 6, 6, 192) │ 393,216 │ average_pooling2… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 6, 6, 320) │ 960 │ conv2d_85[0][0] │\n",
+ "│ (BatchNormalizatio… │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_87 │ (None, 6, 6, 384) │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_88 │ (None, 6, 6, 384) │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_91 │ (None, 6, 6, 384) │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_92 │ (None, 6, 6, 384) │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ batch_normalizatio… │ (None, 6, 6, 192) │ 576 │ conv2d_93[0][0] │\n",
+ "│ (BatchNormalizatio… │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_85 │ (None, 6, 6, 320) │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed9_1 │ (None, 6, 6, 768) │ 0 │ activation_87[0]… │\n",
+ "│ (Concatenate) │ │ │ activation_88[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ concatenate_1 │ (None, 6, 6, 768) │ 0 │ activation_91[0]… │\n",
+ "│ (Concatenate) │ │ │ activation_92[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ activation_93 │ (None, 6, 6, 192) │ 0 │ batch_normalizat… │\n",
+ "│ (Activation) │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ mixed10 │ (None, 6, 6, │ 0 │ activation_85[0]… │\n",
+ "│ (Concatenate) │ 2048) │ │ mixed9_1[0][0], │\n",
+ "│ │ │ │ concatenate_1[0]… │\n",
+ "│ │ │ │ activation_93[0]… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ global_average_poo… │ (None, 2048) │ 0 │ mixed10[0][0] │\n",
+ "│ (GlobalAveragePool… │ │ │ │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ dense (Dense) │ (None, 512) │ 1,049,088 │ global_average_p… │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ dropout (Dropout) │ (None, 512) │ 0 │ dense[0][0] │\n",
+ "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+ "│ dense_1 (Dense) │ (None, 3) │ 1,539 │ dropout[0][0] │\n",
+ "└─────────────────────┴───────────────────┴────────────┴───────────────────┘\n",
+ "
\n"
+ ]
+ },
+ "metadata": {}
+ },
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m22,853,411\u001b[0m (87.18 MB)\n"
+ ],
+ "text/html": [
+ " Total params: 22,853,411 (87.18 MB)\n",
+ "
\n"
+ ]
+ },
+ "metadata": {}
+ },
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m1,050,627\u001b[0m (4.01 MB)\n"
+ ],
+ "text/html": [
+ " Trainable params: 1,050,627 (4.01 MB)\n",
+ "
\n"
+ ]
+ },
+ "metadata": {}
+ },
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m21,802,784\u001b[0m (83.17 MB)\n"
+ ],
+ "text/html": [
+ " Non-trainable params: 21,802,784 (83.17 MB)\n",
+ "
\n"
+ ]
+ },
+ "metadata": {}
+ }
+ ]
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "# tell the model what cost and optimization method to use\n",
+ "model.compile(\n",
+ " loss='sparse_categorical_crossentropy',\n",
+ " optimizer='adam',\n",
+ " metrics=['accuracy']\n",
+ ")"
+ ],
+ "metadata": {
+ "id": "gfNkhjcSjwnS"
+ },
+ "execution_count": 19,
+ "outputs": []
+ },
+ {
+ "cell_type": "markdown",
+ "source": [
+ "# **Model Feature Extraction**"
+ ],
+ "metadata": {
+ "id": "3BpOSE3uwVrh"
+ }
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "callbacks = [\n",
+ " EarlyStopping(monitor='val_loss', patience=3, restore_best_weights=True, verbose = 1),\n",
+ " ModelCheckpoint(\"best_model.h5\", save_best_only=True, monitor='val_loss', verbose = 1),\n",
+ " ReduceLROnPlateau(monitor='val_loss', factor=0.5, patience=2, min_lr=1e-5, verbose=1)\n",
+ "]\n",
+ "\n",
+ "history = model.fit(train_ds, validation_data=val_ds, epochs=5, callbacks=callbacks, verbose = 1)"
+ ],
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "id": "RQdBN1FUm4cd",
+ "outputId": "d8a99641-0bee-4335-bf14-cc9af3262b6b"
+ },
+ "execution_count": 20,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "Epoch 1/5\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 211ms/step - accuracy: 0.8509 - loss: 0.4330\n",
+ "Epoch 1: val_loss improved from inf to 0.06660, saving model to best_model.h5\n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stderr",
+ "text": [
+ "WARNING:absl:You are saving your model as an HDF5 file via `model.save()` or `keras.saving.save_model(model)`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')` or `keras.saving.save_model(model, 'my_model.keras')`. \n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m49s\u001b[0m 410ms/step - accuracy: 0.8521 - loss: 0.4298 - val_accuracy: 0.9741 - val_loss: 0.0666 - learning_rate: 0.0010\n",
+ "Epoch 2/5\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 87ms/step - accuracy: 0.9722 - loss: 0.0880\n",
+ "Epoch 2: val_loss improved from 0.06660 to 0.05993, saving model to best_model.h5\n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stderr",
+ "text": [
+ "WARNING:absl:You are saving your model as an HDF5 file via `model.save()` or `keras.saving.save_model(model)`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')` or `keras.saving.save_model(model, 'my_model.keras')`. \n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 107ms/step - accuracy: 0.9722 - loss: 0.0881 - val_accuracy: 0.9741 - val_loss: 0.0599 - learning_rate: 0.0010\n",
+ "Epoch 3/5\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 85ms/step - accuracy: 0.9845 - loss: 0.0572\n",
+ "Epoch 3: val_loss did not improve from 0.05993\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 95ms/step - accuracy: 0.9845 - loss: 0.0572 - val_accuracy: 0.9741 - val_loss: 0.0714 - learning_rate: 0.0010\n",
+ "Epoch 4/5\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 84ms/step - accuracy: 0.9905 - loss: 0.0337\n",
+ "Epoch 4: val_loss did not improve from 0.05993\n",
+ "\n",
+ "Epoch 4: ReduceLROnPlateau reducing learning rate to 0.0005000000237487257.\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 94ms/step - accuracy: 0.9905 - loss: 0.0340 - val_accuracy: 0.9667 - val_loss: 0.0852 - learning_rate: 0.0010\n",
+ "Epoch 5/5\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 88ms/step - accuracy: 0.9866 - loss: 0.0362\n",
+ "Epoch 5: val_loss improved from 0.05993 to 0.02748, saving model to best_model.h5\n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stderr",
+ "text": [
+ "WARNING:absl:You are saving your model as an HDF5 file via `model.save()` or `keras.saving.save_model(model)`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')` or `keras.saving.save_model(model, 'my_model.keras')`. \n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 108ms/step - accuracy: 0.9867 - loss: 0.0362 - val_accuracy: 0.9889 - val_loss: 0.0275 - learning_rate: 5.0000e-04\n",
+ "Restoring model weights from the end of the best epoch: 5.\n"
+ ]
+ }
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "source": [
+ "# **Model Fine-Tuning**"
+ ],
+ "metadata": {
+ "id": "4vdljsyQcKR5"
+ }
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "#Fine Tuning\n",
+ "for layer in inception.layers[-30:]: # Unfreeze last 30 layers (tune as needed)\n",
+ " layer.trainable = True\n",
+ "\n",
+ "# tell the model what cost and optimization method to use\n",
+ "model.compile(\n",
+ " loss='sparse_categorical_crossentropy',\n",
+ " optimizer='adam',\n",
+ " metrics=['accuracy']\n",
+ ")\n",
+ "\n",
+ "callbacks = [\n",
+ " EarlyStopping(monitor='val_loss', patience=5, restore_best_weights=True, verbose = 1),\n",
+ " ModelCheckpoint(\"best_model.h5\", save_best_only=True, monitor='val_loss', verbose = 1),\n",
+ " ReduceLROnPlateau(monitor='val_loss', factor=0.5, patience=7, min_lr=1e-5, verbose=1)\n",
+ "]\n",
+ "\n",
+ "history = model.fit(train_ds, validation_data=val_ds, epochs=10, callbacks=callbacks, verbose = 1)"
+ ],
+ "metadata": {
+ "id": "tJJmGsVwcJ-y",
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "outputId": "fac049b9-3ff5-41a0-ed75-47c561b0bfaf"
+ },
+ "execution_count": 22,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "Epoch 1/10\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 203ms/step - accuracy: 0.9950 - loss: 0.0395\n",
+ "Epoch 1: val_loss improved from inf to 0.15426, saving model to best_model.h5\n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stderr",
+ "text": [
+ "WARNING:absl:You are saving your model as an HDF5 file via `model.save()` or `keras.saving.save_model(model)`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')` or `keras.saving.save_model(model, 'my_model.keras')`. \n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m56s\u001b[0m 374ms/step - accuracy: 0.9950 - loss: 0.0395 - val_accuracy: 0.9852 - val_loss: 0.1543 - learning_rate: 0.0010\n",
+ "Epoch 2/10\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 107ms/step - accuracy: 0.9968 - loss: 0.0215\n",
+ "Epoch 2: val_loss improved from 0.15426 to 0.04965, saving model to best_model.h5\n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stderr",
+ "text": [
+ "WARNING:absl:You are saving your model as an HDF5 file via `model.save()` or `keras.saving.save_model(model)`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')` or `keras.saving.save_model(model, 'my_model.keras')`. \n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 129ms/step - accuracy: 0.9968 - loss: 0.0216 - val_accuracy: 0.9926 - val_loss: 0.0497 - learning_rate: 0.0010\n",
+ "Epoch 3/10\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 100ms/step - accuracy: 0.9940 - loss: 0.0194\n",
+ "Epoch 3: val_loss did not improve from 0.04965\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 111ms/step - accuracy: 0.9940 - loss: 0.0194 - val_accuracy: 0.9926 - val_loss: 0.0808 - learning_rate: 0.0010\n",
+ "Epoch 4/10\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 99ms/step - accuracy: 0.9963 - loss: 0.0105\n",
+ "Epoch 4: val_loss did not improve from 0.04965\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 110ms/step - accuracy: 0.9963 - loss: 0.0105 - val_accuracy: 0.9889 - val_loss: 0.0528 - learning_rate: 0.0010\n",
+ "Epoch 5/10\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 101ms/step - accuracy: 0.9961 - loss: 0.0122\n",
+ "Epoch 5: val_loss did not improve from 0.04965\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 111ms/step - accuracy: 0.9961 - loss: 0.0123 - val_accuracy: 0.9593 - val_loss: 0.2248 - learning_rate: 0.0010\n",
+ "Epoch 6/10\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 98ms/step - accuracy: 0.9961 - loss: 0.0121\n",
+ "Epoch 6: val_loss did not improve from 0.04965\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 108ms/step - accuracy: 0.9961 - loss: 0.0121 - val_accuracy: 0.9778 - val_loss: 0.1974 - learning_rate: 0.0010\n",
+ "Epoch 7/10\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 95ms/step - accuracy: 0.9940 - loss: 0.0102\n",
+ "Epoch 7: val_loss improved from 0.04965 to 0.04634, saving model to best_model.h5\n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stderr",
+ "text": [
+ "WARNING:absl:You are saving your model as an HDF5 file via `model.save()` or `keras.saving.save_model(model)`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')` or `keras.saving.save_model(model, 'my_model.keras')`. \n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 119ms/step - accuracy: 0.9940 - loss: 0.0103 - val_accuracy: 0.9926 - val_loss: 0.0463 - learning_rate: 0.0010\n",
+ "Epoch 8/10\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 97ms/step - accuracy: 0.9967 - loss: 0.0116\n",
+ "Epoch 8: val_loss did not improve from 0.04634\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 107ms/step - accuracy: 0.9967 - loss: 0.0117 - val_accuracy: 0.9889 - val_loss: 0.0786 - learning_rate: 0.0010\n",
+ "Epoch 9/10\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 99ms/step - accuracy: 0.9905 - loss: 0.0722\n",
+ "Epoch 9: val_loss improved from 0.04634 to 0.04627, saving model to best_model.h5\n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stderr",
+ "text": [
+ "WARNING:absl:You are saving your model as an HDF5 file via `model.save()` or `keras.saving.save_model(model)`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')` or `keras.saving.save_model(model, 'my_model.keras')`. \n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 120ms/step - accuracy: 0.9905 - loss: 0.0717 - val_accuracy: 0.9889 - val_loss: 0.0463 - learning_rate: 0.0010\n",
+ "Epoch 10/10\n",
+ "\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 97ms/step - accuracy: 0.9970 - loss: 0.0081\n",
+ "Epoch 10: val_loss improved from 0.04627 to 0.04424, saving model to best_model.h5\n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stderr",
+ "text": [
+ "WARNING:absl:You are saving your model as an HDF5 file via `model.save()` or `keras.saving.save_model(model)`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')` or `keras.saving.save_model(model, 'my_model.keras')`. \n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r\u001b[1m76/76\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 126ms/step - accuracy: 0.9971 - loss: 0.0081 - val_accuracy: 0.9889 - val_loss: 0.0442 - learning_rate: 0.0010\n",
+ "Restoring model weights from the end of the best epoch: 10.\n"
+ ]
+ }
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "source": [
+ "# **Model Evaluation**"
+ ],
+ "metadata": {
+ "id": "dA81JM_WwY95"
+ }
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "model.evaluate(test_ds)"
+ ],
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "id": "vpHXpVrn3Ywv",
+ "outputId": "769cd0da-c9f8-433f-a5ea-7497b0777260"
+ },
+ "execution_count": 23,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "\u001b[1m10/10\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 1s/step - accuracy: 0.9826 - loss: 0.3225\n"
+ ]
+ },
+ {
+ "output_type": "execute_result",
+ "data": {
+ "text/plain": [
+ "[0.1541966050863266, 0.9900000095367432]"
+ ]
+ },
+ "metadata": {},
+ "execution_count": 23
+ }
+ ]
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "fig, ax = plt.subplots(1, 2)\n",
+ "fig.set_size_inches(20, 8)\n",
+ "\n",
+ "train_acc = history.history['accuracy']\n",
+ "train_loss = history.history['loss']\n",
+ "val_acc = history.history['val_accuracy']\n",
+ "val_loss = history.history['val_loss']\n",
+ "\n",
+ "epochs = range(1, len(train_acc) + 1)\n",
+ "\n",
+ "ax[0].plot(epochs, train_acc, 'g-o', label='Training Accuracy')\n",
+ "ax[0].plot(epochs, val_acc, 'y-o', label='Validation Accuracy')\n",
+ "ax[0].set_title('Training and Validation Accuracy')\n",
+ "ax[0].legend(loc = 'lower right')\n",
+ "ax[0].set_xlabel('Epochs')\n",
+ "ax[0].set_ylabel('Accuracy')\n",
+ "\n",
+ "ax[1].plot(epochs, train_loss, 'g-o', label='Training Loss')\n",
+ "ax[1].plot(epochs, val_loss, 'y-o', label='Validation Loss')\n",
+ "ax[1].set_title('Training and Validation Loss')\n",
+ "ax[1].legend()\n",
+ "ax[1].set_xlabel('Epochs')\n",
+ "ax[1].set_ylabel('Loss')\n",
+ "plt.show()"
+ ],
+ "metadata": {
+ "id": "lrAa3SklclRy",
+ "colab": {
+ "base_uri": "https://localhost:8080/",
+ "height": 587
+ },
+ "outputId": "ae67a623-5d3c-4fe5-c933-e9870021d151"
+ },
+ "execution_count": 25,
+ "outputs": [
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ ""
+ ],
+ "image/png": "iVBORw0KGgoAAAANSUhEUgAABmYAAAK9CAYAAAA+DNPkAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAPYQAAD2EBqD+naQABAABJREFUeJzs3Xd4FOXexvHv7qZ3SggJCQRCCzUQiohBOKKAiHRRUYrt2BDsYkFEjxzLOYJg1xdRRAUE1COCiAYRUSlSpJeQUEJPIT3ZnfePmNWVAAkkmZT747XX5c4+O3PvMElm9jfP81gMwzAQERERERERERERERGRcmc1O4CIiIiIiIiIiIiIiEhNocKMiIiIiIiIiIiIiIhIBVFhRkREREREREREREREpIKoMCMiIiIiIiIiIiIiIlJBVJgRERERERERERERERGpICrMiIiIiIiIiIiIiIiIVBAVZkRERERERERERERERCqICjMiIiIiIiIiIiIiIiIVRIUZERERERERERERERGRCqLCjIjUeGPGjCEyMvKC3jt58mQsFkvZBqpk9u/fj8Vi4f3336/wbVssFiZPnux8/v7772OxWNi/f/953xsZGcmYMWPKNM/FHCsiIiIiIlJI12DnpmuwP+kaTESqKxVmRKTSslgsJXrEx8ebHbXGu++++7BYLOzZs+esbZ544gksFgubN2+uwGSld/jwYSZPnszGjRvNjlKs7du3Y7FY8PLyIjU11ew4IiIiIlKN6Bqs6tA1WPkqKo69/PLLZkcRkWrKzewAIiJn8+GHH7o8/+CDD1i+fPkZy6Ojoy9qO++88w4Oh+OC3vvkk0/y2GOPXdT2q4ORI0cyY8YM5s6dy6RJk4pt8/HHH9O2bVvatWt3wdu5+eabuf766/H09LzgdZzP4cOHeeaZZ4iMjCQmJsbltYs5VsrKnDlzqF+/PikpKSxYsIDbbrvN1DwiIiIiUn3oGqzq0DWYiEjVpsKMiFRaN910k8vzn3/+meXLl5+x/O+ysrLw8fEp8Xbc3d0vKB+Am5sbbm76Vdq1a1eaNm3Kxx9/XOxFwZo1a0hISODf//73RW3HZrNhs9kuah0X42KOlbJgGAZz587lxhtvJCEhgY8++qjSFmYyMzPx9fU1O4aIiIiIlIKuwaoOXYOJiFRtGspMRKq0nj170qZNG9avX0+PHj3w8fHh8ccfB+Dzzz+nf//+hIWF4enpSVRUFM8++yx2u91lHX8fs/avXZbffvttoqKi8PT0pHPnzqxdu9blvcWNb2yxWLj33ntZvHgxbdq0wdPTk9atW7N06dIz8sfHx9OpUye8vLyIiorirbfeKvGYyatWrWL48OE0bNgQT09PIiIiuP/++8nOzj7j8/n5+XHo0CEGDRqEn58fwcHBPPTQQ2fsi9TUVMaMGUNgYCBBQUGMHj26xMNljRw5kh07drBhw4YzXps7dy4Wi4UbbriBvLw8Jk2aRGxsLIGBgfj6+hIXF8f3339/3m0UN76xYRg899xzhIeH4+PjQ69evdi6desZ7z116hQPPfQQbdu2xc/Pj4CAAPr168emTZucbeLj4+ncuTMAY8eOdQ7VUDS2c3HjG2dmZvLggw8SERGBp6cnLVq04OWXX8YwDJd2pTkuzmb16tXs37+f66+/nuuvv54ffviBgwcPntHO4XAwffp02rZti5eXF8HBwfTt25d169a5tJszZw5dunTBx8eHWrVq0aNHD7755huXzH8dX7rI38eOLvp3WblyJXfffTf16tUjPDwcgMTERO6++25atGiBt7c3derUYfjw4cWOUZ2amsr9999PZGQknp6ehIeHM2rUKE6cOEFGRga+vr6MHz/+jPcdPHgQm83G1KlTS7gnRURERORC6RpM12A16RrsfI4dO8att95KSEgIXl5etG/fntmzZ5/R7pNPPiE2NhZ/f38CAgJo27Yt06dPd76en5/PM888Q7NmzfDy8qJOnTpcdtllLF++vMyyikjlolsMRKTKO3nyJP369eP666/npptuIiQkBCg8gfTz8+OBBx7Az8+P7777jkmTJpGens5LL7103vXOnTuX06dP889//hOLxcKLL77IkCFD2Ldv33nv2vnxxx9ZuHAhd999N/7+/rz66qsMHTqUpKQk6tSpA8Bvv/1G3759CQ0N5ZlnnsFutzNlyhSCg4NL9Lnnz59PVlYWd911F3Xq1OHXX39lxowZHDx4kPnz57u0tdvt9OnTh65du/Lyyy/z7bff8p///IeoqCjuuusuoPDkeuDAgfz444/ceeedREdHs2jRIkaPHl2iPCNHjuSZZ55h7ty5dOzY0WXb8+bNIy4ujoYNG3LixAneffddbrjhBm6//XZOnz7Ne++9R58+ffj111/P6Lp+PpMmTeK5557j6quv5uqrr2bDhg1cddVV5OXlubTbt28fixcvZvjw4TRu3JijR4/y1ltvcfnll7Nt2zbCwsKIjo5mypQpTJo0iTvuuIO4uDgALr300mK3bRgG1157Ld9//z233norMTExLFu2jIcffphDhw7xyiuvuLQvyXFxLh999BFRUVF07tyZNm3a4OPjw8cff8zDDz/s0u7WW2/l/fffp1+/ftx2220UFBSwatUqfv75Zzp16gTAM888w+TJk7n00kuZMmUKHh4e/PLLL3z33XdcddVVJd7/f3X33XcTHBzMpEmTyMzMBGDt2rX89NNPXH/99YSHh7N//37eeOMNevbsybZt25x3VmZkZBAXF8f27du55ZZb6NixIydOnOCLL77g4MGDxMTEMHjwYD799FP++9//uty19/HHH2MYBiNHjryg3CIiIiJSOroG0zVYTbkGO5fs7Gx69uzJnj17uPfee2ncuDHz589nzJgxpKamOm8qW758OTfccANXXHEFL7zwAlA4d+jq1audbSZPnszUqVO57bbb6NKlC+np6axbt44NGzZw5ZVXXlROEamkDBGRKuKee+4x/v5r6/LLLzcA48033zyjfVZW1hnL/vnPfxo+Pj5GTk6Oc9no0aONRo0aOZ8nJCQYgFGnTh3j1KlTzuWff/65ARhffvmlc9nTTz99RibA8PDwMPbs2eNctmnTJgMwZsyY4Vw2YMAAw8fHxzh06JBz2e7duw03N7cz1lmc4j7f1KlTDYvFYiQmJrp8PsCYMmWKS9sOHToYsbGxzueLFy82AOPFF190LisoKDDi4uIMwJg1a9Z5M3Xu3NkIDw837Ha7c9nSpUsNwHjrrbec68zNzXV5X0pKihESEmLccsstLssB4+mnn3Y+nzVrlgEYCQkJhmEYxrFjxwwPDw+jf//+hsPhcLZ7/PHHDcAYPXq0c1lOTo5LLsMo/Lf29PR02Tdr16496+f9+7FStM+ee+45l3bDhg0zLBaLyzFQ0uPibPLy8ow6deoYTzzxhHPZjTfeaLRv396l3XfffWcAxn333XfGOor20e7duw2r1WoMHjz4jH3y1/349/1fpFGjRi77tujf5bLLLjMKCgpc2hZ3nK5Zs8YAjA8++MC5bNKkSQZgLFy48Ky5ly1bZgDG119/7fJ6u3btjMsvv/yM94mIiIjIxdE12Pk/n67BClW3a7CiY/Kll146a5tp06YZgDFnzhznsry8PKNbt26Gn5+fkZ6ebhiGYYwfP94ICAg441rpr9q3b2/079//nJlEpHrRUGYiUuV5enoyduzYM5Z7e3s7///06dOcOHGCuLg4srKy2LFjx3nXO2LECGrVquV8XnTnzr59+8773t69exMVFeV83q5dOwICApzvtdvtfPvttwwaNIiwsDBnu6ZNm9KvX7/zrh9cP19mZiYnTpzg0ksvxTAMfvvttzPa33nnnS7P4+LiXD7LkiVLcHNzc969BYXjCY8bN65EeaBwTOqDBw/yww8/OJfNnTsXDw8Phg8f7lynh4cHUDjk1qlTpygoKKBTp07FdsE/l2+//Za8vDzGjRvnMvTAhAkTzmjr6emJ1Vr4Z89ut3Py5En8/Pxo0aJFqbdbZMmSJdhsNu677z6X5Q8++CCGYfD111+7LD/fcXEuX3/9NSdPnuSGG25wLrvhhhvYtGmTy7ABn332GRaLhaeffvqMdRTto8WLF+NwOJg0aZJzn/y9zYW4/fbbzxh/+q/HaX5+PidPnqRp06YEBQW57PfPPvuM9u3bM3jw4LPm7t27N2FhYXz00UfO137//Xc2b9583nHPRURERKTs6BpM12A14RqsJFnq16/vco3m7u7OfffdR0ZGBitXrgQgKCiIzMzMcw5LFhQUxNatW9m9e/dF5xKRqkGFGRGp8ho0aOA8yfyrrVu3MnjwYAIDAwkICCA4ONj55W1aWtp519uwYUOX50UXCCkpKaV+b9H7i9577NgxsrOzadq06RntiltWnKSkJMaMGUPt2rWdYxZffvnlwJmfr2iekbPlgcK5QEJDQ/Hz83Np16JFixLlAbj++uux2WzMnTsXgJycHBYtWkS/fv1cLrBmz55Nu3btnGPnBgcH89VXX5Xo3+WvEhMTAWjWrJnL8uDgYJftQeEFyCuvvEKzZs3w9PSkbt26BAcHs3nz5lJv96/bDwsLw9/f32V5dHS0S74i5zsuzmXOnDk0btwYT09P9uzZw549e4iKisLHx8elULF3717CwsKoXbv2Wde1d+9erFYrrVq1Ou92S6Nx48ZnLMvOzmbSpEnO8Z+L9ntqaqrLft+7dy9t2rQ55/qtVisjR45k8eLFZGVlAYXDu3l5eTkvOkVERESk/OkaTNdgNeEarCRZmjVrdsbNbn/Pcvfdd9O8eXP69etHeHg4t9xyyxnz3EyZMoXU1FSaN29O27Ztefjhh9m8efNFZxSRykuFGRGp8v5611KR1NRULr/8cjZt2sSUKVP48ssvWb58uXM8V4fDcd71/v3O/yLG3yYULOv3loTdbufKK6/kq6++4tFHH2Xx4sUsX77cOUHi3z/f2fKUtXr16nHllVfy2WefkZ+fz5dffsnp06dd5v6YM2cOY8aMISoqivfee4+lS5eyfPly/vGPf5To3+VCPf/88zzwwAP06NGDOXPmsGzZMpYvX07r1q3Ldbt/daHHRXp6Ol9++SUJCQk0a9bM+WjVqhVZWVnMnTu3zI6tkvj7hKVFivtZHDduHP/617+47rrrmDdvHt988w3Lly+nTp06F7TfR40aRUZGBosXL8YwDObOncs111xDYGBgqdclIiIiIhdG12C6BiuJqnwNVpbq1avHxo0b+eKLL5zz4/Tr189lLqEePXqwd+9e/u///o82bdrw7rvv0rFjR959990KyykiFcvN7AAiIuUhPj6ekydPsnDhQnr06OFcnpCQYGKqP9WrVw8vLy/27NlzxmvFLfu7LVu2sGvXLmbPns2oUaOcy8/VNfp8GjVqxIoVK8jIyHC5Y2vnzp2lWs/IkSNZunQpX3/9NXPnziUgIIABAwY4X1+wYAFNmjRh4cKFLl3fixt6qySZAXbv3k2TJk2cy48fP37GHVALFiygV69evPfeey7LU1NTqVu3rvN5aYbyatSoEd9++y2nT592uWOraJiGonwXa+HCheTk5PDGG2+4ZIXCf58nn3yS1atXc9lllxEVFcWyZcs4derUWXvNREVF4XA42LZt2zkn+qxVqxapqakuy/Ly8khOTi5x9gULFjB69Gj+85//OJfl5OScsd6oqCh+//33866vTZs2dOjQgY8++ojw8HCSkpKYMWNGifOIiIiISPnQNVjp6RqsUGW8Bitpls2bN+NwOFx6zRSXxcPDgwEDBjBgwAAcDgd33303b731Fk899ZSzx1bt2rUZO3YsY8eOJSMjgx49ejB58mRuu+22CvtMIlJx1GNGRKqlorti/noXTF5eHq+//rpZkVzYbDZ69+7N4sWLOXz4sHP5nj17zhgT92zvB9fPZxgG06dPv+BMV199NQUFBbzxxhvOZXa7vdRfeg8aNAgfHx9ef/11vv76a4YMGYKXl9c5s//yyy+sWbOm1Jl79+6Nu7s7M2bMcFnftGnTzmhrs9nOuCtq/vz5HDp0yGWZr68vwBmFg+JcffXV2O12Zs6c6bL8lVdewWKxlHis6vOZM2cOTZo04c4772TYsGEuj4ceegg/Pz/ncGZDhw7FMAyeeeaZM9ZT9PkHDRqE1WplypQpZ9yp9td9FBUV5TJWNcDbb7991h4zxSluv8+YMeOMdQwdOpRNmzaxaNGis+YucvPNN/PNN98wbdo06tSpU2b7WUREREQunK7BSk/XYIUq4zVYSVx99dUcOXKETz/91LmsoKCAGTNm4Ofn5xzm7uTJky7vs1qttGvXDoDc3Nxi2/j5+dG0aVPn6yJS/ajHjIhUS5deeim1atVi9OjR3HfffVgsFj788MMK7a58PpMnT+abb76he/fu3HXXXc6TyzZt2rBx48Zzvrdly5ZERUXx0EMPcejQIQICAvjss88uapzcAQMG0L17dx577DH2799Pq1atWLhwYanH/vXz82PQoEHOMY7/2oUe4JprrmHhwoUMHjyY/v37k5CQwJtvvkmrVq3IyMgo1baCg4N56KGHmDp1Ktdccw1XX301v/32G19//fUZPUuuueYapkyZwtixY7n00kvZsmULH330kctdXlBYjAgKCuLNN9/E398fX19funbtWuz8KQMGDKBXr1488cQT7N+/n/bt2/PNN9/w+eefM2HCBJdJJi/U4cOH+f7778+Y3LKIp6cnffr0Yf78+bz66qv06tWLm2++mVdffZXdu3fTt29fHA4Hq1atolevXtx77700bdqUJ554gmeffZa4uDiGDBmCp6cna9euJSwsjKlTpwJw2223ceeddzJ06FCuvPJKNm3axLJly87Yt+dyzTXX8OGHHxIYGEirVq1Ys2YN3377LXXq1HFp9/DDD7NgwQKGDx/OLbfcQmxsLKdOneKLL77gzTffpH379s62N954I4888giLFi3irrvuwt3d/QL2rIiIiIiUJV2DlZ6uwQpVtmuwv1qxYgU5OTlnLB80aBB33HEHb731FmPGjGH9+vVERkayYMECVq9ezbRp05w9em677TZOnTrFP/7xD8LDw0lMTGTGjBnExMQ456Np1aoVPXv2JDY2ltq1a7Nu3ToWLFjAvffeW6afR0QqDxVmRKRaqlOnDv/73/948MEHefLJJ6lVqxY33XQTV1xxBX369DE7HgCxsbF8/fXXPPTQQzz11FNEREQwZcoUtm/f7uz6fDbu7u58+eWX3HfffUydOhUvLy8GDx7Mvffe6/IFdmlYrVa++OILJkyYwJw5c7BYLFx77bX85z//oUOHDqVa18iRI5k7dy6hoaH84x//cHltzJgxHDlyhLfeeotly5bRqlUr5syZw/z584mPjy917ueeew4vLy/efPNNvv/+e7p27co333xD//79Xdo9/vjjZGZmMnfuXD799FM6duzIV199xWOPPebSzt3dndmzZzNx4kTuvPNOCgoKmDVrVrEXBUX7bNKkSXz66afMmjWLyMhIXnrpJR588MFSf5bifPLJJzgcDpehCP5uwIABfPbZZ3z99ddce+21zJo1i3bt2vHee+/x8MMPExgYSKdOnbj00kud75kyZQqNGzdmxowZPPHEE/j4+NCuXTtuvvlmZ5vbb7+dhIQE5zjUcXFxLF++nCuuuKLE+adPn47NZuOjjz4iJyeH7t278+23357xc+jn58eqVat4+umnWbRoEbNnz6ZevXpcccUVhIeHu7QNCQnhqquuYsmSJS55RURERMQ8ugYrPV2DFaps12B/tXTpUpYuXXrG8sjISNq0aUN8fDyPPfYYs2fPJj09nRYtWjBr1izGjBnjbHvTTTfx9ttv8/rrr5Oamkr9+vUZMWIEkydPdg6Bdt999/HFF1/wzTffkJubS6NGjXjuued4+OGHy/wziUjlYDEq060LIiLCoEGD2Lp1K7t37zY7ikilNXjwYLZs2VKi8cBFRERERM5F12AiIlLRNMeMiIiJsrOzXZ7v3r2bJUuW0LNnT3MCiVQBycnJfPXVV+otIyIiIiKlpmswERGpDNRjRkTERKGhoYwZM4YmTZqQmJjIG2+8QW5uLr/99hvNmjUzO55IpZKQkMDq1at59913Wbt2LXv37qV+/fpmxxIRERGRKkTXYCIiUhlojhkRERP17duXjz/+mCNHjuDp6Um3bt14/vnndUEgUoyVK1cyduxYGjZsyOzZs1WUEREREZFS0zWYiIhUBuoxIyIiIiIiIiIiIiIiUkE0x4yIiIiIiIiIiIiIiEgFUWFGRERERERERERERESkgmiOmQvkcDg4fPgw/v7+WCwWs+OIiIiIiJQrwzA4ffo0YWFhWK26v0vOT9dMIiIiIlLTlPi6yTDRypUrjWuuucYIDQ01AGPRokXnfc/3339vdOjQwfDw8DCioqKMWbNmndFm5syZRqNGjQxPT0+jS5cuxi+//OLyenZ2tnH33XcbtWvXNnx9fY0hQ4YYR44cKVX2AwcOGIAeeuihhx566KGHHnrUqMeBAwdKdd4sNZeumfTQQw899NBDDz30qKmP8103mdpjJjMzk/bt23PLLbcwZMiQ87ZPSEigf//+3HnnnXz00UesWLGC2267jdDQUPr06QPAp59+ygMPPMCbb75J165dmTZtGn369GHnzp3Uq1cPgPvvv5+vvvqK+fPnExgYyL333suQIUNYvXp1ibP7+/sDcODAAQICAi7g04uIiIiIVB3p6elEREQ4z4NFzkfXTCIiIiJS05T0usliGIZRQZnOyWKxsGjRIgYNGnTWNo8++ihfffUVv//+u3PZ9ddfT2pqKkuXLgWga9eudO7cmZkzZwKF3ecjIiIYN24cjz32GGlpaQQHBzN37lyGDRsGwI4dO4iOjmbNmjVccsklJcqbnp5OYGAgaWlpusgQERERkWpP579SWjpmRERERKSmKek5cJUaHHrNmjX07t3bZVmfPn1Ys2YNAHl5eaxfv96ljdVqpXfv3s4269evJz8/36VNy5YtadiwobNNcXJzc0lPT3d5iIiIiIiIiIiIiIiIlEaVKswcOXKEkJAQl2UhISGkp6eTnZ3NiRMnsNvtxbY5cuSIcx0eHh4EBQWdtU1xpk6dSmBgoPMRERFRNh9KRERERERERERERERqjCpVmDHTxIkTSUtLcz4OHDhgdiQREREREREREREREali3MwOUBr169fn6NGjLsuOHj1KQEAA3t7e2Gw2bDZbsW3q16/vXEdeXh6pqakuvWb+2qY4np6eeHp6lt2HEREREREREREREZEyZxgGBQUF2O12s6NINWOz2XBzc8NisVzUeqpUYaZbt24sWbLEZdny5cvp1q0bAB4eHsTGxrJixQoGDRoEgMPhYMWKFdx7770AxMbG4u7uzooVKxg6dCgAO3fuJCkpybkeEREREREREREREal68vLySE5OJisry+woUk35+PgQGhqKh4fHBa/D1MJMRkYGe/bscT5PSEhg48aN1K5dm4YNGzJx4kQOHTrEBx98AMCdd97JzJkzeeSRR7jlllv47rvvmDdvHl999ZVzHQ888ACjR4+mU6dOdOnShWnTppGZmcnYsWMBCAwM5NZbb+WBBx6gdu3aBAQEMG7cOLp168Yll1xSsTtARERERERERERERMqEw+EgISEBm81GWFgYHh4eF92zQaSIYRjk5eVx/PhxEhISaNasGVbrhc0WY2phZt26dfTq1cv5/IEHHgBg9OjRvP/++yQnJ5OUlOR8vXHjxnz11Vfcf//9TJ8+nfDwcN5991369OnjbDNixAiOHz/OpEmTOHLkCDExMSxdupSQkBBnm1deeQWr1crQoUPJzc2lT58+vP766xXwiUVERERERERERESkPOTl5eFwOIiIiMDHx8fsOFINeXt74+7uTmJiInl5eXh5eV3QeiyGYRhlnK1GSE9PJzAwkLS0NAICAsyOIyIiIiJSrnT+K6WlY0ZEREQqWk5ODgkJCTRu3PiCvzAXOZ9zHWclPQe+sH42IiIiIiIiIiIiIiIiUmoqzIiIiIiIiIiIiIiIiFQQFWZERERERERERERERP7C7rATvz+ej7d8TPz+eOwOu9mRSi0yMpJp06aVuH18fDwWi4XU1NRyyySF3MwOICIiIiIiIiIiIiJSWSzcvpDxS8dzMP2gc1l4QDjT+05nSPSQMt+exWI55+tPP/00kydPLvV6165di6+vb4nbX3rppSQnJxMYGFjqbZVGfHw8vXr1IiUlhaCgoHLdVmWlwoyIiIiIiIiIiIiICIVFmWHzhmFguCw/lH6IYfOGseC6BWVenElOTnb+/6effsqkSZPYuXOnc5mfn5/z/w3DwG634+Z2/q/2g4ODS5XDw8OD+vXrl+o9cmE0lJmIiIiIiIiIiIiIVEuGYZCZl1miR3pOOvd9fd8ZRRnAuWz81+NJz0kv0foM48z1FKd+/frOR2BgIBaLxfl8x44d+Pv78/XXXxMbG4unpyc//vgje/fuZeDAgYSEhODn50fnzp359ttvXdb796HMLBYL7777LoMHD8bHx4dmzZrxxRdfOF//+1Bm77//PkFBQSxbtozo6Gj8/Pzo27evSyGpoKCA++67j6CgIOrUqcOjjz7K6NGjGTRoUAn/hc6UkpLCqFGjqFWrFj4+PvTr14/du3c7X09MTGTAgAHUqlULX19fWrduzZIlS5zvHTlyJMHBwXh7e9OsWTNmzZp1wVnKi3rMiIiIiIiIiIiIiEi1lJWfhd9Uv/M3LAEDg4OnDxL4QsmG+sqYmIGvR8mHEjuXxx57jJdffpkmTZpQq1YtDhw4wNVXX82//vUvPD09+eCDDxgwYAA7d+6kYcOGZ13PM888w4svvshLL73EjBkzGDlyJImJidSuXbvY9llZWbz88st8+OGHWK1WbrrpJh566CE++ugjAF544QU++ugjZs2aRXR0NNOnT2fx4sX06tXrgj/rmDFj2L17N1988QUBAQE8+uijXH311Wzbtg13d3fuuece8vLy+OGHH/D19WXbtm3OXkVPPfUU27Zt4+uvv6Zu3brs2bOH7OzsC85SXlSYERERERERERERERGpxKZMmcKVV17pfF67dm3at2/vfP7ss8+yaNEivvjiC+69996zrmfMmDHccMMNADz//PO8+uqr/Prrr/Tt27fY9vn5+bz55ptERUUBcO+99zJlyhTn6zNmzGDixIkMHjwYgJkzZzp7r1yIooLM6tWrufTSSwH46KOPiIiIYPHixQwfPpykpCSGDh1K27ZtAWjSpInz/UlJSXTo0IFOnToBhb2GKiMVZkRERERERERERESkWvJx9yFjYkaJ2v6Q+ANXz736vO2W3LiEHo16lGjbZaWo0FAkIyODyZMn89VXX5GcnExBQQHZ2dkkJSWdcz3t2rVz/r+vry8BAQEcO3bsrO19fHycRRmA0NBQZ/u0tDSOHj1Kly5dnK/bbDZiY2NxOByl+nxFtm/fjpubG127dnUuq1OnDi1atGD79u0A3Hfffdx1111888039O7dm6FDhzo/11133cXQoUPZsGEDV111FYMGDXIWeCoTzTEjIiIiIiIiIiIiItWSxWLB18O3RI+roq4iPCAcC5bi14WFiIAIroq6qkTrs1iKX8+F8PV1HRLtoYceYtGiRTz//POsWrWKjRs30rZtW/Ly8s65Hnd3d9fPZLGcs4hSXPuSzp1TXm677Tb27dvHzTffzJYtW+jUqRMzZswAoF+/fiQmJnL//fdz+PBhrrjiCh566CFT8xZHhRkRERERERERERERqfFsVhvT+04HOKM4U/R8Wt9p2Ky2Cs/2d6tXr2bMmDEMHjyYtm3bUr9+ffbv31+hGQIDAwkJCWHt2rXOZXa7nQ0bNlzwOqOjoykoKOCXX35xLjt58iQ7d+6kVatWzmURERHceeedLFy4kAcffJB33nnH+VpwcDCjR49mzpw5TJs2jbfffvuC85QXDWUmIiIiIiIiIiIiIgIMiR7CgusWMH7peA6mH3QuDw8IZ1rfaQyJHmJiuj81a9aMhQsXMmDAACwWC0899dQFDx92McaNG8fUqVNp2rQpLVu2ZMaMGaSkpJSot9CWLVvw9/d3PrdYLLRv356BAwdy++2389Zbb+Hv789jjz1GgwYNGDhwIAATJkygX79+NG/enJSUFL7//nuio6MBmDRpErGxsbRu3Zrc3Fz+97//OV+rTFSYERERERGp5OwOO6uSVpF8OplQ/1DiGsZVirv0REREBAzDTmrqKvLykvHwCCUoKA6LRX+nRaqyIdFDGNhiYKU+B//vf//LLbfcwqWXXkrdunV59NFHSU9Pr/Acjz76KEeOHGHUqFHYbDbuuOMO+vTpg812/n3Vo4frPD02m42CggJmzZrF+PHjueaaa8jLy6NHjx4sWbLEOaya3W7nnnvu4eDBgwQEBNC3b19eeeUVADw8PJg4cSL79+/H29ubuLg4Pvnkk7L/4BfJYpg9IFwVlZ6eTmBgIGlpaQQEBJgdR0RERESqqYXbFxZ7t970vtMr9G49nf9KaemYEZGa4PjxhezZM57c3D//Tnt6htO06XSCgyvHXfUiNUlOTg4JCQk0btwYLy8vs+PUSA6Hg+joaK677jqeffZZs+OUi3MdZyU9B9YcMyIiIiIildTC7QsZNm+YS1EG4FD6IYbNG8bC7QtNSiYiIiLHjy9k69ZhLkUZgNzcQ2zdOozjx/V3WkSqv8TERN555x127drFli1buOuuu0hISODGG280O1qlpsKMiIiIiEglZHfYGb90PAZndnAvWjZh6QTsDntFRxMREanxDMPOnj3joZi/00XL9uyZgGHo77SIVG9Wq5X333+fzp070717d7Zs2cK3335bKed1qUw0x4yIiIiISCW0KmnVGT1l/srA4ED6AVYlraJnZM+KCyYiIiKkpq46o6eMK4Pc3AOkpq6iVq2eFZRKRKTiRUREsHr1arNjVDnqMSMiIiIiUgkln04u03YiIiJSdvLySvb3t6TtRESkZlFhRkRERESkEgr1Dy3TdiIiIlJ2PDxK9ve3pO1ERKRmUWFGRERERKQSMgwDC5azvm7BQkRABHEN4yowlYiIiAAEBcXh6RkO5/hb7eZWm6Ag/Z0WEZEzqTAjIiIiIlLJfL7jc/p91A/jj8mD/16gKXo+re80bFZbhecTERGp6SwWG02bTj9nm4KCUyQnv1dBiUREpCpRYUZEREREpBJ5b8N7DJk3hFx7LgNbDGTukLk0CGjg0iY8IJwF1y1gSPQQk1KKiIhIcPAQWrdegMXi7rLc0zOC2rX7A7Br1z85dOg1M+KJiEgl5mZ2ABERERERKRy67N8//pvHv3scgFtibuGtAW/hZnXjutbXsSppFcmnkwn1DyWuYZx6yoiIiFQCdesOxmLxwDDyadLkRfz9O/8xfJmVffse4cCBl9m9+14cjnwiIiaYHVdERCoJ9ZgRERERETGZw3Bw/7L7nUWZiZdN5N1r38XNWngflc1qo2dkT25oewM9I3uqKCMiIlJJ5OYexOHIxGJxIzx8PLVq9cRisWGxWGjS5EUaNpwIwN6995OU9JK5YUWkVAzDTkpKPEePfkxKSjyGYTc70nn17NmTCRMmOJ9HRkYybdq0c77HYrGwePHii952Wa2nplBhRkRERETERHn2PG5edDPTfykcp/6VPq/w/BXPY7GcfTJhERERqRyysrYD4O3dDKvVw+U1i8VC48b/olGjSQDs2/cIiYnPV3hGESm948cX8vPPkWza1Ivt229k06Ze/PxzJMePLyyX7Q0YMIC+ffsW+9qqVauwWCxs3ry51Otdu3Ytd9xxx8XGczF58mRiYmLOWJ6cnEy/fv3KdFt/9/777xMUFFSu26goKsyIiIiIiJgkIy+Daz++lrlb5uJmdWPO4DlMuGSC2bFERESkhDIztwHg4xNd7OuFxZlniIycAkBCwhPs3/8MhmFUWEYRKZ3jxxeydeswcnMPuizPzT3E1q3DyqU4c+utt7J8+XIOHjx4xmuzZs2iU6dOtGvXrtTrDQ4OxsfHpywinlf9+vXx9PSskG1VByrMiIiIiIiY4ETWCa744AqW7V2Gj7sPX97wJSPbjTQ7loiIiJRCVlZhYcbXt9U520VGPkWTJv8GYP/+ySQkPKXijEgFMQwDuz2zRI+CgnR2774PKO7ns3DZnj3jKShIL9H6Svpzfs011xAcHMz777/vsjwjI4P58+dz6623cvLkSW644QYaNGiAj48Pbdu25eOPPz7nev8+lNnu3bvp0aMHXl5etGrViuXLl5/xnkcffZTmzZvj4+NDkyZNeOqpp8jPzwcKe6w888wzbNq0CYvFgsVicWb++1BmW7Zs4R//+Afe3t7UqVOHO+64g4yMDOfrY8aMYdCgQbz88suEhoZSp04d7rnnHue2LkRSUhIDBw7Ez8+PgIAArrvuOo4ePep8fdOmTfTq1Qt/f38CAgKIjY1l3bp1ACQmJjJgwABq1aqFr68vrVu3ZsmSJRec5Xzcym3NIiIiIiJSrKS0JK768Cp2ntxJbe/aLLlxCV3Du5odS0REREqpaCizs/WY+auGDR/FYnFn794HSUr6F4aRT5Mm/9bwpSLlzOHIYtUqvzJam0Fu7kF+/DGwRK3j4jKw2XzP287NzY1Ro0bx/vvv88QTTzh/L8yfPx+73c4NN9xARkYGsbGxPProowQEBPDVV19x8803ExUVRZcuXc67DYfDwZAhQwgJCeGXX34hLS3NZT6aIv7+/rz//vuEhYWxZcsWbr/9dvz9/XnkkUcYMWIEv//+O0uXLuXbb78FIDDwzH2RmZlJnz596NatG2vXruXYsWPcdttt3HvvvS7Fp++//57Q0FC+//579uzZw4gRI4iJieH2228/7+cp7vMVFWVWrlxJQUEB99xzDyNGjCA+Ph6AkSNH0qFDB9544w1sNhsbN27E3d0dgHvuuYe8vDx++OEHfH192bZtG35+ZXXcnEmFGRERERGRCrT12Fb6zOnDodOHiAiIYNlNy4gOPv+XOSIiIlK5GIbxl6HMzt1jpkhExANYLG7s2TOeAwdexDDyiYr6j4ozIsItt9zCSy+9xMqVK+nZsydQOIzZ0KFDCQwMJDAwkIceesjZfty4cSxbtox58+aVqDDz7bffsmPHDpYtW0ZYWBgAzz///Bnzwjz55JPO/4+MjOShhx7ik08+4ZFHHsHb2xs/Pz/c3NyoX7/+Wbc1d+5ccnJy+OCDD/D1LSxMzZw5kwEDBvDCCy8QEhICQK1atZg5cyY2m42WLVvSv39/VqxYcUGFmRUrVrBlyxYSEhKIiIgA4IMPPqB169asXbuWzp07k5SUxMMPP0zLli0BaNasmfP9SUlJDB06lLZt2wLQpEmTUmcoDRVmREREREQqyE8HfuKaudeQkpNCdN1olt20jIjACLNjiYiIyAXIzz9OQcEpwIKPT4sSvy88/D4sFnd2776bgwdfwTAKaNp0uoozIuXEavUhLi7j/A2B1NQf2LLl6vO2a9t2CUFBPUq07ZJq2bIll156Kf/3f/9Hz5492bNnD6tWrWLKlMI5qux2O88//zzz5s3j0KFD5OXlkZubW+I5ZLZv305ERISzKAPQrVu3M9p9+umnvPrqq+zdu5eMjAwKCgoICAgo8eco2lb79u2dRRmA7t2743A42Llzp7Mw07p1a2w2m7NNaGgoW7ZsKdW2/rrNiIgIZ1EGoFWrVgQFBbF9+3Y6d+7MAw88wG233caHH35I7969GT58OFFRUQDcd9993HXXXXzzzTf07t2boUOHXtC8PiWlOWZERERERCrAV7u+ovcHvUnJSeGS8EtYNXaVijIiIiJVWFFvGS+vxths3qV6b4MGd9G8+TuAhUOHZrB7990YhqMcUoqIxWLBZvMt0aN27avw9AwHzlYoteDpGUHt2leVaH2lLbjeeuutfPbZZ5w+fZpZs2YRFRXF5ZdfDsBLL73E9OnTefTRR/n+++/ZuHEjffr0IS8v7+J20F+sWbOGkSNHcvXVV/O///2P3377jSeeeKJMt/FXRcOIFbFYLDgc5fe7cPLkyWzdupX+/fvz3Xff0apVKxYtWgTAbbfdxr59+7j55pvZsmULnTp1YsaMGeWWRYUZEREREZFy9sGmDxj4yUCyC7Lp17Qf3978LXV86pgdS0RERC5CVlZhYcbXt2TDmP1dWNhttGw5C7Bw+PCb7Nx5h4ozIiazWGw0bTq96NnfXwWgadNpWCw2ysN1112H1Wpl7ty5fPDBB9xyyy3O4s7q1asZOHAgN910E+3bt6dJkybs2rWrxOuOjo7mwIEDJCcnO5f9/PPPLm1++uknGjVqxBNPPEGnTp1o1qwZiYmJLm08PDyw2+3n3damTZvIzMx0Llu9ejVWq5UWLUrew7A0ij7fgQMHnMu2bdtGamoqrVr9+Xu6efPm3H///XzzzTcMGTKEWbNmOV+LiIjgzjvvZOHChTz44IO888475ZIVVJgRERERESlXL//0MqMXj8Zu2Lm53c18fv3n+HqcfwJQERERqdyysrYDJZ9fpjj1648mOvpDwMqRI++xY8dYDOPcX3iKSPkKDh5C69YL8PRs4LLc0zOc1q0XEBw8pNy27efnx4gRI5g4cSLJycmMGTPG+VqzZs1Yvnw5P/30E9u3b+ef//wnR48eLfG6e/fuTfPmzRk9ejSbNm1i1apVPPHEEy5tmjVrRlJSEp988gl79+7l1VdfdfYoKRIZGUlCQgIbN27kxIkT5ObmnrGtkSNH4uXlxejRo/n999/5/vvvGTduHDfffLNzGLMLZbfb2bhxo8tj+/bt9O7dm7Zt2zJy5Eg2bNjAr7/+yqhRo7j88svp1KkT2dnZ3HvvvcTHx5OYmMjq1atZu3Yt0dGF831OmDCBZcuWkZCQwIYNG/j++++dr5UHFWZERERERMqBYRg8svwRHl7+MAAPdnuQ9we9j7vN/TzvFBERkaqgaCgzH5+L++IuJGQkrVrNBWwcPfoB27ePwuEoKIOEInKhgoOHcMkl+2nf/nuio+fSvv33XHJJQrkWZYrceuutpKSk0KdPH5f5YJ588kk6duxInz596NmzJ/Xr12fQoEElXq/VamXRokVkZ2fTpUsXbrvtNv71r3+5tLn22mu5//77uffee4mJieGnn37iqaeecmkzdOhQ+vbtS69evQgODubjjz8+Y1s+Pj4sW7aMU6dO0blzZ4YNG8YVV1zBzJkzS7czipGRkUGHDh1cHgMGDMBisfD5559Tq1YtevToQe/evWnSpAmffvopADabjZMnTzJq1CiaN2/OddddR79+/XjmmWeAwoLPPffcQ3R0NH379qV58+a8/vrrF533bCyGYRjltvZqLD09ncDAQNLS0ko9+ZGIiIiIVG/59nxu//J2Zm+aDcCLvV/k4e4Pm5zq4uj8V0pLx4yIVHc//RRKXt4ROnb8hYCALhe9vuPHP2PbtusxjAKCg68jOnoOVqtu6BApjZycHBISEmjcuDFeXl5mx5Fq6lzHWUnPgdVjRkRERESkDGXlZzH408HM3jQbm8XGrIGzqnxRRkRERFzl56eQl3cEAB+flmWyzuDgobRuvQCLxZ3jx+exbdv1OBzlM+G2iIiYS4UZEREREZEycir7FFd+eCVf7f4KLzcvFo1YxJiYMWbHEhERkTJWNL+Mp2c4bm5l1yuwbt2BtGmzCIvFgxMnFrJ163AcjjPnbxARkapNhRkRERERkTJwKP0QPWb14KcDPxHkFcS3N3/LgBYDzI4lIiIi5aCoMOPj06rM112nTn/atv0Cq9WLkye/4Pffh2C355T5dkRExDwqzIiIiIiIXKSdJ3Zy6f9dytbjWwnzD2PV2FV0b9jd7FgiIiJSTjIztwHg4xNdLuuvXbsPbdv+D6vVm1OnlvD77wOx27PLZVsiIlLxVJgREREREbkIaw+t5bJZl5GUlkTzOs356ZafaFOvjdmxREREpBxlZRUWZnx9y77HTJFata6gbdslWK2+pKR8w5Yt12C3Z5bb9kSqE8MwzI4g1VhZHF8qzIiIiIiIXKBv9n5Dr9m9OJF1gs5hnflx7I80CmpkdiwREREpZ5mZ5TeU2V/VqtWTdu2WYrP5kZr6HZs3X01BQUa5blOkKnN3dwcgKyvL5CRSnRUdX0XH24VwK6swIiIiNY3dYWdV0iqSTycT6h9KXMM4bFab2bFEpIJ8vOVjRi8eTb4jnyubXMnCEQvx8/AzO5aIiIiUs4KCDHJzEwHw9S2focz+KijoMtq1W87mzX1IS/uBzZv70q7dEtzcAsp92yJVjc1mIygoiGPHjgHg4+ODxWIxOZVUF4ZhkJWVxbFjxwgKCsJmu/DvgFSYERERuQALty9k/NLxHEw/6FwWHhDO9L7TGRI9xMRkIlIRXv3lVcYvHQ/A9W2uZ/ag2XjYPExOJSIiIhUhK2sHAO7u9XB3r1Mh2wwMvIT27b9l8+arSE9fzaZNV9Gu3VLc3YMqZPsiVUn9+vUBnMUZkbIWFBTkPM4ulAozIiIipbRw+0KGzRuGgeuYoofSDzFs3jAWXLdAxRmRasowDJ76/in+tepfAIzrMo5pfadhtWiEYBERkZoiK6toGLPy7y3zVwEBnWnffgWbNl3J6dO/sHnzlbRrtwx399oVmkOksrNYLISGhlKvXj3y8/PNjiPVjLu7+0X1lCmiwoyIiEgp2B12xi8df0ZRBsDAwIKFCUsnMLDFQA1rJlLNFDgKuPuru3lnwzsAPNfrOR6Pe1xDI4iIiNQwWVnbAPD1Ld/5ZYrj79+RmJjv2LSpN6dPr2PTpito3/7bCuu5I1KV2Gy2MvkCXaQ86NY+ERGRUliVtMpl+LK/MzA4kH6AVUmrKjCViJS3nIIchs8fzjsb3sFqsfL2NW/zRI8nVJQRERGpgTIzi3rMVHxhBsDPrz3t23+Pu3s9MjI2snFjL/LyjpuSRURELox6zIiIiJzH8czjrExcycr9K1m8c3GJ3vPO+nfwdfelQ2gH3Kz6cytSlaXlpDHwk4GsTFyJp82Tj4d+zODowWbHEhEREZP82WOmYocy+ys/vzbExMSzadM/yMzcwsaNPWnffgWenhc354GIiFQMfVMkIiLyN8cyj7Fy/0ri98ezMnElW49vLfU65v4+l7m/z8Xfw5+4RnFc3uhyekb2pGNoRxVqRKqQ5NPJ9PuoH5uObiLAM4Avrv+CyyMvNzuWiIiImMRuzyE7ey9gXo+ZIr6+0cTErGTjxn+QlbWNjRt7EhPzHZ6eYabmEhGR89M3QyIiUuMdyTjCyv0rWZlYWIzZfmL7GW3a1GtDz0Y9iWsUx/1L7yc5I7nYeWYAgjyDuKzhZaxKWkVabhpLdi9hye4lAPh5+HFZw8vo2agnl0deTmxoLO4293L9fCJyYfac2sNVH15FQmoCIb4hLL1pKTH1Y8yOJSIiIibKzt4NOLDZAvHwML93io9Pczp0WMnGjb3Izt75R8+Z7/DyCjc7moiInIMKMyIiUuMkn052FmFWJq5kx4kdZ7RpF9LOWTzp0agHdX3qOl9zs7oxbN4wLFhcijMWCueaeG/gewyJHoLdYWfz0c3E748nPjGeHxJ/IDUnlaV7lrJ0z1IAfN19uazhZc4eNZ3COqlQI1IJbEjeQL+P+nEs8xhRtaJYdtMyompHmR1LRERETPbnMGatKs1cc97eUcTErGTTpn+Qnb2bjRsvJybmO7y8GpkdTUREzsJiGEbxt/vKOaWnpxMYGEhaWhoBAQFmxxERkXM4lH7IOUdMfGI8u07ucnndgoX29ds7iyNxDeOo41PnnOtcuH0h45eO52D6QeeyiIAIpvWdxpDoIcW+x+6ws+XYFmeOlftXkpKT4tLGx92H7hHd6RnZk8sbXU7nBp3xsHlc4CcXkQvxXcJ3DPpkEKfzThNTP4alI5cS4hdidizT6fxXSkvHjIhURwkJT5OYOIX69W+lZct3zY7jIicniY0be5GTsw9Pz0bExHyPt3djs2OJiNQoJT0HVmHmAukiQ0Sk8jqYftBljpjdp3a7vG7BQkz9GGfxI65RHLW9a5d6O3aHnVVJq0g+nUyofyhxDeOwWW0lfr/DcPD7sd+dOVfuX8nJ7JMubbzdvOnesLuzaNQ5rDOebp6lzioiJbNg2wJGLhxJnj2PnpE9WTxiMYFegWbHqhR0/iulpWNGRKqjrVuv4/jx+URFvUxExINmxzlDTs5BZ88ZT88I2rf/Dh+fpmbHEhGpMVSYKWe6yBARqTwOpB1wFjfi98ezN2Wvy+tWi5UO9Ts4ixuXNbyMWt61TEp7dg7DwdZjW12GWTuRdcKljbebN90iujmHWevaoKsKNSJl5M11b3L3V3djYDAkeggfDfkILzcvs2NVGjr/ldLSMSMi1dGvv7YhK2srbdsuoU6dfmbHKVZubjKbNv2DrKwdeHiEERPzHT4+LcyOJSJSI6gwU850kSEiYp7E1ESX4sW+lH0ur1stVjqGdnQWLy5reBlBXkHmhL0IDsPB9uPbXYpOx7OOu7TxcvOiW3g3Z9Gpa3hXfZEsUkqGYTBl5RQmr5wMwD9j/8lrV79Wqh5wNYHOf6W0dMyISHXjcBSwapUPhpHPJZfsr9RzuOTlHWXjxivIytqKh0d92rdfga9vK7NjiYhUeyrMlDNdZIiIVJz9qftdihP7U/e7vG6z2IgNi3UWJ7pHdK+WQw8ZhsGOEzuI3x/vnKPmaOZRlzaeNk8uCb/EOUzbJeGX4O3ubVJikcrP7rBz39f38fq61wGY1GMSk3tOrjST+VYmOv+V0tIxIyLVTVbWTn79tSVWqw9xcaexWKxmRzqnvLzjbNrUm8zMzbi7B9O+/Xf4+bUxO5aISLVW0nNgtwrMJCIicl6GYTgLMUXFh8S0RJc2NouNTmGdnMWH7g27E+BZ/b/wsVgsRAdHEx0czV2d78IwDHae3OlStDqScaRwvprElQB42Dy4JPwSZ9GqW3g3FWpE/pBbkMvNi25m/rb5WLAwo98M7ulyj9mxREREpJLKzNwGgI9PdKUvygB4eAQTE/MdmzZdSUbGb2zc2JOYmBX4+bU3O5qISI2nHjMXSHd/iYiUDcMw2Jeyz6W4cCD9gEsbN6sbncM6O4sLl0Zcir+nv0mJKy/DMNh9andhUeuPR3JGsksbD5sHXRp0cQ7zdmnEpfi4+5iUWMQ8p3NPM/jTwaxIWIG71Z05Q+ZwXevrzI5Vqen8V0pLx4yIVDeJic+TkPAEISE3ER39odlxSiw/P4XNm/tw+vRa3Nxq0779cvz9O5odS0SkWtJQZuVMFxkiIhfGMAz2pux1Fg5WJq7kYPpBlzZuVrczigd+Hn4mJa66DMNgz6k9LkWvQ6cPubRxt7rTpUEXl6KXr4evSYlFKsaxzGP0+6gfG5I34Ofhx6IRi+jdpLfZsSo9nf9KaemYEZHqZtu2mzh27CMaN36eRo0mmh2nVAoK0ti8uS/p6T/j5hZEu3bLCAjoYnYsEZFqR4WZcqaLDBGRkvlrL46i4sDh04dd2rhb3eka3tVluC0VB8reX3snxScWFsaKK4p1DuvsMkycimJSnSSkJHDVnKvYc2oPwT7BLBm5hE5hncyOVSXo/FdKS8eMiFQ369bFkpGxgTZtFlO37kCz45RaQUE6mzdfTXr6amy2ANq1W0pgYDezY4mIVCsqzJQzXWSIVB92h51VSatIPp1MqH8ocQ3jsFltZseqNEq7fwzDYNfJXS5zxBQ3nFbXBl3pGdmTnpE9uST8Eg2nZQLDMEhITWDl/pXOQk1SWpJLGzerG53COjmLZt0jupdoGDn9XJ2d9o15Nh/dTJ85fTiScYRGgY345uZvaF6nudmxqgyd/0pp6ZgRkerEMBysWuWHw5FNly478fGpmucQBQUZbNlyDWlpK7HZ/GjbdglBQXFmxxIRqTZUmClnusgQqR4Wbl/I+KXjXXoNhAeEM73vdIZEDzExWeVQkv1jGAY7Tuxw6RFzNPOoy3o8bZ5cEn6JsxfGJeGXaAL6Smp/6n6XYeb2p+53ed1msREbFuscZu6yhpcR4On6d1A/V2enfWOeHxJ/4NqPryUtN4229dqy9KalhPmHmR2rStH5r5SWjhkRqU6ysxP45ZcmWCwexMVlYrW6mR3pgtntmWzZci2pqd9htfrQtu1X1KrV0+xYIiLVQpUpzLz22mu89NJLHDlyhPbt2zNjxgy6dCl+jMv8/HymTp3K7NmzOXToEC1atOCFF16gb9++zjanT5/mqaeeYtGiRRw7dowOHTowffp0Onfu7GwzZswYZs+e7bLuPn36sHTp0hLn1kWGSNW3cPtChs0bhoHrr0ELFgAWXLegRn9Rer79c1vH20jNSWVl4kqOZR5zaeNp86RbRDd6NirsEdM1vCtebl4Vll3KTmJqorPgFr8/noTUBJfXrRYrsaGxzh41p7JPMXrxaP1cFUO/c8zz+Y7PGbFgBLn2XC5reBlf3vAlQV5BZseqcnT+K6WlY0ZEqpOTJ5ewZUt/fH3b0rnzZrPjXDS7PZvffx9ESso3WK3etGnzBbVra849EZGLVSUKM59++imjRo3izTffpGvXrkybNo358+ezc+dO6tWrd0b7Rx99lDlz5vDOO+/QsmVLli1bxgMPPMBPP/1Ehw4dABgxYgS///47b7zxBmFhYcyZM4dXXnmFbdu20aBBA6CwMHP06FFmzZrlXLenpye1atUqcXZdZIhUbXaHncjpkWfMr1HEgoUQvxCW37S8Rg4xZHfY6f1h7zN6vpyNl5sXl0Zc6uxF0aVBFxViqqmktCRW7l/pLNbsTdlb4vfW5J+r8/1MWbAQHhBOwviEGrdvytt7G97jjv/dgcNwcG2La/lk6CfqsXeBdP4rpaVjRkSqk6Skl9m372GCg6+jdetPzY5TJuz2HLZuHcapU19htXrRuvUi6tTpe/43iojIWVWJwkzXrl3p3LkzM2fOBMDhcBAREcG4ceN47LHHzmgfFhbGE088wT333ONcNnToULy9vZkzZw7Z2dn4+/vz+eef079/f2eb2NhY+vXrx3PPPQcUFmZSU1NZvHjxBWfXRYZI1Ra/P55es3uZHaPKuyXmFsZ2GEvnsM54unmaHUdMcDD9YOEcNfvj+XrP1xw6fcjsSFXa96O/p2dkT7NjVAuGYfDvH//N4989DhT+vnprwFu4VeFhR8ym818pLR0zIlKd7NhxC0eOzCIycjKRkU+bHafMOBy5bN06gpMnP8di8aB168+oW/cas2OJiFRZJT0HNu3KNC8vj/Xr1zNx4kTnMqvVSu/evVmzZk2x78nNzcXLy/UObG9vb3788UcACgoKsNvt52xTJD4+nnr16lGrVi3+8Y9/8Nxzz1GnTp2z5s3NzSU3N9f5PD09vWQfVEQqpeTTyedvBPi5+9XIgkNuQS4Z+Rnnbde7SW8ua3hZBSSSyio8IJyR7UYyst1IPt7yMTcuvPG876mJP1cl/Zm6YcENXNP8msL5mCIvJzwgvALSVT8Ow8EDyx5g+i/TAXis+2M8f8XzWCwWk5OJiIhIVZWVtR0AH59ok5OULavVk9at57Ft2w2cOLGQrVuH0KrVPIKDB5kdTUSkWjOtMHPixAnsdjshISEuy0NCQtixY0ex7+nTpw///e9/6dGjB1FRUaxYsYKFCxdit9sB8Pf3p1u3bjz77LNER0cTEhLCxx9/zJo1a2jatKlzPX379mXIkCE0btyYvXv38vjjj9OvXz/WrFmDzVb88CFTp07lmWeeKaNPL+XF7rCzKmkVyaeTCfUPJa5hnIaEkWIVOApK1O7LG7+skXevl7RHUah/aAWkkaqipMdDTfy5KunP1JHMI7z727u8+9u7AETVinLO39MzsicRgRHlHbXKy7PnMfbzsczdMheAV/q8woRLJpgbSkRERKo0wzDIzNwGgI9PK5PTlD2r1YNWrT5h+/abOH58Htu2DSc6+mPq1RtmdjQRkWqrSo3lMH36dG6//XZatmyJxWIhKiqKsWPH8n//93/ONh9++CG33HILDRo0wGaz0bFjR2644QbWr1/vbHP99dc7/79t27a0a9eOqKgo4uPjueKKK4rd9sSJE3nggQecz9PT04mI0JcjlcnC7QsZv3S8y5wh4QHhTO87XZMpi5PdYWf6L9N5fMXj52xXNN9DXMO4CkpWucQ1jCM8IJxD6YfOmKgctH+keDpuzq4k+ybMP4zXrn6NVUmrWJm4kg3JG9ibspe9KXv5v42F5zqNgxo7izSXN7qcRkGNKvqjVGoZeRkMmzeMZXuX4WZ14/2B7zOy3UizY4mIiEgVl5eXjN2eDtjw8WlmdpxyYbW6Ex39ERaLO8eOfcS2bdcDH1Gv3gizo4mIVEtWszZct25dbDYbR4+6ToJ79OhR6tevX+x7goODWbx4MZmZmSQmJrJjxw78/Pxo0qSJs01UVBQrV64kIyODAwcO8Ouvv5Kfn+/S5u+aNGlC3bp12bNnz1nbeHp6EhAQ4PKQymPh9oUMmzfsjIncD6UfYti8YSzcvtCkZFKZ7Dixg7hZcTz4zYPk2nNpF9IOyx///VXR82l9p9XYHlc2q43pfQuHANL+kZLScXN2Jdk3r/Z7lYEtB/LyVS+z9va1nHrkFF/d+BUPX/owncM6Y7PYSEhNYNbGWYxePJrI6ZE0nt6YMYvH8P7G99mfur+iP1alciLrBFd8cAXL9i7Dx92HL2/4UkUZERERKRNFvWW8vaOwWqvvkLxWqxvR0bOpX38MYGfbths5cmSO2bFERKol0wozHh4exMbGsmLFCucyh8PBihUr6Nat2znf6+XlRYMGDSgoKOCzzz5j4MCBZ7Tx9fUlNDSUlJQUli1bVmybIgcPHuTkyZOEhmpInqrI7rAzfun4Yu9ALlo2YekE7A57RUeTSsLusPPS6peIeTOGNQfX4O/hzzsD3mHjPzey4LoFNAho4NI+PCCcBdctqPE9rYZED9H+kVLTcXN2pd03gV6BXN3sal688kV+vf1XTj16iiU3LuHR7o/StUFXbBYb+1P3M3vTbMZ+PpbG0xvTaFojRi8ezazfZrEvZR+GcebfxuooKS2JuFlx/HroV2p71+a7Ud/Rt2lfs2OJiIhINZGVVViY8fWtfsOY/Z3FYqNFi/cIDb0NcLBjxyiSk983O5aISLVjMUy8Yv/0008ZPXo0b731Fl26dGHatGnMmzePHTt2EBISwqhRo2jQoAFTp04F4JdffuHQoUPExMRw6NAhJk+eTEJCAhs2bCAoKAiAZcuWYRgGLVq0YM+ePTz88MN4eXmxatUq3N3dycjI4JlnnmHo0KHUr1+fvXv38sgjj3D69Gm2bNmCp2fJ7nxIT08nMDCQtLQ09Z4xWUnH7f9+9Pc1bk4DgW3HtzH287H8euhXAPo27cvb17ztMk+D5iY6N+0fuRA6bs6urPbN6dzT/HTgJ+L3xxOfGM+6w+vOmD8rIiDCOexZz8ieNKnVBIvFcpY1Vk3bjm+jz5w+HEw/SERABMtuWkZ0cPWalLey0PmvlJaOGRGpLnbtuovDh9+kYcOJNGnyvNlxKoRhONi9+14OH34DgObN3yYs7HaTU4mIVH4lPQc2dY6ZESNGcPz4cSZNmsSRI0eIiYlh6dKlhISEAJCUlITV+mennpycHJ588kn27duHn58fV199NR9++KGzKAOQlpbGxIkTOXjwILVr12bo0KH861//wt3dHQCbzcbmzZuZPXs2qamphIWFcdVVV/Hss8+WuCgjlcPuk7tZsG0Bb61/q0Ttk08nl3MiqUwKHAW8tPolJq+cTJ49j0DPQF7p8wpjYsac8aWkzWpT0e4ctH/kQui4Obuy2jf+nv70adqHPk37AIXzq/x04CdW7l9JfGI8vx76lQPpB/hw84d8uPlDABr4N3Ap1DSt3bRKF2rWHFhD/7n9SclJIbpuNMtuWuZSeBcREREpC0VDmfn4VP8eM0UsFivNmr2GxeLGoUMz2LXrDgwjnwYN7jY7mohItWBqj5mqTHd/mWPXyV3M3zqf+dvms+noplK99/l/PM9jlz1Wpb+AkpL5/djvjP18LOsOrwOgf7P+vHXNW2cMHyQiUp1l5mWy5uAa4vfHszJxJb8c/IV8R75LmzD/MGeRpmdkT5rVblZl/k4u2b2EYfOGkV2QzSXhl/C/G/5HHZ86Zseq1nT+K6WlY0ZEqovVq4PJzz9BbOx6/P07mh2nQhmGwd69D3Hw4H8BaNp0OuHh95mcSkSk8irpObAKMxdIFxkVZ8eJHc5izJZjW5zLbRYbVzS5gqHRQ3km/hmSM5KLnWfmr/o27csrfV6hZd2W5R1bTJBvz+fF1S/yzMpnyHfkE+QVxPS+07m53c1V5otGEZHykpWfxZoDa1iZuJL4/fH8cugX8ux5Lm3q+9V36VHTok6LSvn788NNHzL287HYDTv9mvZj/vD5+Hr4mh2r2tP5r5SWjhkRqQ7y8o7z00/1AIiLy8Bmq3nnHIZhkJDwOElJ/wYgKuplIiIeNDmViEjlpMJMOdNFRvnafnw787cVFmN+P/a7c7mb1Y0rGl/B8FbDGdRykPPO2IXbFzJs3jAAl+KMhcIvkwa2HMiS3UvIs+fhZnVjXJdxTLp8EkFeQRX3oaRcbT66mTGLx/Dbkd8AuLbFtbzR/w3C/MNMTiYiUjll52fz88GfnT1qfj74M7n2XJc2Ib4hXB55OT0bFfaoaVm3pemFmv/89B8eWv4QADe1u4n/u/b/cLe5m5qpptD5r5SWjhkRqQ5SU39g48bL8fKK5JJLEsyOYxrDMNi//2kSE58FoHHjqTRq9JjJqUREKh8VZsqZLjLK3tZjW53FmG3HtzmXu1nduLLJlQxvNZyBLQdS27t2se9fuH0h45eO52D6QeeyiIAIpvWdxpDoIew5tYcHv3mQL3Z+AUCwTzD/+se/uKXDLZqQugrLs+cxddVUnlv1HAWOAmp712ZGvxnc0OYG0788FBGpSnIKcvjl4C/E748nPjGeNQfWnFGoqedbj8sbXe7sUdMquFWF/a41DINHv32Ul356CYAHLnmAl656CavFep53SlnR+a+Ulo4ZEakODh9+i1277qR27atp1+4rs+OYbv/+Kezf/zQAkZHPEBk5yeREIiKViwoz5UwXGRfPMAy2Ht/qHKZs+4ntztfcre5cGfVHMabFQGp51yrROu0OO6uSVpF8OplQ/1DiGsadUXRZtmcZ9y+737m9mPoxvNr3VeIaxZXdh5MKsfHIRsYsHuOcb2hQy0G80f8N6vvVNzmZiEjVl1OQw6+HfmXl/pXEJ8bz04GfyCnIcWkT7BNMj0Y9nHPUtApuVS6FkgJHAbd/eTvvb3wfgBd6v8DDlz6sAnwF0/mvlJaOGRGpDnbvHs+hQ68SHv4gTZu+bHacSiExcSoJCY8D0KjRU0RGPqPzMhGRP6gwU850kXFhDMNgy7EtzN86nwXbF7DjxA7nax42D66KuorhrYZzbYtry3WYsXx7Pq+vfZ2n458mLTcNgBGtR/DilS/SMLBhuW1XykaePY/nfniOqT9OpcBRQB3vOsy8eiYjWo/QyaCISDnJLchl7eG1hT1q9hcWarILsl3a1PGuw+WRf/aoaVOvzUUXarLysxixYAT/2/U/bBYb7wx4h7Edxl7UOuXC6Py38nvttdd46aWXOHLkCO3bt2fGjBl06dKl2LbvvPMOH3zwAb//XjhscGxsLM8//7xLe8MwePrpp3nnnXdITU2le/fuvPHGGzRr1qxEeXTMiEh1sGnTlaSkfEuLFu8RGnqL2XEqjaSkl9m372EAGjZ8jMaNn9f1uIgIKsyUO11klJxhGGw6uokF2xYwf9t8dp3c5XzNw+ZB36Z9GRY9jGtbXEugV2CFZjueeZxJ30/i7Q1v4zAceLt582j3R3m4+8P4uPtUaBYpmfWH1zP287FsObYFgGGthvHa1a9Rz7eeyclERGqWPHseaw+tZWXiSuL3x7P6wGqy8rNc2tT2rl3Yo6ZRTy6PvJx2Ie3OWaj5e8/XNsFtGPTpIFYfWI2Xmxfzhs1jQIsB5f3R5Cx0/lu5ffrpp4waNYo333yTrl27Mm3aNObPn8/OnTupV+/M86SRI0fSvXt3Lr30Ury8vHjhhRdYtGgRW7dupUGDBgC88MILTJ06ldmzZ9O4cWOeeuoptmzZwrZt2/Dy8jpvJh0zIlId/PRTOHl5h+jQ4ScCA7uZHadSOXhwOnv2TAAgPPxBoqJeUnFGRGo8FWbKmS4yzs0wDDYe2eicM2bPqT3O1zxtnvRt2pfhrYYzoMUAAjzN338bj2xk/NLx/JD4AwANAxvy0pUvMbzVcJ1UVBK5BblMWTmFF1a/gN2wE+wTzGtXv8bw1sPNjiYiIhQWatYfXk/8/nhWJq7kx6QfyczPdGlTy6sWPRr1cPaoaRfSzjnkaHFzxblZ3ShwFBDkFcSXN3zJZQ0vq9DPJK50/lu5de3alc6dOzNz5kwAHA4HERERjBs3jsceO//kzHa7nVq1ajFz5kxGjRqFYRiEhYXx4IMP8tBDDwGQlpZGSEgI77//Ptdff/1516ljRkSquoKCNH78MQiA7t1TcHcPMjVPZXTo0Ovs3n0PAA0a3EfTptP0PYqI1GglPQd2q8BMUs0ZhsGG5A3OnjF7U/Y6X/Ny86Jf034MbzWc/s37V4pizF/F1I8hfnQ8C7Yt4KHlD5GUlsSIBSOY2XAm0/tOp0NoB7Mj1mhrD61l7Odj2Xp8K1A47NyMfjMI9g02OZmIiBTxsHnQLaIb3SK6MTFuIvn2fNYnr3fOUfNj0o+k5KTw+c7P+Xzn5wAEegbSo1EPanvXZvam2Wess8BRAMCkHpNUlBE5h7y8PNavX8/EiROdy6xWK71792bNmjUlWkdWVhb5+fnUrl0bgISEBI4cOULv3r2dbQIDA+natStr1qwptjCTm5tLbm6u83l6evqFfiQRkUohM7NwbloPjzAVZc6iQYO7sVjc2LXrnxw69CqGkU+zZjOxlMO8gyIi1YkKM3JRDMNgffJ655wx+1L2OV/zcvPi6mZXFxZjmvXH39PfxKTnZ7FYGN56ONc0v4aXf3qZqT9OZVXSKmLfjuX2jrfz3D+eUyGgguUU5DA5fjIv/fQSDsNBPd96vNH/DYZEDzE7moiInIe7zZ1Lwi/hkvBLePSyRylwFLAheYOzR82qxFWk5abx5a4vz7keCxZe+fkV7ut6n7N3jYi4OnHiBHa7nZCQEJflISEh7Nix4yzvcvXoo48SFhbmLMQcOXLEuY6/r7Potb+bOnUqzzzzTGnji4hUWllZhYUZH59ok5NUbmFhd2CxuLNz560cPvwGhpFP8+ZvqTgjInIOKsxIqRmGwdrDa53FmP2p+52vebt50795f4a3Gs7Vza7Gz8PPvKAXyNvdm6cuf4oxMWN49NtH+fj3j3l7w9t8uvVTJveczD2d78Hd5m52zGrv54M/M/bzsew4Ufhlwo1tb+TVvq9Sx6eOyclERORCuFnd6NKgC10adOGR7o9Q4Chg45GNvLfhPd5c/+ZZ32dgcCD9AKuSVtEzsmfFBRapQf7973/zySefEB8fX6K5Y85m4sSJPPDAA87n6enpRERElEVEERFTZGVtA8DXt5XJSSq/0NCxWCzu7NgxmuTkdzGMAlq0eBeA1NRV5OUl4+ERSlBQHBaLbrYREVFhRkrEMAx+PfQr87fNZ8G2BSSmJTpf83H3oX+zP4sxvh6+JiYtOxGBEcwdOpe7O9/NfV/fx29HfuP+Zffz1vq3eKXPK/Rt2tfsiNVSdn42k76fxH9//i8Ow0F9v/q82f9NBrYcaHY0EREpQ25WNzqFdWL3yd3nLMwUST6dXAGpRKqmunXrYrPZOHr0qMvyo0ePUr9+/XO+9+WXX+bf//433377Le3atXMuL3rf0aNHCQ0NdVlnTExMsevy9PTE09PzAj+FiEjlUzSUmY+PCjMlUb/+TVgsbmzffhNHjrxPdvY+srP3kZf35xyCnp7hNG06neBgjYQhIjWb+hTKWTkMB2sOrOGBZQ/QaFojLnnvEv6z5j8kpiXi6+7LiNYjWDB8AccfPs684fMY3np4tSnK/NVlDS9j7e1reWfAOwT7BLPjxA76fdSPAR8PYPfJ3WbHq1Z+OvATMW/F8PKal3EYDm5udzNb796qooyISDUW6h96/kalaCdSE3l4eBAbG8uKFSucyxwOBytWrKBbt25nfd+LL77Is88+y9KlS+nUqZPLa40bN6Z+/fou60xPT+eXX3455zpFRKqTP3vMaCizkgoJuZ5WrT4BrKSl/eBSlAHIzT3E1q3DOH58oTkBRUQqCfWYERcOw8HPB392DlN2MP3PP6B+Hn4MaD6AYa2G0bdpX3zcfUxMWrFsVhu3dbyN4a2G8+wPzzL9l+n8b9f/WLZnGRMumcCTPZ4kwDPA7JhVVlZ+Fk9+9yTTfp6GgUGYfxhvXfMW1zS/xuxoIiJSzuIaxhEeEM6h9EMYGGe8bsFCeEA4cQ3jTEgnUnU88MADjB49mk6dOtGlSxemTZtGZmYmY8eOBWDUqFE0aNCAqVOnAvDCCy8wadIk5s6dS2RkpHPeGD8/P/z8/LBYLEyYMIHnnnuOZs2a0bhxY5566inCwsIYNGiQWR9TRKTC2O1Z5OTsB9RjprSCgwfj5laLgoKTxbxqABb27JlA3boDNayZiNRYKswIDsPBTwd+Yv7W+Xy2/TMOnT7kfM3fw58BLQYwvNVw+kT1wdvd28Sk5gv0CuTlq17m9o63c/+y+/l6z9e89NNLfLDpA6ZeMZXRMaOxanK7UlmVuIpbvriFPaf2ADAmZgz/veq/1PKuZXIyERGpCDarjel9pzNs3jAsWFyKMxYsAEzrOw2bVRftIucyYsQIjh8/zqRJkzhy5AgxMTEsXbqUkJAQAJKSkrBa/zxPfeONN8jLy2PYsGEu63n66aeZPHkyAI888giZmZnccccdpKamctlll7F06dKLmodGRKSqyMraCRi4udXBwyPY7DhVSmrqqrMUZYoY5OYeIDV1FbVq9aygVCIilYvFMIwzb02U80pPTycwMJC0tDQCAqpeTwm7w15YjNlWWIw5fPqw87UAzwCubXEtw6KH0adpH7zcdOF1Nkt2L+H+Zfez6+QuAGJDY3m136tcGnGpyckqv8y8TB5f8Tgzfp2BgUED/wa8M+Ad+jXrZ3Y0ERExwcLtCxm/dLxLb92IgAim9Z3GkGiNQV4ZVPXzX6l4OmZEpCo7evQjtm+/icDAODp0+MHsOFXK0aMfs337jedtFx09l5CQGyogkYhIxSnpObB6zNQgdoedH5N+dBZjjmQccb4W4BnAwBYDGd5qOFdFXYWnmybtLImrm11N7ya9mfHLDKb8MIX1yevp/n/dGdl2JP/u/W/CA8LNjlgprdy/klu+uIV9KfsAuK3Dbbx81csEegWanExERMwyJHoIA1sMZFXSKpJPJxPqH0pcwzj1lBERERFTZGZuBzSM2YXw8CjZ3IAlbSciUh2pMFMF2R32En9pYXfY+SHxB+Zvm8/C7Qs5mnnU+VqgZyCDWg5ieKvh9G7SW8WYC+Rh8+DBSx/kpnY38eR3T/Leb+/x0ZaPWLRjEY9f9jgPXvqgeh39ISMvg8e+fYzX1r4GFN4J/c6Ad+jTtI/JyUREpDKwWW30jOxpdgwRERERsrK2AeDrG21ykqonKCgOT89wcnMPQTFzCIIFT89wgoI0h6CI1FwqzFQxxQ3zER4QzvS+053DfBQ4CgqLMVvns3DHQo5lHnO2DfIKcinGeNg8KvwzVFchfiG8c+073NnpTsYvHc/qA6t58vsnefe3d/nPVf9hcMvBWCwWs2Oa5ruE77j1i1vZn7ofgDs63sFLV71EgKeGtRAREREREZHKJTOzsDCjHjOlZ7HYaNp0Olu3DgMsuBZnCr8Xadp0GhaLekaLSM2lOWYukBnjJS/cvpBh84a5TIoLf06M+9TlT3Hk9BEW7VjE8azjztdredVicMvBDG89nH80/oeKMRXAMAw++f0THvn2EWcRrVdkL6b3nU7bkLYmp6tYp3NP88jyR3hz/ZsANApsxLvXvkvvJr1NTiYiIiKloflCpLR0zIhIVeVw5PHDDz6AnUsuOYCXl4YpvxDHjy9kz57x5Ob+eXOxp2c4TZtOJzhYcwiKSPVU0nNgFWYuUEVfZNgddiKnR7r0lDmX2t61C4sxrQqLMe4293JOKMXJzMvkhdUv8NJPL5FTkIPVYuXO2DuZ0msKdXzqmB2v3C3fu5zbvryNpLQkAO7udDf/7v1v/D39TU4mIiIipaUv2aW0dMyISFWVmbmVtWvbYLP5c9llaTV69IuLZRh2UlN/4PffB2K3n6ZDh58JDOxqdiwRkXJT0nNgawVmkouwKmlViYoy/Zv155ubvuHIg0d499p36dO0j4oyJvL18GVKrylsv2c7w1sNx2E4eH3d6zSb0YyZv86kwFFgdsRykZaTxh1f3sFVc64iKS2JxkGN+W7Ud7zW/zUVZURERERERKRS++swZirKXByLxUatWr0ICCgsxmRmbjE5kYhI5aDCTBWRfDq5RO1Gth3JlVFXqhhTyUQGRTJv+Dy+H/097ULakZKTwrivxxHzZgwr9q0wO16ZWrpnKW3eaMM7G94BYFyXcWy+azO9GvcyOZmIiIiIiIjI+WVlbQfA1zfa5CTVh69vewAyMzeZnEREpHJQYaaKCPUPLdN21ZVh2ElJiefo0Y9JSYnHMOxmR3LRM7In6+9Yzxv936COdx22Ht9K7w97M/jTwexL2Veu2y7vfZOak8otn99Cv4/6cTD9IFG1oogfHc+r/V7Fz8OvTLclIpVfZf99LCIiIiJyNn/tMSNlw8+vsDCTkaHCjIgIgJvZAaRk4hrGER4QzqH0QxicOS2QBQvhAeHENYwzIV3lUFUmlXOzunFnpzsZ0XoEz6x8hpm/zmTxjsUs2b2EB7s9yONxj5d5IaO8981Xu77ijv/dweHTh7FgYXzX8Tz3j+fw9fC96HWLSNVTVX4fi4iIiIgU588eMyrMlBU/vxigsDBjGIaGiBORGk89ZqoIm9XG9L7TgcIizF8VPZ/Wdxo2q63Cs1UGx48vZOvWYS5fAgLk5h5i69ZhHD++0KRkZ1fLuxbT+k5j812bubLJleTZ85j641Saz2jOh5s+xGE4ymQ75blvUrJTGL14NNd8fA2HTx+mWe1mrBq7ilf6vqKijEgNVRV/H4uIiIiIFHE4CsjK2gmAj4+GMisrPj4tsVg8sNvTycnZb3YcERHTqTBThQyJHsKC6xbQIKCBy/LwgHAWXLeAIdE18y5kw7CzZ894KKYnUdGyPXsmVNphdFoFt2LZTcv4/PrPiaoVRXJGMqMWj6L7/3Xn10O/XtS6y3PffLHzC1q/3poPNn2ABQsPdnuQjXdupHvD7heVWUSqrqr++1hEREREJCcnAcPIxWr1xsurkdlxqg2r1R1f39YAZGRsNDeMiEgloKHMqpgh0UMY2GIgq5JWkXw6mVD/UOIaxtXYnjIAqamrzrgz25VBbu4Bfv21FW5uARWWq7QaAB9fGsTRjHCSMw5jd/zMhvVdSdpeh/CAcNxt7qVeZ0FBeon2TWrqKmrV6lmidZ7MOsn4peP5aMtHALSo04JZA2fRLaJbqfOJSPVS0t/HpfmdIyIiIiJSkYqGMfPxaYHFUnO/aykPfn7tycj4jYyMTQQHDzY7joiIqVSYqYJsVhs9I3uaHaPSyMtLLlG77Oxd5ZykbPgBzVymmDlJTtZJcspxmyXdh4u2L+Kur+7iaOZRrBYrD3V7iMk9J+Pt7l2O6USkqijp75KSthMRERERqWiZmdsA8PHR/DJlzde3PQCZmZtMTiIiYj4VZqTK8/AILVG7xo2n4ufXrpzTlK1dJ3fy5rq32HmycHzbUL/63N7xdro06FqiifIyMjaTkDDxvO3Otw9PZJ1g3Nfj+OT3TwCIrhvNrIGz6BretQSfQkRqipL+Pi5pOxERERGRipaVVViY8fVVYaas+fnFABrKTEQEVJiRaiAoKA4Pj3Dy8s42fI4FT89wGjZ8uMp1Q+5W52q6NhvPR5s/4tFvH+WXpGQWJz3LlU2uZFrfabQKPveJYu3afTh8+DVycw9R/JwPAG64uQWedR0Lti3g7q/u5njWcWwWG490f4RJl0/Cy83rwj+YiFRLQUFxeHqGn/N3jtXq47wgExERERGpbP4cyiza5CTVj59fYY+ZnJz9FBSknfO7CBGR6s5qdgCRi2Wx2AgOHn62VwFo2nRalSvKFLFarNzc/mZ2jdvFxMsm4mHzYPm+5bR7ox3jvx5PSnbKWd9rsdho2nR60bOztCpgw4ZuHD78Nobx5xepxzKPcd386xg+fzjHs47Tpl4bfr7tZ56/4nkVZUSkWH/+zjlbIRgcjiw2bOhKRoaGLxARERGRysUwHGRmFhVm1GOmrLm718LTsyFQOMKHiEhNpsKMVHm5uUc4enQ2ADZbgMtrnp7htG69gODgIWZEK1N+Hn48f8XzbL9nO4NaDsJu2Hn111dpNqMZb657E7vDXuz7goOH0Lr1Ajw9G7gs9/SMoEWLWdSufTWGkcuuXf9k+/Ybyc9P49PfP6X1662Zv20+NouNJ+OeZN3t6+gU1qkiPqqIVGHBwUMICup9xnJPzwgaN/4Xnp7hZGfvYv36rhw+/JZLQVhERERExEy5uQdxODKxWNzx9o4yO061VNRrRsOZiUhNp6HMpEozDINdu+6koOAUfn4d6NDhJ9LTfyYvLxkPj1CCguKqbE+Zs2lSqwmLRizi233fMmHpBLYe38pdX93FG+veYHrf6fSM7HnGe4KDh1C37kBSU1edsW/q1x/FgQP/JSFhIseOfcKuw1/y5KZMTmRBu5B2zBo4i46hHSv+g4pIleRw5JOR8RsAUVEv4+ER5vI7Jyzsn+zYMYaTJ//Hrl13kpLyHS1avIObW8B51iwiIiIiUr4yMwvnl/H2bobV6m5ymurJz689J09+qR70IlLjqTAjVdqxY3M5efJzLBZ3WrZ8H5vNi1q1epodq0L0btKbjXdu5M11bzLp+0lsPrqZXrN7MazVMF6+8mUaBTVyaW+x2IrdNxaLlYiIB/nlRCYcn0KwZyavdYD91n6MvmwRnm6eFfOBRKRaSE39noKCk7i7B9OgwXisVtdTDXf3OrRp8wUHD77Cvn2Pcvz4PE6fXk/r1vPw91cRWERERETMk5VVWJjx9dUwZuWlaL5J9ZgRkZpOQ5lJlZWbm8zu3eMAiIx8Gj+/diYnqnhuVjfu7XIvu8ft5u5Od2O1WFmwbQEtX2vJpO8nkZmX6Wxrd9iJ3x/Px1s+Jn5/vHPos+TTyQz+dDDXfTmZW9c52HI6AA8rNOdrdu8YSX5+qkmfTkSqouPH5wMQHDz0jKJMEYvFQkTEA3To8COeno3IydnLhg3dOHhwpoY2ExERERHTZGUVzS8TbXKS6svXt3Aos8zM33E4CkxOIyJiHhVmpEoqHMLsnxQUpODnF0tExKNmRzJVHZ86vNb/NTb+cyO9InuRU5DDsz88S8vXWvLxlo/5bNtnRE6PpNfsXty48EZ6ze5F5PRI7vv6Plq/3prPd36Ou9Wdhy+bwj/7HScq6hUsFndOnPiM9es7kp6+1uyPKCJVgMORz/HjCwEIDh5+3vYBAV3p1Ok36tYdhGHksWfPOLZuHaaCsIiIiIiYomgoMx8f9ZgpL97eTbDZ/DCMXLKzd5kdR0TENCrMSJV09OiHnDz5JRaLB9HRs896V3ZN0zakLStGreCz6z4jMiiSg+kHuXHhjQybP4yD6Qdd2h5MP8iMX2eQkpNCx9COrL9jPU9d/hQebh5EREygQ4fVeHk1Jicngd9+686BA9N0J7uInFNq6ncUFJzC3b0egYE9SvQed/datG69kKZNp/9REF7I+vUdSE//tZzTioiIiIj8yTAMDWVWASwWK76+hSOeaDgzEanJVJiRKic39xC7d98HQGTkM/j6tjY5UeVisVgYEj2EbXdvY0rPKViwnLN9oGcgq8eupm1IW5flAQGdiY3dQN26QzGMfPbuvZ/ffx9Mfv6p8owvIlXYsWPzgHMPY1Yci8VCePh9dOjwE15eTcjJ2c9vv13GgQOvqCAsIiIiIhUiL+8oBQWpgBVv7+Zmx6nW/PwKhzPLyNhkchIREfOoMCNVimEY7Nx5B3Z7Gv7+XYiIeMjsSJWWt7s3cY3iMDj3l5ppuWn8fOjnYl9zdw+idev5NGs2E4vFg5MnP2fdug6kpRXfXkRqLocjjxMnFgEQHHzdBa0jIKATnTptIDh42B8F4Qf4/fdBKgiLiIiISLkr6i1TONSWl8lpqjc/vxhAPWZEpGZTYUaqlCNH3ufUqSVYLJ60bPm+hjA7j+TTyRfdzmKx0KDBPXTsuAYvryhyc5PYuDGOpKSXMAxHWUUVkSouJWUFBQUpuLuHEBQUd8HrcXMLpFWreTRr9joWiycnT37BunUxpKWtKcO0IiIiIiKusrK2A5pfpiKox4yIiAozUoXk5Bxgz54JADRu/Cy+vtHmBqoCQv1Dy6ydv3/HP+5kH4FhFLBv3yNs2XIteXknLjamiFQDx48XDWM2DIvFdlHrKiwI30XHjj/j7d2M3NwDbNzYQwVhERERESk3mZmFPWZ8fPRdQ3nz9W0LWMnPP0pu7hGz44iImEKFGakSCocwux27PZ2AgEuIiHjA7EhVQlzDOMIDws86z4wFCxEBEcQ1LNnd7W5uAbRq9THNm7+FxeLJqVNfsX59B1JTfyzL2CJSxRQOY7YYgHr1LmwYs+L4+8cQG7ueevVu+EtBeIAKwiIiIiJS5oqGMvP1VY+Z8maz+eDt3QyAzEz1mhGRmkmFGakSkpPfIyVlGVarFy1bvn/Rd2PXFDarjel9pwOcUZwpej6t7zRs1pLvT4vFQljYHcTG/oq3d3Nycw+ycWNPEhOn6k52kRoqJeVbCgpS8fCoT2Bg9zJdt5ubP9HRH9G8+dtYrV6cOrWEdetiSE1dVabbEREREZGaLTOzaCgz9ZipCBrOTERqOhVmpNLLyUlk797CHjKNG/8LH58WJieqWoZED2HBdQtoENDAZXl4QDgLrlvAkOghF7ReP792xMauJyTkJsBOQsLjbN7cj7y8Y2WQWkSqkmPHym4Ys+IUFoRvp2PHX/D2bkFe3iE2buxFYuLzKgiLiIiIyEXLzz9Ffv5RAHx8Wpqcpmbw84sBICNjo6k5RETMopnTpVIrHMLsNuz20wQEdCc8fLzZkaqkIdFDGNhiIKuSVpF8OplQ/1DiGsaVqqdMcdzc/GjZ8gOCgnqxe/e9pKR8w7p1MURHz6VWrZ5lkl1EKjeHI9c5jFlwcNkNY1acwoLwOnbvvpujRz8kIeEJUlPjiY6eg4dHvXLdtoiIiIhUX1lZhb1lPD0b4ubmb3KamkE9ZkSkplOPGanUkpPfJiXlW6xWb1q2nKUhzC6CzWqjZ2RPbmh7Az0je150UaaIxWIhNPQWYmPX4uMTTV5eMps2XcH+/VMwDHuZbENEKq9Tp5Zjt6fh4RFa5sOYFcfNzY/o6A9o0WIWVqs3KSnLWbcuhpSU+HLftoiIiIhUT5mZhfPLaBizilPUYyYrayd2e7a5YURETKDCjFRa2dkJ7NnzIABNmkzFx6eZyYnkXHx9WxMbu5b69ccCDvbvf5pNm/qQm3vE7GgiUo6OHy8axmw4FkvFnVaEho4hNnYdPj6t/1IQfkYFYREREREptayswsKMr28rk5PUHB4eobi71wXsZGZuNTuOiEiFU2FGKiXDcLBz5604HJkEBsbRoME4syNJCdhsvrRs+X+0bPkBVqsPqakr/riTfYXZ0USkHBQOY/Y5APXqle8wZsXx9W1FbOyv1K9/C4UF4cls2nSVCsIiIiIiUiqZmYVDmanHTMWxWCz4+hYOZ5aZqeHMRKTmUWFGKqXDh98gNfV7rFafP4Yw06FaldSvfzOxsevx9W1Lfv5RNm26koSESTgcBWZHE5EydOrUN9jt6Xh4NCAgoJspGWw2H1q2fI+WLT/EavUlNfU71q1rz6lT35qSR0RERESqHvWYMUfRcGYZGRtNzSEiYgZ92y2VTnb2PvbufQSAJk1ewNs7yuREciF8fVvSseMvhIbeDhgkJj7Lpk1XkJt72OxoIlJG/hzGbJjpBfT69W8iNnYdvr7tyM8/xubNV7Fv35MqCIuIiIjIORUUnCY39wCgHjMVzc+vsMdMRoZ6zIhIzaPCjFQqhuFgx46xOBxZBAX1pEGDu82OJBfBZvOmRYu3iY6ei83mR1raD3/cyb7M7GgicpHs9hxThzErTmFB+GdCQ/8JGCQl/euPgvAhs6OJiIiISCWVlbUDAHf3ENzda5ucpmb5s8fMJgzDMDeMiEgFU2FGKpVDh2aSlvYDVqsvLVr8n+l3YEvZCAm5gdjY9fj5xZCff4LNm/uyb99E3ckuUoWlpCzDbj+Np2c4AQGXmB3HqbAg/CbR0R9js/n/URCO4eTJpWZHExEREZFKSMOYmcfHpyUWiwd2ezo5OfvNjiMiUqH0rbdUGllZe9i37zEAoqJextu7scmJpCz5+DSnQ4c1hIUV9oJKSvo3Gzf2JCfngMnJRORCHDtWNIzZ8EpZRA8Juf6PgnAH8vNPsGVLP/bufQyHI9/saCIiIiJSiWRmbgfAx0eFmYpmtbo7C2IazkxEaprK902K1EiGYWfHjjE4HNkEBV1BWNg/zY4k5cBm86J589do1WoeNlsA6emr/7iT/Suzo4lIKdjt2Zw8+QUAwcGVYxiz4vj4NKNDh58IC7sHgAMHXlBBWERERERc/NljRvPLmOHP4cw2mppDRKSiqTAjlcLBg6+Snr4am82Pli3fw2KxmB1JylG9esPp1GkDfn6xFBScYsuWa9iz5yHdyS5SRZw6tQy7PQNPz4YEBHQ1O845FRaEZ9K69QJstkDS039i3boYTpz40uxoIiIiIlIJZGYWFmbUY8Ycvr7tAcjMVI8ZEalZVJgR02Vl7SIh4XEAoqL+i5dXI5MTSUXw9o6iY8fVNGhwHwAHD/6H336LIzt7v7nBROS8jh8vGsZsWJUppAcHD6VTpw34+3emoOAUv/9+LXv2PIjDkWd2NBERERExid2eTU5OAgA+PuoxYwY/v8LCjIYyE5GaRoUZMdWfQ5jlUKvWVYSG3mZ2JKlAVqsnzZpNp3XrRbi5BXH69C+sX9+B48cXmx1NRM7Cbs/mxInCYczq1au8w5gVx9u7CR06/Eh4+AQADh78L7/91kMFYREREZEaKjt7F+DAza0WHh4hZsepkYoKMzk5CRQUpJmcRkSk4qgwI6Y6cOAV0tPXYLMF0KLFu1XmzmspW8HBg4iN/Q1//64UFKSydetgdu+eoDvZRSqhU6e+xuHIxNOzIf7+XcyOU2pWqwdNm75CmzaLVRAWERERqeH+OoyZvo8wh7t7bTw9IwDIyNhschoRkYqjwoyYJjNzOwkJTwLQtOkreHlFmJxIzOTtHUmHDj8QHv4gAIcOTee337qTnb3P5GQi8lfHjhUOY1av3nVV+uK1bt2BdOq0kYCAS/5SEB6Pw5FrdjQRERERqSBZWdsB8PXVMGZm8vOLASAjY6OpOUREKpIKM2IKh6OAHTvGYBi51K7dj/r1x5odSSqBwjvZX6ZNmy9xc6vN6dPrWLeuA8eOLTA7mogAdnsWJ09+CUBwcNUaxqw4Xl6NiIn5gYiIhwE4dOhVNmzoTnb2XpOTiYiIiEhF+GuPGTGP5pkRkZrI9MLMa6+9RmRkJF5eXnTt2pVff/31rG3z8/OZMmUKUVFReHl50b59e5YuXerS5vTp00yYMIFGjRrh7e3NpZdeytq1a13aGIbBpEmTCA0Nxdvbm969e7N79+5y+XxSvIMH/8Pp079iswXSvPnbVfquayl7dete88ed7Jdit6ezbdtwdu26B7s9x+xoIjXayZNLcDiy8PKKxN+/k9lxyoTV6k5U1Iu0bfs/3NzqkJGxnnXrOnLs2Hyzo4mIiIhIOfuzx4wKM2by9S0szGRmqjAjIjWHqYWZTz/9lAceeICnn36aDRs20L59e/r06cOxY8eKbf/kk0/y1ltvMWPGDLZt28add97J4MGD+e2335xtbrvtNpYvX86HH37Ili1buOqqq+jduzeHDh1ytnnxxRd59dVXefPNN/nll1/w9fWlT58+5OToS9+KkJm5lYSESQA0azYdL69wkxNJZeTlFUFMTDwNGz4GwOHDr/Pbb93IylIRVcQsx48XFiuCg6v2MGbFqVOn/x8F4e5/FISvY9euu1UQFhEREammHI58srN3AeDjo6HMzPTnUGZbcDgKzA0jIlJBLIZhGGZtvGvXrnTu3JmZM2cC4HA4iIiIYNy4cTz22GNntA8LC+OJJ57gnnvucS4bOnQo3t7ezJkzh+zsbPz9/fn888/p37+/s01sbCz9+vXjueeewzAMwsLCePDBB3nooYcASEtLIyQkhPfff5/rr7++RNnT09MJDAwkLS2NgICAi9kNNYrDUcBvv3Xj9Ol11KlzDW3afFHtvtyTsnfy5FJ27LiZ/PwT2Gx+NG/+DiEhJftZFZGyYbdnsnp1PRyOLDp2XEtAQPXoMfN3DkcB+/dPIilpKlB4kdiq1Tx8fJqZnEzEfDr/ldLSMSMilVlm5nbWrm2FzebHZZel67sJExmGgx9/DMRuz6Bz563qwSQiVVpJz4FN6zGTl5fH+vXr6d27959hrFZ69+7NmjVrin1Pbm4uXl5eLsu8vb358ccfASgoKMBut5+zTUJCAkeOHHHZbmBgIF27dj3rdou2nZ6e7vKQ0jtw4EVOn16Hm1sQzZu/pRMfKZE6dfrSqdNGAgN7YLdnsH37Dezc+U/s9myzo4nUGH8OY9YYf/9Ys+OUG6vVjSZNnqddu6W4uweTkbGR9es7cvTox2ZHExEREZEyVDSMmY9PS303YTKLxYqvbzsAMjI2mhtGRKSCmFaYOXHiBHa7nZCQEJflISEhHDlypNj39OnTh//+97/s3r0bh8PB8uXLWbhwIcnJyQD4+/vTrVs3nn32WQ4fPozdbmfOnDmsWbPG2aZo3aXZLsDUqVMJDAx0PiIiIi74s9dUGRlb2L9/MgBNm87A0zPM3EBSpXh6NqB9+xU0avQkYCE5+W02bOhKZuYOs6OJ1AjHj88DqucwZsWpXbvPHwXhy/8oCN/Izp23qyAsIiIiUk1kZm4DwMdHvTMqAz+/wnlmMjI0z4yI1AymzjFTWtOnT6dZs2a0bNkSDw8P7r33XsaOHYvV+ufH+PDDDzEMgwYNGuDp6cmrr77KDTfc4NLmQkycOJG0tDTn48CBAxf7cWoUhyOfHTtGYxj51KkzkJCQkWZHkirIanWjceNnadfuG9zd65GZuYX16ztx5MiHZkcTqdYKCjI4efIrAOrVu87kNBXH0zOM9u2/pVGjSRQWhN9lw4YuKgiLiIiIVANFPWY0bFbloMKMiNQ0phVm6tati81m4+jRoy7Ljx49Sv369Yt9T3BwMIsXLyYzM5PExER27NiBn58fTZo0cbaJiopi5cqVZGRkcODAAX799Vfy8/OdbYrWXZrtAnh6ehIQEODykJJLSvo3GRm/4eZWm+bN36wRd1tL+alduzedOm0kKOgfOByZ7Ngxih07bsFuzzQ7mki1dOrUVzgc2Xh5ReHn18HsOBWqsCD8DO3bL8fdPYTMzN9Zvz6WI0c+MDuaiIiIiFyErKyiHjPRJicRKJzbETSUmYjUHKYVZjw8PIiNjWXFihXOZQ6HgxUrVtCtW7dzvtfLy4sGDRpQUFDAZ599xsCBA89o4+vrS2hoKCkpKSxbtszZpnHjxtSvX99lu+np6fzyyy/n3a5cmNOnN5KYOAWAZs1m4ul59gKYSEl5eobSvv03REY+A1g5cmQW69d3ITNzq9nRRKqdY8cKhzGrV69mDGNWnFq1rvijIHwFDkcWO3aMZseOsSoIi4iIiFRBhmEnK6uwF7SGMqscfH3bABby84+Sm3v2qQZERKoLU4cye+CBB3jnnXeYPXs227dv56677iIzM5OxY8cCMGrUKCZOnOhs/8svv7Bw4UL27dvHqlWr6Nu3Lw6Hg0ceecTZZtmyZSxdupSEhASWL19Or169aNmypXOdFouFCRMm8Nxzz/HFF1+wZcsWRo0aRVhYGIMGDarQz18TOBx57NgxBsMooG7dIdSrd73ZkaQasVhsREZOon37FXh4hJKVtY316zuTnDwLwzDMjidSLRQUZHDq1BKgcH6ZmszTsz7t2y8jMnIKhQXh91UQFhEREamCcnIScThysFg88fZubHYcAWw2X7y9mwOQmanhzESk+jO1MDNixAhefvllJk2aRExMDBs3bmTp0qWEhIQAkJSURHJysrN9Tk4OTz75JK1atWLw4ME0aNCAH3/8kaCgIGebtLQ07rnnHlq2bMmoUaO47LLLWLZsGe7u7s42jzzyCOPGjeOOO+6gc+fOZGRksHTpUry8vCrss9cUiYnPk5m5CXf3ujRv/kaNvdNayletWj3p1GkjtWpdhcORzc6dt7BjxygKCjLMjiZS5Z08+T8cjhy8vZs6x32uyQoLwk8RE/Pd3wrC76kgLCIiIlJFZGYWDWPWAovFZnIaKaJ5ZkSkJrEY+hbhgqSnpxMYGEhaWprmmzmL06c3sGFDVwyjgFatPq1RE0aLOQzDQVLSv0lIeApw4O3dgtat5+Hn187saCJV1u+/D+HEiUU0bPg4TZr8y+w4lUpe3jG2bx9FSsoyAOrVG0nz5m/i5uZncjKR8qHzXyktHTMiUlklJb3Ivn2PUq/e9bRq9bHZceQPiYnPk5DwBPXq3UirVh+ZHUdE5IKU9BzY1B4zUn05HLnOIcyCg4erKCMVwmKx0qjR48TExOPh0YDs7J2sX9+Fw4ff1p3sIhegoOA0J08WDmOm3+Nn8vCoR7t2S2jceCpg49ixj1i/PlZ3+ImIiIhUcllZ2wHw8Yk2OYn8lZ9fDAAZGRtNzSEiUhFUmJFysX//s2RmbsHdPZhmzV4zO47UMEFBcXTqtJHata/GMHLZteufbN9+IwUF6WZHE6lSTp78EsPIxdu7Ob6+6nlWnMKC8GPExMTj6RlOdvYu1q/vyuHDb6kgLCIiIlJJ/TmUWSuTk8hfFQ1llpW1E7s9x+Q0IiLlS4UZKXPp6etISvo3AM2bv4GHR7DJiaQm8vCoS9u2X9KkyYtYLG4cO/YJ69fHcvr0b2ZHE6kyjh2bBxT2ltEcYecWFHQZsbG/Ubt2/z8KwneybdsNKgiLiIiIVDKGYTh7zPj6qjBTmXh4hOHuXhewk5W11ew4IiLlSoUZKVOFQ5iNBuzUq3c9wcFDzY4kNZjFYqVhw4eJifkBT8+GZGfvYcOGSzh06DXdyS5yHgUF6Zw6tRSA4GANY1YShQXhL2jS5CUsFjeOH/+Udes6cvr0BrOjiYiIiMgfcnMPYbefBmx4ezc1O478hcViwde3sNeMhjMTkepOhRkpU/v3TyYraxvu7iE0azbT7DgiAAQGdqNTp9+oU+daDCOP3bvvZdu268jPTwXAMOykpMRz9OjHpKTEYxh2cwOLVAJ/DmPWAl/fNmbHqTIKC8IPEROzCk/PRuTk7GXDhm4cPDjTWRDW75yz0745O+0bERGRspGVVTSMWTOsVg+T08jfFQ1npnkbRaS6czM7gFQf6em/kpT0IgDNm7+Ju3sdkxOJ/MndvTZt2izm4MHp7Nv3CMePL+D06fWEht7J4cMzyM096Gzr6RlO06bTCQ4eYmJiEXNpGLOLExh4CZ06/caOHWM5efJz9uwZR2pqPHXrDiQh4XH9zinG8eML2bNnvPZNMbRvREREyk7RMGY+PtEmJ5Hi+PnFAOoxIyLVn3rMSJmw23P+GMLMQb16IwkOHmR2JJEzWCwWIiIm0KHDary8IsnJSSAh4VGXL7qgsGv71q3DOH58oUlJRcxVUJCmYczKgLt7Ldq0WUTTptOwWNw5ceIzduwYpd85xTh+fCFbtw7TvimG9o2IiEjZysws6jGj+WUqoz97zGzWEOQiUq2pMCNlYv/+SWRl7cDDoz7Nmr1qdhyRcwoI6EzHjuuwWr3P0qLw5G/PngkaKkZqpBMnvuD/2bvz+Ljqev/j7zNrZsnWJt33Nd2LiICKIBaRcpFd8foTLV5RFK+AyhVkE/WyCAgiV7wqiCgKCpfrdeHeUgVBNgVK9yXd96TN1sxk9vP7YzhJY7ckncl3ltfz8cjjcZ2ezLxTvenMvOf7+dh2QsHgDIVCs0zHKWqWZWnMmC9p/vy/SHIf5qry/p1j22k1Nn5Jzt/DP/ypJP5u+LsBACB3nFFmoRDFTCEKBhtkWT6l0+2KxbaYjgMAecMoMxyz9vaXtW3bXZKkadP+U17vEMOJgKOLRJYrk+k6whW24vFtamt7QbW1pw1SKqAwNDdnx5jV1zPGLFcymZikI715nv2d8/e/nyCvt3awYhWEZLL1oNMgvfF3c3j8WwUAQH/Ytn3AiRlGmRUil8unUGimOjuXqrNzqQKBCaYjAUBeUMzgmKTTXVqz5lOSbA0ffqnq6s4xHQnok0RiV06vA0pFMtmmlpb/lSQNG3ax4TSlo6+/SyKRN/OcpHjxd3N4/FsFAEDfJJPNSqVaJFkKBqebjoPDCIXmqbNzqSKRtxiVD6BkUczgmGzadIO6utbJ5xulKVPuNR0H6DOfb2ROrwNKxb59v5VtJxUMzmKMWQ719XfJuHE3lt1YjUhklbZu/eZRr+Pv5vD4twoAgL6JRldLkioqJsrtDhpOg8MJh+drz55H1Nm51HQUAMgbihkMWFvbi9q+/buSpOnTf1R240VQ3GpqTpHfP0bx+A4dena/Jb9/jGpqThnsaIBRTU3ZMWaclsmtvv7OmTjxZlnW4XbRlCbbTmvPnof5uzmEvv7d8G8VAAB9wxiz4hAOz5MkdXa+ZTgJAOSPy3QAFKd0Oqq1axdJsjVixCINHbrQdCSgXyzLrSlT7nP+0z/+qSRpypR7y+5NQJS3ZLJVra3/J0mqr6eYySV+5xwefzeHx98NAAC5FY1mi5lyO4VbbJxiJhbbpFSq3XAaAMgPihkMyMaN16urq1F+/xhNnnyP6TjAgNTXX6BZs34jv390r9u93mGaNes3qq+/wFAywIy9e/9btp1UKDSbF6t5cLjfOX7/mLL/ncPfzeHxdwMAQO5EItlRZpyYKWxe7xD5/WMlSZ2dywynAYD8YJQZ+q2t7Xnt2JH99Ob06T+W11tjNhBwDOrrL1Bd3blqa3tB69Z9Tl1dazV58nd4owtlqbk5O8asvv4jhpOUrgN/5yQSu+TzjVRNzSmceBB/N0fC3w0AALnBiZniEQ7PUzy+TZ2dbzG2FUBJophBv6TTEa1Zc5kkaeTIz2jIkDMNJwKOnWW5VVt7mmpqTlNX11pFo2tMRwIGXTLZotbWxZIYY5Zvzu8cHIy/m8Pj7wYAgGOTTLYpkdgliRMzxSAcnq99+36nzs6lpqMAQF4wygz9snHj1xSLbZTfP06TJ99lOg6QU6HQLElSJLLScBJg8O3d+7RsO6VQaK5CoQbTcQAAAICcikazY8x8vtHyeKoMp8HRhELZPTORyFuGkwBAflDMoM9aW/+sHTu+L0maPv0nPJFByXGKmWiUYgblp7n515KkYcMYYwYAAIDSwxiz4hIOz5ckRSIrlMmkzIYBgDygmEGfpFKdWrs2O8Js1KjPaciQBYYTAbnnFDNdXRuUTncZTgMMnmRyn1pbn5XEGDMAAACUpkgke2KGMWbFIRCYJLc7rEwmpq6udabjAEDOUcygTzZuvFax2Gb5/eM1adKdpuMAeeH1DpPHM1SSzZ4ZlJWeMWbzFAxOMx0HAAAAyDlOzBQXy3IpFJojSersZJwZgNJDMYOjaml5Vjt3/kCS1NDwkDyeSsOJgPywLIs9MyhLTU1PSGKMGQAAAEpXJJItZoJBipli4Ywz6+xcajQHAOQDxQyOKJXq0Nq1n5YkjRr1BdXWnm44EZBf7JlBuUkk9qq1dYkkxpgBAACgNKXTEcXjWyQxyqyYhMPzJHFiBkBpopjBEW3Y8BXF41tVUTFRkybdbjoOkHfOp6c4MYNysXfvf0lKKxw+TsHgVNNxAAAAgJxzRlV7vfXy+eoMp0FfOSdmIhGKGQClh2IGh9XS8n/atetHkqSGhofl8YQNJwLyj1FmKDfNzb+WJNXXM8YMAAAApSkSWS2JMWbFJhSaLclSIrFbicQe03EAIKcoZnBIqVR79wiz0aP/VTU1pxpOBAwOp5iJxTYpnY4aTgPkVyLRrNbWP0mShg1jjBkAAABKUzSa3S8TCjHGrJi43SEFAtlT/YwzA1BqKGZwSI2N1yge365AYIomTfp303GAQePzDZPXWyfJVjS62nQcIK96xpi9Q4HAZNNxAAAAgLyIRLLFDCdmio8zzqyzc6nRHACQaxQzOMi+fX/U7t0PSbI0ffrDcrtDpiMBgyoYZJwZykNT0xOSpGHDGGMGAACA0uV86C4Y5MRMsQmH50nixAyA0kMxg16SyVatXfsvkqQxY65STc17DScCBh97ZlAOEokmtbX9WZJUX88YMwAAAJSmTCaurq5GSVIoxImZYtNzYoZiBkBpoZhBL42NVyuR2KlAYJomTvyW6TiAEU4x48whBkpRc/NTkjKqrHynAoFJpuMAAAAAeRGNrpOUkdtdLZ9vpOk46CfnxEw0ukbpdMxwGgDIHYoZdNu793+0Z88jkiw1NDwstztoOhJgBCdmUA6am7NjzOrrGWMGAACA0uWMMQuFZsiyLMNp0F8+3yh5PEMlpRWN8hodQOmgmIEkKZls0bp1n5UkjR37ZVVXv9twIsAcZ8dMLLZJ6XTEcBog9xKJPWpre14SY8wAAABQ2iKR7CSEYJAxZsXIsqwDxpktNZoFAHKJYgaSpMbGLymR2KVgsEETJtxqOg5glM9XJ693mCQpElltOA2Qez1jzE5QIDDBdBwAAAAgb3pOzFDMFCtnnBl7ZgCUEooZaO/e/9aePT+X5FJDw0/ldgdMRwKM69kzw1FplB7GmAEAAKBcOLtDg8EZhpNgoHpOzFDMACgdFDNlLpncp7VrnRFmX1VV1YmGEwGFgT0zKFXx+O7uMWbDhjHGDAAAAKUrk0kpGl0riVFmxezAEzO2bRtOAwC5QTFT5tav/6KSyT0KBmdqwoRbTMcBCoazZ4ZiBqVm794nJdmqrDxRFRXjTccBAAAA8iYW2yjbTsrlCqqiYpzpOBigYLBBluVVOt2uWGyL6TgAkBMUM2WsuflJNTX9UpL77RFmFaYjAQWDEzMoVU1N2TFmw4YxxgwAAAClLRJxxpg1yLJ4C6xYuVy+7tfonZ1LzYYBgBzhX6UylUg0a926KyRJ48b9m6qqTjCcCCgszmLIeHyLUqlOw2mA3IjHd6m9/QVJUn39RYbTAAAAAPnl7JdxXt+heIVC2XFmkQh7ZgCUBoqZMrV+/ZVKJpsVCs3WhAk3mY4DFByvd6i83uGSep7MA8WuuTk7xqyq6mRGOQAAAKDkRaOrJUnB4AzDSXCswuH5krJ7ZgCgFFDMlKGmpifU3PyEsiPMHpHL5TcdCShIjDNDqcn+7pfq6y82nAQAAADIv55RZpyYKXbhcPbEDKPMAJQKipkyk0g0af36L0iSxo+/XpWV7zCcCChcFDMoJfH4DrW3vyiJMWYAAAAofbadUTS6RhKjzEqBU8zEYpuUSrUbTgMAx45ipozYtq116z6vZHKvQqG5Gj/+BtORgILmFDPRKMUMil/PGLN3q6JirOk4AAAAQF7FYluVyURlWT5VVEwyHQfHyOsdIr8/+zqms3OZ4TQAcOwoZspIU9Pj2rv3SVmW5+0RZj7TkYCCFgxyYgalo6kpO8Zs2LCPGE4CAAAA5J+zKzQYnCaXy2M4DXKhZ5wZe2YAFD+KmTIRj+8+YITZjaqsnG82EFAEnBMz8fg2pVIdhtMAAxeLbVdHx18lMcYMAAAA5SEaXS1JCgZnGE6CXAmFssVMJEIxA6D4UcyUgewIs88plWpROHycxo27znQkoCh4vbXy+UZK6nlSDxSjvXuflCRVV79Xfv9ow2kAAACA/ItEnBMz7JcpFeHwfElSZ+dSozkAIBcoZspAU9Nj2rfvv2VZXjU0/FQul9d0JKBoOKdmGGeGYuaMMauvZ4wZAAAAyoMzyiwUopgpFc4os0hkhTKZlOE0AHBsKGZKXDy+U+vXf1GSNGHCzQqH5xpOBBQX9syg2MVi29TR8ZIkS/X1F5qOAwAAAOSdbduKRBhlVmoCgclyuULKZGLq6lpvOg4AHBOKmRKWHWH2WaVSrQqHj9fYsf9mOhJQdDgxg2LX3PwbSc4Ys1GG0wAAAAD5l0jsUjrdLsmlYHCa6TjIEctydX/gmHFmAIodxUwJ27PnUe3b9ztZlu/tEWYe05GAouMUM9EoxQyKU3MzY8wAAABQXpwdodkTFn7DaZBLzjizzs63DCcBgGPDO/UlKh7fofXr/1WSNGHCNxQOzzacCChOzqLIeHy7Uql2eTzVhhMBfReLbVFHxytijBkAoNzYdlptbS8okdgln2+kampOkWW5TccCMEgikex+Gef1HEpHODxfEidmABQ/ipkSZNu21q79jNLpdlVWvktjx37FdCSgaHm9NfL5RimR2KlIZJWqq082HQnos54xZu+T3z/ScBoAAAZHc/NTamz8kuLx7d23+f1jNGXKfaqvv8BgMgCDJRrNFjOhEMVMqQmFsidmIhFOzAAobowyK0G7dz+slpY/yrL8amh4mBFmwDFizwyKVVPTryVJw4YxxgwAUB6am5/SypUX9SplpOxEgZUrL1Jz81OGkgEYTJFIdpRZMDjDcBLkWjg8R5KlRGK3Eok9puMAwIBRzJSYWGybGhuvliRNnPhNPh0C5AB7ZlCMuro2a//+VyVZqqvj08EAgNJn22k1Nn5Jkn2oP5UkNTZeJdtOD2ouAIOPEzOly+0OKRCYKok9MwCKG8VMCcmOMPsXpdMdqqo6SWPHXmM6ElASgkFOzKD4OGPMampOld8/wnAaAADyr63thYNOyvRmKx7fpra2FwYtE4DBl0jsVTLZLEkKBhsMp0E+hMPZcWYUMwCKGcVMCdm168dqbf0/uVwVamj4KcstgRxhlBmKUXPzE5Kk+nrGmAEAykMisSun1wEoTtFodoyZ3z9ebnfIcBrkQzg8X5LU2bnUaA4AOBYUMyUiFtuiDRu+LEmaOPHbCganG04ElA7n+HsisVPJZJvZMEAfdHVt0v79f5PkYskxAKBs+Hwjc3odgOLEGLPSx4kZAKWAYqYE2LatNWs+rXR6v6qq3qMxY75kOhJQUjyeavn9YySxZwbFobn515KkmprT5PMNN5wGAIDBUVNzytvP2azDXGHJ7x+rmppTBjMWgEEWiWRPzASDMwwnQb44J2ai0TVKp2NmwwDAAFHMlICdO3+otrYlcrkCamh4mBFmQB707JlZZTgJcHRNTdkxZsOGMcYMAFA+LMutKVPuc/7TIa+ZMuVeXi8BJY4TM6XP5xslj2eopDQfngRQtChmipBtp9Xa+pz27Pml9uz5pRobsyPMJk26TcHgVMPpgNLEnhkUi66ujersfF2SS3V1jDEDAJSX+voLNGvWb+T3j+51u8sV0qxZv2HEJ1AGnA/TBYMUM6XKsizGmQEoeh7TAdA/zc1PqbHxS4rHt/e6PRicodGjv2goFVD6nGKGT+Og0PWMMXu/fL56w2kAABh89fUXqK7uXLW1vaCWlme0bdsd8niGqK7ufNPRAORZKtWhRGKHJEaZlbpweL7a2v6kzs6lpqMAwIBwYqaINDc/pZUrLzqolJGyczX37n168EMBZYITMygWjDEDACA71qy29jRNmHCjLMunRGKburoaTccCkGfRaHa/jM83Ul5vjdkwyCtOzAAodhQzRcK202ps/JIk+7DXNDZeJdtOD14ooIw4x+ATiV1KJlsNpwEOLRptVGfnG5LcfCoYAMrQAw88oAkTJqiiokInnniiXnvttcNeu3LlSl144YWaMGGCLMvSvffee9A1t9xyiyzL6vXV0NCQx58g99zukKqr3y1Jam1dbDgNgHyLRLLFDGPMSl84PF9Stpix7cO/VwYAhcp4MdOfFw/JZFK33nqrJk+erIqKCs2bN0/PPPNMr2vS6bRuvPFGTZw4UYFAQJMnT9Y3v/nNXr+kP/WpTx30AuNDH/pQ3n7GXGhre+GQJ2V62IrHt6mt7YVBywSUE4+nUn7/OEmcmkHhcsaY1daezhgzACgzjz/+uK655hrdfPPNeuONNzRv3jydeeaZampqOuT10WhUkyZN0u23364RI0Yc9n5nzZqlXbt2dX+9+OKL+foR8qa2doEkqbX1WcNJAORbNJrdLxMKMcas1AWDDbIsr9LpdsViW0zHAYB+M1rM9PfFww033KAf/vCHuv/++7Vq1Sp97nOf0/nnn68333yz+5o77rhDP/jBD/T9739fq1ev1h133KE777xT999/f6/7+tCHPtTrBcYvf/nLvP6sxyqR2JXT6wD0H3tmUOiam7NjzOrrGWMGAOXmnnvu0Wc+8xktWrRIM2fO1IMPPqhgMKiHHnrokNefcMIJ+s53vqNLLrlEfr//sPfr8Xg0YsSI7q+6urp8/Qh5U1t7hiSptfVPymRShtMAyKdIJFvMcGKm9Llcvu7/niMRxpkBKD5Gi5n+vnh49NFHdf3112vhwoWaNGmSrrjiCi1cuFB333139zUvvfSSzj33XJ199tmaMGGCLrroIn3wgx886CSO3+/v9QKjtrY2rz/rsfL5Rub0OgD91/Okj2IGhScaXf/24ku36usZYwYA5SSRSOj111/XggULum9zuVxasGCBXn755WO67/Xr12vUqFGaNGmSPv7xj2vr1q2HvTYej6ujo6PXVyGorDxebne10ul2dXa+bjoOgDxydswEg5yYKQc948yWGs0BAANhrJgZyIuHeDyuioqKXrcFAoFex+nf/e53a8mSJVq3bp0k6a233tKLL76os846q9f3Pffccxo2bJimT5+uK664Qvv27TtiXtMvMmpqTpHfP0aSdZgrLPn9Y1VTc8pgxgLKinNihmIGhahnjNkCeb1DDacBAAymvXv3Kp1Oa/jw4b1uHz58uHbv3j3g+z3xxBP105/+VM8884x+8IMfaNOmTTrllFO0f//+Q15/2223qbq6uvtr7NixA37sXLIst2prT5fEODOglKXTUcVimyRJoRAnZspBODxPUnbPDAAUG2PFzEBePJx55pm65557tH79emUyGS1evFhPPfWUdu3qGd/1ta99TZdccokaGhrk9Xp13HHH6aqrrtLHP/7x7ms+9KEP6Wc/+5mWLFmiO+64Q88//7zOOusspdPpw+Y1/SLDstyaMuU+5z/9459KkqZMuVeW5R7UXEA5oZhBIWtqyo4xGzbsYsNJAACl4qyzztLFF1+suXPn6swzz9Qf/vAHtbW16Yknnjjk9dddd53a29u7v7Zt2zbIiQ/PGWfW0rLYcBIA+RKNrpVky+MZKq+XfYvloOfEDMUMgOJjdJRZf913332aOnWqGhoa5PP5dOWVV2rRokVyuXp+jCeeeEK/+MUv9Nhjj+mNN97QI488orvuukuPPPJI9zWXXHKJPvzhD2vOnDk677zz9Lvf/U5/+9vf9Nxzzx32sQvhRUZ9/QWaNes38vtH97rd7x+jWbN+o/r6CwY9E1BOnFFmyeQeJZNHPmUHDKZodK0ikbdkWR7V1Z1nOg4AYJDV1dXJ7XZrz549vW7fs2ePRowYkbPHqamp0bRp09TY2HjIP/f7/aqqqur1VShqa7OTGjo6XlI6HTGcBkA+OGPMQqEZsqzDTRtBKXFOzMRiG5VKFcb4TADoK2PFzEBePNTX1+vpp59WJBLRli1btGbNGoXDYU2aNKn7mq9+9avdp2bmzJmjT3ziE7r66qt12223HTbLpEmTVFdXd9gXGFLhvMior79AJ520WfPm/VkzZjymefP+rJNO2kQpAwwCjycsv3+8JE7NoLA0NTHGDADKmc/n0/HHH68lS5Z035bJZLRkyRKdfPLJOXuczs5ObdiwQSNHFt9ey0Bgivz+8bLtpNra/mI6DoA8iERWSer5QB1Kn9c75O2x/1Jn5zLDaQCgf4wVM8fy4qGiokKjR49WKpXSk08+qXPPPbf7z6LRaK8TNJLkdruVyWQOe3/bt2/Xvn37iuYFRnZG8mkaPvxjqq09jfFlwCBinBkKUXNzdqRMff1HDCcBAJhyzTXX6Ec/+pEeeeQRrV69WldccYUikYgWLVokSbr00kt13XXXdV+fSCS0dOlSLV26VIlEQjt27NDSpUt7fVjtK1/5ip5//nlt3rxZL730ks4//3y53W597GMfG/Sf71hZltV9aoY9M0Bpck7MBIMzDCfBYOoZZ7bUaA4A6C+PyQe/5ppr9MlPflLvfOc79a53vUv33nvvQS8eRo8e3X3a5dVXX9WOHTs0f/587dixQ7fccosymYyuvfba7vs855xz9O1vf1vjxo3TrFmz9Oabb+qee+7RZZddJin7Ka9vfOMbuvDCCzVixAht2LBB1157raZMmaIzzzxz8P8SABSVUGiWWlr+oGh0lekogCQpElmjSGS5LMvLGDMAKGMf/ehH1dzcrJtuukm7d+/W/Pnz9cwzz3Tv9Ny6dWuvD7Dt3LlTxx13XPd/vuuuu3TXXXfp1FNP7R7xvH37dn3sYx/Tvn37VF9fr/e+97165ZVXVF9fnLsbhgw5Q7t3/0StreyZAUqR8xotFOLETDkJheZp377fKRJhzwyA4mK0mOnvi4dYLKYbbrhBGzduVDgc1sKFC/Xoo4+qpqam+5r7779fN954oz7/+c+rqalJo0aN0mc/+1nddNNNkrKnZ5YtW6ZHHnlEbW1tGjVqlD74wQ/qm9/8pvx+/6D+/ACKDydmUGiam50xZmfI6601nAYAYNKVV16pK6+88pB/9o/7NCdMmCDbto94f7/61a9yFa0g1NScLkmKRJYrHt8tvz93+3cAmJXJJBSNrpfEKLNy03NihmIGQHGx7KM9G8chdXR0qLq6Wu3t7QW11BJAfnV0/F1vvHGCvN5hes979hz9G4A8+9vf5igSWaHp0x/WyJGfMh0HQAnj+S/6qxD/N/P3vx+nzs6lmjHjFxo+/J9NxwGQI5HIKv3tb7Pkdlfqve9tl2VZpiNhkESj6/Xaa9PkclXove/dL5fL6GfQAaDPz4GN7ZgBgGIUCmXnFSeTTUok9hpOg3IXiaxSJLLi7TFm5x79GwAAKHO1tWdIEuPMgBITiWTHmAWDMyhlykwgMFkuV0iZTExdXetNxwGAPqOYAYB+cLtDqqiYKEmKRhlnBrN6xph9kDFmAAD0QW3tAklSa+uzRx3lBqB4sF+mfFmWS+HwXElSZ+dSs2EAoB8oZgCgn9gzg0LR1PSEJGnYsI8YTgIAQHGorj5FluVXPL5d0eha03EA5Eg0ulpS9sQMyk84PE8Se2YAFBeKGQDop2CQYgbmRSIrFY2ukmX5NHToh03HAQCgKLjdAVVXv0dS9tQMgNLQM8qMEzPlKByeL0mKRChmABQPihkA6CfneDzFDExyTssMGXKmvN4as2EAACgi7JkBSottp7tPwDk7QVFeQiHnxMxSs0EAoB8oZgCgn5xRZuyYgSm2bXfvl6mvZ4wZAAD94eyZaWv7szKZlOE0AI5VV9cm2XZcLleFKiommI4DA8LhOZIsJRK7lUjsMR0HAPqEYgYA+ik7t9hSMrlXiUST6TgoQ9kxZqtlWT7V1Z1jOg4AAEWlsvI4eTxDlE7v1/79r5mOA+AYRaPOGLMGWZbbcBqY4HaHFAhMlcSeGQDFg2IGAPrJ7Q6qomKiJMaZwYzmZmeM2Yfk8VQbTgMAQHGxLLdqa0+XxJ4ZoBREo6slOR+gQ7kKh51xZhQzAIoDxQwADIAzzoxiBoPNtu3u/TLDhjHGDACAgXDGmVHMAMUvEnFOzMw0nAQmhcPzJVHMACgeFDMAMADsmYEpkchydXWtlWX5NXQoY8wAABiI2tozJEkdHS8rldpvOA2AY+GcmAmFKGbKWc+JmaVmgwBAH1HMAMAABIOcmIEZzmmZoUPPksdTZTgNAADFKRCYpIqKibLtlNrb/2I6DoABsm37gBMzjDIrZ6FQtpiJRtconY4ZTgMAR0cxAwADcOAoM9u2DadBubBtW83Nv5Yk1dczxgwAgGPhnJppaVlsOAmAgYrHtymTiciyPAoEppiOA4P8/tHyeIZKSisaXWU6DgAcFcUMAAxAMNggyaVUqkXJZJPpOCgTkcgydXWtk8tVoaFD/8l0HAAAihp7ZoDi54wxCwSmyuXyGk4DkyzLYpwZgKJCMQMAA+B2BxQITJLEODMMHmeM2ZAhZ8njqTScBgCA4lZbe7okS9HoSsXjO03HATAAPWPM2C8DKRyeL0nq7HzLbBAA6AOKGQAYIPbMYDBlx5hlixnGmAEAcOy83qEKh98hSWptXWI4DYCBcEZWhUIUMxAnZgAUFYoZABigA/fMAPnW2blUXV2NjDEDACCHGGcGFLdIJDvKLBicYTgJCkFPMfMWu2ABFDyKGQAYIKeYiUYpZpB/zmmZIUPOlscTNpwGAIDSMGTIGZKk1tbFvIkHFBnbtjkxg16CwRmyLK/S6XbF41tNxwGAI6KYAYABOvDEDC/kkU+2bXfvlxk2jDFmAADkSlXVe+RyVSiR2NW9RBxAcUgmm5RKtUpyKRCYZjoOCoDL5eveN8Q4MwCFjmIGAAYoEJguyaVUqlWJxG7TcVDCOjvfVCy2US5XQEOHnm06DgAAJcPtrlB19SmSsqdmABSPSCR7WqaiYqLc7oDhNCgU4fB8SdlxZgBQyChmAGCA3O4KBQKTJbFnBvnlnJYZOvRsud0hw2kAACgt7JkBihNjzHAoPXtmlpoNAgBHQTEDAMeAPTPIN9u2u/fL1NczxgwAgFyrrc3umWlre06ZTNJwGgB9FYlkxw8GgzMMJ0Eh6SlmODEDoLBRzADAMQgGe/bMAPmwf//risU2yeUKaujQhabjAABQcsLhefJ4hiqd7lRHx6um4wDoI07M4FCcYiYW26hUqsNwGgA4PIoZADgGzokZihnki3NaZujQf2KMGQAAeWBZLtXWfkAS48yAYuLsmHGWvQOS5PUOld8/RpLU2bnMcBoAODyKGQA4BgcWM7ZtG06DUmPbdvd+mWHDGGMGAEC+OOPMWlsXG04CoC+SyRYlk3skScFgg+E0KDShUPbUTCTCODMAhYtiBgCOQTA4XZJb6XS7EoldpuOgxOzf/zfF41vkcoU0ZMhZpuMAAFCyamsXSJI6Ol5l9A1QBKLR7H4Zv3+sPJ5Kw2lQaMLh+ZKkzs6lRnMAwJFQzADAMXC5/AoEpkhinBlyr7n515Kkurpz5HYHDacBAKB0BQIT3n5Ol1Zb23Om4wA4ikgkW8wEgzMMJ0EhcvbMdHZyYgZA4aKYAYBjxJ4Z5MOBY8zq6y82nAYAgNLnnJphzwxQ+KLR7H6ZUIj9MjiYc2ImElmuTCZlNgwAHAbFDAAcI6eYiUYpZpA7+/e/pnh8K2PMAAAYJOyZAYpHJJItZoJBihkcLBCYLJcrpEwmpq6u9abjAMAhUcwAwDHixAzywTktU1f3YbndAcNpAAAofTU175fkUjS6RrHYdtNxAByBs2OGUWY4FMtyKRyeI4lxZgAKF8UMAByjYLCnmLFt23AalALbznTvl6mv/4jhNAAAlAevt1aVle+UJLW1LTGcBsDhpFL7FY9vlSSFQhQzODRnnFln51KjOQDgcChmAOAYBYPTJLmVTncoHt9hOg5KQEfHq4rHt8ntDmvIkA+ZjgMAQNlw9sy0tDDODChU0egaSZLXO1xe71DDaVCoQqF5kqRIhBMzAAoTxQwAHCOXy6dgcKok9swgN5zTMkOHniu3u8JwGgAAyodTzLS2PstJaKBAOWPMOC2DI+k5MUMxA6AwUcwAQA4cOM4MOBYHjjEbNowxZgAADKbq6nfL5QoomdyjSGSF6TgADiESWSVJCgZnGk6CQpbdMWMpkdilRKLJdBwAOAjFDADkQChEMYPc6Oh4RfH4drndlaqt/aDpOAAAlBWXy6/q6vdJyp6aAVB4nBMzwSAnZnB4bndIgcAUSZyaAVCYKGYAIAcoZpArTU1PSJLq6hhjBgCACUOGnCFJam1lzwxQiKLR7ImZUIgTMziynnFmS43mAIBDoZgBgBxwiplodBXzyDFgB44xq69njBkAACY4e2ba2p5XJpMwnAbAgdLpmLq6NkpilBmOLhyeJ4kTMwAKE8UMAORAIDBVluVROr1f8fg203FQpNrbX1IisVNud5WGDGGMGQAAJoRCc+T1DlMmE1VHxyum4wA4QFfXOkkZeTw18vmGm46DAuecmIlEKGYAFB6KGQDIAZfLp0BgmiTGmWHgmpudMWbnyeXyG04DAEB5siyXams/IIlxZkChiUSyY8yCwZmyLMtwGhS6UCh7YiYSWa10OmY4DQD0RjEDADly4DgzoL+yY8x+I0kaNowxZgAAmOSMM2ttfdZwEgAHikZXS2K/DPrG7x8tj2eIpDSv0wEUHIoZAMgRp5jhxAwGor39r0okdsntrlZt7Rmm4wAAUNacYqaj4zUlk21mwwDo5ry5HgzOMJwExcCyrO5xZp2dS41mAYB/RDEDADkSDFLMYOB6jzHzGU4DAEB5q6gY9/aY2oza2p4zHQfA2w4cZQb0RTicHWfW2cmeGQCFhWIGAHLkwFFmtm0bToNiYttpxpgBAFBgnBOs7JkBCkMmk1RX13pJUijEiRn0Tc+JGYoZAIWFYgYAciQQmCLL8iqd7lQ8vtV0HBSR9vYXlUjslsdT0z06BQAAmMWeGaCwdHVtkG0n5XKF5PePNR0HRaLnxMxSPkAJoKBQzABAjrhcXgWD0yUxzgz909TkjDE7nzFmAAAUiNra90tyqatrnWIxPnQDmObslwmFZsiyeDsLfRMMznj7A5TtfIASQEHhXzIAyCH2zKC/smPMnpQk1dczxgwAgELh8VSrqupdkjg1AxSCaHS1pOwb7UBfuVy+7p1EnZ1LzYYBgANQzABADoVC2Sd8FDPoq7a2F5RM7pHHU6va2g+YjgMAAA7AODOgcEQi2RMzzpvsQF/1jDNjzwyAwkExAwA5FAplT8xEoxQz6Jvm5gPHmHkNpwEAAAeqrT1DUraYse2M4TRAeXNOzIRCnJhB/4TD8yVRzAAoLBQzAJBDPaPMVvHiHUeVyaS6x5gNG8YYMwAACk1V1UlyuUJKJpsViSw3HQcoW7adPmCUGSdm0D89J2aWmg0CAAegmAGAHAoEpsiyfMpkoorFtpiOgwLX3v4XJZNN8niGqKbmdNNxAADAP3C5fKqpOVWS1NKy2HAaoHzFYluUycRkWX5VVEw0HQdFxilmYrGNSqU6DKcBgCyKGQDIIZfLo2BwuiT2zODompqyY8zq6y9gjBkAAAWKPTOAeT2nZabJ5fIYToNi4/UOld8/RpLU2bnMcBoAyKKYAYAcY88M+iKTSWnv3uwYs/p6xpgBAFConD0z7e1/UTodM5wGKE+RyCpJjDHDwIVC2VMzkQh7ZgAUBooZAMixnj0zFDM4vPb255VM7pXHM1Q1Ne83HQcAABxGKDRLXu9wZTJd6uh42XQcoCxFo9liJhSimMHAhMPzJUmdnRQzAAoDxQwA5JhzYsb5VBdwKD1jzC5kHAMAAAXMsizGmQGGRSLOKLMZhpOgWDl7Zjo7l5oNAgBv63cxM2HCBN16663aunVrPvIAQNHrGWW2WradMZwGhSiTSam52RljdrHhNAAA4GiGDMmOM2ttXWw4CVB+bNvmxAyOmVPMRCIrZNtpw2kAYADFzFVXXaWnnnpKkyZN0hlnnKFf/epXisfj+cgGAEUpEJgsy/Irk4kqFttsOg4KUFvbn5VK7ZPXW6eamtNMxwEAAEfhnJjZv//vSiZbDacByksisVPp9H5JbgUCU03HQZEKBCbL5Qopk+lSNLredBwAGFgxs3TpUr322muaMWOGvvjFL2rkyJG68sor9cYbb+QjIwAUFctyKxhskMSeGRxac3N2jFldHWPMAAAoBn7/6LdHKNlqa/uT6ThAWXFGRAcCU+Ry+QynQbGyLLfC4TmSGGcGoDAMeMfMO97xDn3ve9/Tzp07dfPNN+vHP/6xTjjhBM2fP18PPfSQbNvOZU4AKCo9e2YoZtBbJpNUc/NTkqRhwz5iOA0AAOgr9swAZjDGDLkSDs+XJEUib5kNAgA6hmImmUzqiSee0Ic//GF9+ctf1jvf+U79+Mc/1oUXXqjrr79eH//4x3OZEwCKSs+eGYoZ9NbW9ielUi3yeoepuvp9puMAAIA+qq3N7plpaWHPDDCYIpHVkvT2qTVg4EKh7J4ZTswAKAT9np/yxhtv6OGHH9Yvf/lLuVwuXXrppfrud7+rhoaG7mvOP/98nXDCCTkNCgDFJBjMfpqLEzP4R01Nv5Yk1dczxgwAgGJSU3OqJLdisQ3q6tqkQGCi6UhAWeDEDHIlHHaKGU7MADCv3ydmTjjhBK1fv14/+MEPtGPHDt111129ShlJmjhxoi655JKchQSAYtNzYma1bDttOA0KRSaT1N692TFm9fUXG04DAAD6w+OpUlXVSZKk1tYlhtMA5cPZMeN8+A0YqFBojiRLicQuJRJNpuMAKHP9LmY2btyoZ555RhdffLG8Xu8hrwmFQnr44YePORwAFKtAYJJcrgplMjF1dW0yHQcForV1iVKpVnm9w1RTwxgzAACKTc+eGcaZAYMhkWhWKrVPkqVgcLrpOChyHk9YgcAUSZyaAWBev4uZpqYmvfrqqwfd/uqrr+rvf/97vwM88MADmjBhgioqKnTiiSfqtddeO+y1yWRSt956qyZPnqyKigrNmzdPzzzzTK9r0um0brzxRk2cOFGBQECTJ0/WN7/5Tdm23X2Nbdu66aabNHLkSAUCAS1YsEDr16/vd3YAOBzLcisYzJ4mZM8MHM3NT0iS6usvkmW5DacBAAD91VPMLJFtZwynAUqfM8asomKC3O6g4TQoBeHwfEkUMwDM63cx84UvfEHbtm076PYdO3boC1/4Qr/u6/HHH9c111yjm2++WW+88YbmzZunM888U01Nhz5OeMMNN+iHP/yh7r//fq1atUqf+9zndP755+vNN9/svuaOO+7QD37wA33/+9/X6tWrdccdd+jOO+/U/fff333NnXfeqe9973t68MEH9eqrryoUCunMM89ULBbrV34AOJJgMDvOjD0zkKRMJqG9e/9LkjRs2EcMpwEAAANRVXWi3O6wUql9LI8GBkEkslqSFAzOMJwEpaJnz8xSs0EAlL1+FzOrVq3SO97xjoNuP+6447Rq1ap+3dc999yjz3zmM1q0aJFmzpypBx98UMFgUA899NAhr3/00Ud1/fXXa+HChZo0aZKuuOIKLVy4UHfffXf3NS+99JLOPfdcnX322ZowYYIuuugiffCDH+w+iWPbtu69917dcMMNOvfcczV37lz97Gc/086dO/X000/3Kz8AHImzZ4ZiBpLU2vqsUqk2+XwjVF39XtNxAADAALhcXtXUnCYp+287gPxyTsyEQuyXQW6EQtliJhLhxAwAs/pdzPj9fu3Zs+eg23ft2iWPx9Pn+0kkEnr99de1YMGCnjAulxYsWKCXX375kN8Tj8dVUVHR67ZAIKAXX3yx+z+/+93v1pIlS7Ru3TpJ0ltvvaUXX3xRZ511liRp06ZN2r17d6/Hra6u1oknnnjYx3Ueu6Ojo9cXABwJxQwO1NTEGDMAAEpBbe0ZktgzAwyGSCRbzASDFDPIDWeUWTS6Ruk0k3MAmNPvYuaDH/ygrrvuOrW3t3ff1tbWpuuvv15nnHFGn+9n7969SqfTGj58eK/bhw8frt27dx/ye84880zdc889Wr9+vTKZjBYvXqynnnpKu3bt6r7ma1/7mi655BI1NDTI6/XquOOO01VXXaWPf/zjktR93/15XEm67bbbVF1d3f01duzYPv+sAMqTU8xEo2tk22nDaWBSdozZ05Kk+nrGmAEAUMycPTNtbS8one4ynAYobdEoo8yQW37/aHk8Q2Tbqe4TWQBgQr+Lmbvuukvbtm3T+PHj9f73v1/vf//7NXHiRO3evbvXSLF8uO+++zR16lQ1NDTI5/Ppyiuv1KJFi+Ry9fwYTzzxhH7xi1/oscce0xtvvKFHHnlEd911lx555JFjemynjHK+DrVnBwAOVFExUS5XQLYdV1fXRtNxYFBr62Kl0+3y+Uaquvo9puMAAIBjEAzOkM83SrYdV0fHS6bjACUrlWpXIrFTkhQKUcwgNyzL6j4109nJODMA5vS7mBk9erSWLVumO++8UzNnztTxxx+v++67T8uXL+/XKZK6ujq53e6DxqLt2bNHI0aMOOT31NfX6+mnn1YkEtGWLVu0Zs0ahcNhTZo0qfuar371q92nZubMmaNPfOITuvrqq3XbbbdJUvd99+dxpewIt6qqql5fAHAkluXq/mQX48zKW+8xZv3+pxcAABQQy7K6T820tDDODMiXSCR7WsbnGyWPp9pwGpSScDi7Z6azc6nZIADK2oDeHQqFQrr88sv1wAMP6K677tKll14qr9fbr/vw+Xw6/vjjtWTJku7bMpmMlixZopNPPvmI31tRUaHRo0crlUrpySef1Lnnntv9Z9FotNcJGklyu93KZDKSpIkTJ2rEiBG9Hrejo0OvvvrqUR8XAPqrZ5wZxUy5ymTijDEDAKDEOMVMa+uzhpMApcsZMxUKsV8GudVTzHBiBoA5noF+46pVq7R161YlEolet3/4wx/u831cc801+uQnP6l3vvOdete73qV7771XkUhEixYtkiRdeumlGj16dPdpl1dffVU7duzQ/PnztWPHDt1yyy3KZDK69tpru+/znHPO0be//W2NGzdOs2bN0ptvvql77rlHl112maTsp5uuuuoqfetb39LUqVM1ceJE3XjjjRo1apTOO++8gf51AMAhOcUMJ2bKV0vL/ymd7pDPN0rV1e82HQcAAOSAU8x0dr6hZHKfvN6hhhMBpScSyRYzwSDFDHLLGWUWibwl27ZlWZbZQADKUr+LmY0bN+r888/X8uXLZVmWbNuWpO5fYul03xdcf/SjH1Vzc7Nuuukm7d69W/Pnz9czzzyj4cOHS5K2bt3a6/RLLBbTDTfcoI0bNyocDmvhwoV69NFHVVNT033N/fffrxtvvFGf//zn1dTUpFGjRumzn/2sbrrppu5rrr32WkUiEV1++eVqa2vTe9/7Xj3zzDOqqKjo718HABxRMEgxU+6am50xZhczxgwAcJBt27bJsiyNGTNGkvTaa6/pscce08yZM3X55ZcbTofD8ftHKhicpWh0pVpb/6Rhwy42HQkoOdFodpSZMx4ayJVgcIYsy6tUqk3x+FZVVIw3HQlAGbJsp1npo3POOUdut1s//vGPNXHiRL322mvat2+fvvzlL+uuu+7SKaeckq+sBaWjo0PV1dVqb29n3wyAw+rq2qRXX50ky/LplFMicrkGfFARRSidjumll4Yrne7Qccf9lRMzAIoaz3/z45RTTtHll1+uT3ziE9q9e7emT5+uWbNmaf369friF7/Y6wNmxabU/zfT2Hi1tm+/VyNHfkbTp/+n6ThAyXnllYmKxTZr/vznVVPzPtNxUGL+9rf5ikTe0uzZ/626ur5P/wGAo+nrc+B+f3T35Zdf1q233qq6ujq5XC65XC69973v1W233aZ//dd/PabQAFBqKirGy+UKyrYTisU2mI6DQdbamh1j5vePUVXVSabjAAAK0IoVK/Sud71LkvTEE09o9uzZeumll/SLX/xCP/3pT82GwxGxZwbIn3Q6olhssyRGmSE/evbMLDUbBEDZ6ncxk06nVVlZKUmqq6vTzp07JUnjx4/X2rVrc5sOAIqcZbm6j94zzqz8NDU5Y8wuYowZAOCQksmk/H6/JOnZZ5/t3tnZ0NCgXbt2mYyGo6iuPlWW5VEstkldXRtNxwFKSjSafX/J662Tz1dnOA1KUU8x85bhJADKVb/fJZo9e7beeiv7S+vEE0/UnXfeqb/+9a+69dZbNWnSpJwHBIBiFwqxZ6YcpdNd2rfvvyVJ9fUfMZwGAFCoZs2apQcffFAvvPCCFi9erA996EOSpJ07d2roUBbKFzKPJ6yqqpMlSa2tiw2nAUpLJLJKEqdlkD/h8HxJFDMAzOl3MXPDDTcok8lIkm699VZt2rRJp5xyiv7whz/oe9/7Xs4DAkCxo5gpTy0t/6t0ulN+/1hVVZ1oOg4AoEDdcccd+uEPf6jTTjtNH/vYxzRvXvYTvL/97W+7R5yhcDHODMiPaHS1JHVPHwByzTkxE4ttUCrVYTgNgHLU7y3UZ555Zvf/PWXKFK1Zs0YtLS2qra2VZVk5DQcApcApZqJRiply0tzsjDG7mDFmAIDDOu2007R37151dHSotra2+/bLL79cwWDQYDL0RW3tGdq8+Wa1ti6RbadlWW7TkYCSEI1mT8yEQpyYQX54vUPl949RPL5dkchyVVe/x3QkAGWmX+8UJZNJeTwerVixotftQ4YMoZQBgMMIBp1iZq0ymaThNBgM6XSX9u79rSRp2DDGmAEADq+rq0vxeLy7lNmyZYvuvfderV27VsOGDTOcDkdTWXmC3O4qpVKt2r//TdNxgJLBKDMMhlDI2TOz1GwQAGWpX8WM1+vVuHHjlE6n85UHAEpORcU4uVwh2XZSXV2NpuNgELS0PKNMJiK/f5wqKxlDAwA4vHPPPVc/+9nPJEltbW068cQTdffdd+u8887TD37wA8PpcDQul0c1Ne+XxJ4ZIFcymbi6ujZIkkIhRpkhf5xxZuyZAWBCv2erfP3rX9f111+vlpaWfOQBgJJjWa7uI/jOkXyUNmeM2bBhH+FEKQDgiN544w2dcsopkqTf/OY3Gj58uLZs2aKf/exn7PAsEuyZAXIrGl0vKS23u0o+3yjTcVDCwuH5kihmAJjR7x0z3//+99XY2KhRo0Zp/PjxCoVCvf78jTfeyFk4ACgVodAs7d//N0UiK1Vff6HpOMijdDqqvXv/R1J2vwwAAEcSjUZVWVkpSfq///s/XXDBBXK5XDrppJO0ZcsWw+nQF0OGnCFJam9/Uel0VG43u4GAY3Hgfhk+5IR8ck7MRCLL2RMGYND1u5g577zz8hADAEqbs2cmEllpOAnyraXlj2+PMRuvysoTTMcBABS4KVOm6Omnn9b555+v//3f/9XVV18tSWpqalJVVZXhdOiLQGBa9wLp9vYXNWTIB01HAopaNLpakhQMMsYM+RUITJbLFVQmE1U0ul6hUIPpSADKSL+LmZtvvjkfOQCgpIVCFDPloqmJMWYAgL676aab9M///M+6+uqrdfrpp+vkk0+WlD09c9xxxxlOh76wLEu1tWdo9+6H1dr6LMUMcIwikeyJmWBwpuEkKHWW5VY4PFcdHa+os3MpxQyAQdXvHTMAgP5zipmurnXKZJKG0yBf0umI9u37nSSpvv4jhtMAAIrBRRddpK1bt+rvf/+7/vd//7f79g984AP67ne/azAZ+qNnz8xiw0mA4uecmAmFODGD/AuFnHFm7JkBMLj6fWLG5XId8RPA6XT6mAIBQCny+8fK7Q4rne5UV9d6hUJ8+qsU7dv3B2UyUVVUTFRl5fGm4wAAisSIESM0YsQIbd++XZI0ZswYvetd7zKcCv1RW/sBSVJn51IlEs3y+eoNJwKKUyaTUjS6VhInZjA4wuH5kqTOTooZAIOr38XMf/3Xf/X6z8lkUm+++aYeeeQRfeMb38hZMAAoJZZlKRicqf37X1MkspJipkQ1N/9aUva0DGPMAAB9kclk9K1vfUt33323Ojs7JUmVlZX68pe/rK9//etyuRhyUAx8vuEKheYqElmm1tYlGj78EtORgKIUi22UbSfkcgVUUTHedByUgXA4e2Kms3Op2SAAyk6/i5lzzz33oNsuuugizZo1S48//rg+/elP5yQYAJSaUGhWdzEjXWw6DnLswDFmw4bx3y8AoG++/vWv6yc/+Yluv/12vec975Ekvfjii7rlllsUi8X07W9/23BC9FVt7YK3i5lnKWaAAXLGmAWDDbIsimnkXyg0R5KlRGKXEokm+XzDTEcCUCZy9q/cSSedpCVLluTq7gCg5Dh7ZqLRlYaTIB/27fu9MpkuVVRMUjj8DtNxAABF4pFHHtGPf/xjXXHFFZo7d67mzp2rz3/+8/rRj36kn/70p6bjoR9qa8+QlN0zY9u24TRAcYpEVklijBkGj8cTViAwRRLjzAAMrpwUM11dXfre976n0aNH5+LuAKAkBYPZYiZ7YgalpqnpCUnSsGGMMQMA9F1LS4saGhoOur2hoUEtLS0GEmGgampOkWX5FI9vVVdXo+k4QFGKRrPFDKOfMZh6xplRzAAYPP0uZmprazVkyJDur9raWlVWVuqhhx7Sd77znXxkBICS4JyY6epar0wmYTgNcimV6lRLy+8lZffLAADQV/PmzdP3v//9g27//ve/r7lz5xpIhIFyu0Oqrn63JKm19VnDaYDiFIk4o8xmGE6CchIOz5ckRSIUMwAGT793zHz3u9/t9Ulgl8ul+vp6nXjiiaqtrc1pOAAoJX7/GLndVUqnOxSNrlM4PNt0JOTIvn2/UyYTUyAwpftJPQAAfXHnnXfq7LPP1rPPPquTTz5ZkvTyyy9r27Zt+sMf/mA4HfqrtnaB2tqeU2vrYo0efYXpOEBRse1M944ZTsxgMIVCzomZpWaDACgr/S5mPvWpT+UhBgCUPsuyFArNVEfHK4pGV1LMlADbTqut7QVt3363JKmu7iLGmAEA+uXUU0/VunXr9MADD2jNmjWSpAsuuECXX365vvWtb+mUU04xnBD9UVu7QJs23aDW1j/JttOyLLfpSEDRiMe3KZOJyrK8qqiYbDoOyogzyiwaXaN0Oia3u8JwIgDloN/FzMMPP6xwOKyLL7641+2//vWvFY1G9clPfjJn4QCg1ASDs9TR8Ur3UksUr+bmp9TY+CXF49u7b9u9+2FVVZ2g+voLDCYDABSbUaNG6dvf/nav29566y395Cc/0X/+538aSoWBqKx8p9zuaqXT7dq//++qqjrRdCSgaDivkQKBaXK5+v12FTBgfv8YeTxDlEq1KBpdpcrKd5iOBKAM9HvHzG233aa6urqDbh82bJj+/d//PSehAKBUOXtmIpGVhpPgWDQ3P6WVKy/qVcpIUjLZpJUrL1Jz81OGkgEAAJMsy63a2tMlsWcG6K9oNFvMMMYMg82yrO5TM52d7JkBMDj6Xcxs3bpVEydOPOj28ePHa+vWrTkJBQClyilmolGKmWJl22k1Nn5Jkn2oP5UkNTZeJdtOD2ouAABQGGprz5AktbQsNpwEKC6RSHa/TDA4w3ASlCNnVyjFDIDB0u9iZtiwYVq2bNlBt7/11lsaOnRoTkIBQKnqKWbWK5OJG06DgWhre+GgkzK92YrHt6mt7YVBywQAAApHbe0CSVJHx0tKpyOG0wDFgxMzMKnnxMxSs0EAlI1+D+382Mc+pn/9139VZWWl3ve+90mSnn/+eX3pS1/SJZdckvOAAFBKfL5R3XPHo9F1CofnmI6EfkokduX0OgBAebrggiPvI2traxucIMi5QGCK/P7xise3qK3tBQ0d+iHTkYCCZ9u2olFOzMCcUChbzEQib8m2bVmWZTgRgFLX72Lmm9/8pjZv3qwPfOAD8niy357JZHTppZeyYwYAjsKyLIVCs9TR8ZIikZUUM0XI5xuZ0+sAAOWpurr6qH9+6aWXDlIa5JJlWaqtXaDdu3+i1tbFFDNAHyQSu5VKtUlyKRCYZjoOylAoNFOW5VUq1aZ4fJsqKsaZjgSgxPW7mPH5fHr88cf1rW99S0uXLlUgENCcOXM0fvz4fOQDgJLjFDPsmSlONTWnyO8fo3h8hw69Z8aS3z9GNTWnDHY0AEARefjhh01HQB71FDPPmo4CFAVnjFkgMFlud4XhNChHLpdPweAMRSLL1Nm5lGIGQN71u5hxTJ06VVOnTs1lFgAoC8FgdmZyJEIxU4wsy60pU+7TypUXHepPJUlTptwry3IPbjAAAFAwams/IEmKRJYpkdgjn2+44URAYYtEGGMG88Lh+W8XM2+pru7DpuMAKHGu/n7DhRdeqDvuuOOg2++8805dfPHFOQkFAKUsFJoliWKmmNXXX6CRIz9z0O1+/xjNmvUb1dcfeW8AAAAobT5fvcLh+ZKk1tYlZsMARcA5MRMKzTScBOUsHM7umensXGo2CICy0O9i5i9/+YsWLlx40O1nnXWW/vKXv+QkFACUMqeY6epqVDodM5wGA5VM7pUkDR9+mWbMeEzz5v1ZJ520iVIGAABIkmprz5AktbYuNpwEKHyRSLaYcaYLACb0FDNvGU4CoBz0u5jp7OyUz+c76Hav16uOjo6chAKAUubzjZTHUyMpo66utabjYABsO622tj9JkkaN+oyGD/+YamtPY3wZAADoVlu7QJLU2vqsbPtQe+kAOKJRRpnBvFAoW8zEYhuUSu03nAZAqet3MTNnzhw9/vjjB93+q1/9SjNn8skGADgay7IUDDLOrJjt3/+6Uqk2ud3Vqqx8p+k4AACgAFVXnyLL8ise366urnWm4wAFK5ncp2SySZIUDDYYToNy5vPVyecbLSm7IwwA8snT32+48cYbdcEFF2jDhg06/fTTJUlLlizRY489pt/85jc5DwgApSgUmqWOjr9SzBQpZyRJbe3pcrn6/U8pAAAoA253QNXV71Fb25/U0rJYweB005GAghSJZE/L+P3j5PGEDadBuQuH56ulZYc6O99SdfV7TMcBUML6fWLmnHPO0dNPP63GxkZ9/vOf15e//GXt2LFDf/rTnzRlypR8ZASAkuPsmaGYKU6trc9K6hlRAgAAcCg9e2aeNZwEKFzRaHa/TCjEFBaY17NnZqnZIABKXr+LGUk6++yz9de//lWRSEQbN27URz7yEX3lK1/RvHnzcp0PAEqSU8w4L0JQPNLpiNrb/yqp580WAACAQ3E+xNHW9mdlMinDaYDCFIlkXxMFgxQzMK+nmHnLcBIApW5AxYwk/eUvf9EnP/lJjRo1SnfffbdOP/10vfLKK7nMBgAly9kx09W1Qel0zHAa9Edb219k20n5/eMVCHBSFAAAHF5l5XHyeGqVTndo//6/mY4DFKRoNDvKLBicYTgJkB1lJkmRyHLZdtpsGAAlrV/FzO7du3X77bdr6tSpuvjii1VVVaV4PK6nn35at99+u0444YR85QSAkuLzDZfHM0RSRtHoGtNx0A8HjjGzLMtwGgAAUMgsy63a2g9I6tlRB6A3RpmhkAQCk+VyBZXJdCkaXW86DoAS1udi5pxzztH06dO1bNky3Xvvvdq5c6fuv//+fGYDgJJlWdYB48zYM1NMnDdVhgxhjBkAADg6Z5wZe2aAg6VSHYrHt0vixAwKg2W5FQ7PlSRFIowzA5A/fS5m/vjHP+rTn/60vvGNb+jss8+W2+3OZy4AKHlOMROJUMwUi3h8tyKR5ZKkmpoPGE4DAACKgbOTrqPjZaVS+w2nAQqLMz3A5xshr7fWcBogKxRy9swsNRsEQEnrczHz4osvav/+/Tr++ON14okn6vvf/7727t2bz2wAUNKcPTMUM8WjrW2JJCkcPk4+X53hNAAAoBgEApNUUTFRtp1Se/tfTMcBCkokkh1jFgwyxgyFIxx2ihlOzADInz4XMyeddJJ+9KMfadeuXfrsZz+rX/3qVxo1apQymYwWL16s/fv55A8A9AcnZopPS0t2jJnzyVcAAIC+cJ47MM4M6C0aXS2JMWYoLOHwfEkUMwDyq8/FjCMUCumyyy7Tiy++qOXLl+vLX/6ybr/9dg0bNkwf/vCH85ERAEqSs9wyFtuodDpqOA2Oxrbt7v0yFDMAAKA/nD0zzoc8AGRFo9kTM85rI6AQhEJzJFlKJHYqkWg2HQdAiep3MXOg6dOn684779T27dv1y1/+MleZAKAseL3D5PEMlWR3z1ZG4YpG1yiR2CnL8qu6+j2m4wAAgCJSW3u6JEvR6ErF47tMxwEKBqPMUIg8nrACgSmSODUDIH+OqZhxuN1unXfeefrtb3+bi7sDgLJgWRbjzIqIc1qmpuYUud0Bw2kAADjYAw88oAkTJqiiokInnniiXnvttcNeu3LlSl144YWaMGGCLMvSvffee8z3icPzeocqHH6HJMaZAY50ukux2CZJUijEKDMUlp49M0vNBgFQsnJSzAAABoZipngwxgwAUMgef/xxXXPNNbr55pv1xhtvaN68eTrzzDPV1NR0yOuj0agmTZqk22+/XSNGjMjJfeLInHFmFDNAVjS6VpItj2eIvN5hpuMAvYRC2WImEuHEDID8oJgBAIOcYiYapZgpZJlMUm1tz0mimAEAFKZ77rlHn/nMZ7Ro0SLNnDlTDz74oILBoB566KFDXn/CCSfoO9/5ji655BL5/f6c3CeObMiQ7HOI1tbFsm3bcBrAvGh0tSQpGJwhy7IMpwF6C4fnS2KUGYD8oZgBAIOCQU7MFIOOjleVTnfK663rPtIOAEChSCQSev3117VgwYLu21wulxYsWKCXX3550O4zHo+ro6Oj1xd6VFW9Ry5XhRKJXd1vSAPlLBrN7pcJhdgvg8LjvO6LRlcrk4kbTgOgFFHMAIBBzomZWGyz0umI4TQ4nJ79Mh+QZfFPJwCgsOzdu1fpdFrDhw/vdfvw4cO1e/fuQbvP2267TdXV1d1fY8eOHdBjlyq3u0LV1adIYpwZIEmRSLaYCQYpZlB4/P4x8niGyLZT3f9bBYBc4t0lADDI56uX11svyVY0usZ0HByGU8w4I0gAAMDBrrvuOrW3t3d/bdu2zXSkgtOzZ2ax4SSAec7JsVBohuEkwMEsy+o+NdPZudRsGAAliWIGAAxzTs0wzqwwpVLt6uh4TVLPmykAABSSuro6ud1u7dmzp9fte/bs0YgRIwbtPv1+v6qqqnp9oTfnuURb23PKZJKG0wDmZDIJdXWtl8SJGRSunmKGPTMAco9iBgAMY89MYWtre05SWoHAVFVUjDcdBwCAg/h8Ph1//PFasmRJ922ZTEZLlizRySefXDD3iewyaY9nqNLpTnV0vGo6DmBMV1ejbDsltzssv3+M6TjAIYXD8yVJkQjFDIDco5gBAMM4MVPYWlqyo0ZqaxljBgAoXNdcc41+9KMf6ZFHHtHq1at1xRVXKBKJaNGiRZKkSy+9VNddd1339YlEQkuXLtXSpUuVSCS0Y8cOLV26VI2NjX2+T/SfZblUW/sBSeyZQXlzxpgFgzNkWZbhNMChhUI9o8xs2zacBkCp8ZgOAADlzilmolGKmULkvGnCGDMAQCH76Ec/qubmZt10003avXu35s+fr2eeeUbDhw+XJG3dulUuV8/n8nbu3Knjjjuu+z/fdddduuuuu3Tqqafqueee69N9YmBqa89Qc/MTam1drIkTbzEdBzDCWabOGDMUslBopizLq1SqTfH4NlVUjDMdCUAJoZgBAMOcYiYW26xUqlMeT9hwIjhisW3q6loryaWamvebjgMAwBFdeeWVuvLKKw/5Z07Z4pgwYUKfPv17pPvEwDgf9ujoeFWpVIc8HnbxoPw4J2ZCoRmGkwCH53L5FAzOUCSyTJ2dSylmAOQUo8wAwDCvd6i83mGSel6goDC0tmbHmFVVvUteb43ZMAAAoCQEAhMUCEyRlFZb2/Om4wBGcGIGxSIcdsaZsWcGQG5RzABAAWDPTGFijBkAAMgH57mF8yEQoJzYdlrR6BpJ2VFRQCELh+dLopgBkHsUMwBQANgzU3hsO3NAMXOG4TQAAKCUOM8tnOcaQDmJxTbLtuOyLL8qKiaYjgMcUc+JmaVmgwAoORQzAFAAgkFOzBSazs5lSiab5XKFVFV1kuk4AACghGR311mKRlcrFttuOg4wqHrGmDXIstyG0wBHFgpli5lYbINSqf2G0wAoJRQzAFAAGGVWeJxPsNbUnCqXy2c4DQAAKCVeb60qK98pSWprW2I4DTC4otFsMcMYMxQDn69OPt9oSVIkssxwGgClhGIGAAqAU8zE41v5FE6BcGa+M8YMAADkg/Mco6WFPTMoL5HIaklSMDjDcBKgb3rGmbFnBkDuFEQx88ADD2jChAmqqKjQiSeeqNdee+2w1yaTSd16662aPHmyKioqNG/ePD3zzDO9rpkwYYIsyzro6wtf+EL3NaeddtpBf/65z30ubz8jAByJ1ztEPt8IST2fIIM56XRM7e1/kSQNGUIxAwAAcq+2doGk7Cld27YNpwEGDydmUGzC4fmSKGYA5JbxYubxxx/XNddco5tvvllvvPGG5s2bpzPPPFNNTU2HvP6GG27QD3/4Q91///1atWqVPve5z+n888/Xm2++2X3N3/72N+3atav7a/Hi7CeQLr744l739ZnPfKbXdXfeeWf+flAAOIqePTMUM6Z1dLykTCYmn2+kgkFeMAIAgNyrrn63XK6Aksk9jLNF2bBtW9EoJ2ZQXHpOzCw1GwRASTFezNxzzz36zGc+o0WLFmnmzJl68MEHFQwG9dBDDx3y+kcffVTXX3+9Fi5cqEmTJumKK67QwoULdffdd3dfU19frxEjRnR//e53v9PkyZN16qmn9rqvYDDY67qqqqq8/qwAcCTsmSkcPWPMFsiyLMNpAABAKXK5/Kqufp+knuceQKmLx7crne6UZXkUCEwxHQfoE6eYiUSWy7bThtMAKBVGi5lEIqHXX39dCxYs6L7N5XJpwYIFevnllw/5PfF4XBUVFb1uCwQCevHFFw/7GD//+c912WWXHfTm2i9+8QvV1dVp9uzZuu666xSNRg+bNR6Pq6Ojo9cXAOSSU8xEoxQzpjmz3tkvAwAA8skZmdra+qzhJMDgcMaYBQJT5XL5DKcB+iYQmCKXK6hMpkvR6HrTcQCUCKPFzN69e5VOpzV8+PBetw8fPly7d+8+5PeceeaZuueee7R+/XplMhktXrxYTz31lHbt2nXI659++mm1tbXpU5/6VK/b//mf/1k///nP9ec//1nXXXedHn30Uf2///f/Dpv1tttuU3V1dffX2LFj+/fDAsBRcGKmMCST+9TZ+YakntnvAAAA+eA812hre16ZTMJwGiD/IhHGmKH4WJZbodAcSVIkwp4ZALlhfJRZf913332aOnWqGhoa5PP5dOWVV2rRokVyuQ79o/zkJz/RWWedpVGjRvW6/fLLL9eZZ56pOXPm6OMf/7h+9rOf6b/+67+0YcOGQ97Pddddp/b29u6vbdu25fxnA1DenB0z8fg2pVKcyjOltfVPkmwFg7Pk9480HQcAAJSwUGiOvN56ZTIRdXS8YjoOkHfOiZlQiD2OKC7h8HxJUmcnxQyA3DBazNTV1cntdmvPnj29bt+zZ49GjBhxyO+pr6/X008/rUgkoi1btmjNmjUKh8OaNGnSQddu2bJFzz77rP7lX/7lqFlOPPFESVJjY+Mh/9zv96uqqqrXFwDkktdbI58vWyJHIqsMpylfzox3Z7QIAABAvliWq/vUDHtmUA6c1znBIMUMiouzZ6azc6nZIABKhtFixufz6fjjj9eSJUu6b8tkMlqyZIlOPvnkI35vRUWFRo8erVQqpSeffFLnnnvuQdc8/PDDGjZsmM4+++yjZlm6dKkkaeRIPh0NwBz2zJhl23b3myLslwEAAIOhp5hhzwxKm23b3SdmGGWGYtNTzHBiBkBueEwHuOaaa/TJT35S73znO/Wud71L9957ryKRiBYtWiRJuvTSSzV69GjddtttkqRXX31VO3bs0Pz587Vjxw7dcsstymQyuvbaa3vdbyaT0cMPP6xPfvKT8nh6/5gbNmzQY489poULF2ro0KFatmyZrr76ar3vfe/T3LlzB+cHB4BDCAZnqrV1MXtmDInFNioW2yzL8qq6+n2m4wAAgDLgFDMdHa8plWqXx1NtOBGQH8lkk1KpVkmWgsHppuMA/ZLdMWMpkdipRKJZPl+96UgAipzxYuajH/2ompubddNNN2n37t2aP3++nnnmGQ0fPlyStHXr1l77Y2KxmG644QZt3LhR4XBYCxcu1KOPPqqamppe9/vss89q69atuuyyyw56TJ/Pp2effba7BBo7dqwuvPBC3XDDDXn9WQHgaJwTMxQzZrS0ZE/LVFWdLI8nbDgNAAAoBxUV4xQITFNX1zq1tv5Z9fXnmY4E5EUkslqSVFExUW53wHAaoH88nkoFApPV1dWozs63NGTIAtORABQ548WMJF155ZW68sorD/lnzz33XK//fOqpp2rVqqPvXvjgBz8o27YP+Wdjx47V888/3++cAJBvFDNmMcYMAACYUFt7xtvFzLMUMyhZzhizUIj9MihO4fB8dXU1KhKhmAFw7IzumAEA9OYswUwkdiiZbDMbpszYdlptbX+S1DNSBAAAYDD07JlZbDgJkD+RiLNfhmIGxSkUcvbMLDUbBEBJoJgBgALi9dbI5xstqecTZRgc+/e/rlSqTW53tSor32k6DgAAKCM1NadJcqmra51isa2m4wB5EY1mR5kFgzMMJwEGJhx2ipm3DCcBUAooZgCgwDDOzIyeMWany+UqiEmfAACgTHi9NaqqepckqbX1WcNpgPxglBmKXTg8X1K2ZMxk4mbDACh6FDMAUGAoZsxw3gRhjBkAADChZ5wZxQxKTzLZqkRityQpGGwwnAYYGL9/jDyeWtl2qns0HwAMFMUMABQYp5hhlNngSacjam//q6Ts8l0AAIDB5jwHaW19VradMZwGyC1njFn2je0qw2mAgbEsq/vUDOPMABwrihkAKDDBICdmBltb219k20n5/eMVCEwxHQcAAJShqqqT5HKFlEw2KxJZbjoOkFPO6YJgkDFmKG49e2aWmg0CoOhRzABAgXFmLicSO5VMtpkNUyYOHGNmWZbhNAAAoBy5XD7V1JwqiXFmKD3OiZlgcIbhJMCxCYWyxUwkwokZAMeGYgYACozHUyW/f6wkKRrl1MxgaG1dLEkaMoQxZgAAwBxnz0xLy2LDSYDccsY0Ox9CA4rVgaPMbNs2GwZAUaOYAYAC5OyZYZxZ/sXju7vHhdTUfMBwGgAAUM6cYqa9/S/KZOKG0wC5wygzlIpQaIYsy6NUqlXx+DbTcQAUMYoZAChA7JkZPG1tSyRJ4fBx8vnqDKcBAADlLBSaLa93uDKZLrW3v2Q6DpATqVSn4vGtkrJvagPFzOXydxeMnZ2MMwMwcBQzAFCAnCP+FDP554wKqa1ljBkAADDLsqzuUzPsmUGpiEbXSJK83mHyeocaTgMcu3A4u2ems3Op2SAAihrFDAAUIGeUGTtm8su27e79MhQzAACgEDg775znKECxi0ZXS5KCQU7LoDT0FDOcmAEwcBQzAFCAnKPRicRuJZMthtOUrmh0jRKJnXK5KlRd/V7TcQAAALpPzOzf/3clk62G0wDHLhrN7pdxpgIAxS4cni9JikQoZgAMHMUMABQgj6dSfv84SYwzyyfnk6jV1e+V211hOA0AAIDk949++2SBrba2P5uOAxyzSCRbzDgfPgOKXSiUPTHT1dWoVGq/4TQAihXFDAAUKGecGcVM/jDGDAAAFKKePTOMM0Pxc0aZhUKMMkNp8Pnq5PONliRFIssNpwFQrChmAKBAsWcmvzKZpNranpNEMQMAAAqL89yktfVZw0mAY5NOx9TVtUESJ2ZQWnr2zCw1GwRA0aKYAYACFQxyYiafOjpeVTrdKa+3rvtJNQAAQCGoqTlVkltdXY3q6tpsOg4wYF1d6yRl5PHUyOcbYToOkDM9xQx7ZgAMDMUMABSonlFmqwwnKU3OaJCamg/IsvjnEAAAFA6Pp0pVVSdK4tQMipszxiwYnCHLsgynAXInHJ4viWIGwMDxThQAFKjs0lcpmdyjZHKf4TSlxylmhgxhjBkAACg8PePM2DOD4uV8yIwxZig1zomZSGSZbDttOA2AYkQxAwAFyuMJq6JigiTGmeVaKtWujo7XJPUs1wUAACgkznOU1tYlsu2M4TTAwDgnZkKhGYaTALkVCEyRyxVUJtOlrq5G03EAFCGKGQAoYOyZyY+2tuckpRUITFVFxXjTcQAAAA5SVXWi3O6wUql9jMoZRLadVmvrc9qz55dqbX2OT8IfI07MoFRZlluh0BxJUmfnUrNhABQlihkAKGA9e2YoZnKppSU7EsQZEQIAAFBoXC6vampOk8Q4s8HS3PyUXnllgt566/1avfqf9dZb79crr0xQc/NTpqMVpUwmpa6udZKkUIhiBqXHGWdGeQ5gIChmAKCAOcVMNEoxk0vOEl3GmAEAgELWs2fmWcNJSl9z81NaufIixePbe90ej+/QypUXUc4MQCy2QbadlMsVlN8/1nQcIOfC4fmSKGYADAzFDAAUME7M5F4stk1dXWsluVRT837TcQAAAA7L+RBJe/sLSqdjhtOULttOq7HxS5LsQ/2pJKmx8SrGmvVTzxizGbIs3n5C6ek5MbPUbBAARYl/GQGggAWDDZKkZLJZiUSz4TSlwRkFUlX1Lnm9NWbDAAAAHEEwOEM+3yhlMjF1dPzVdJyS1db2wkEnZXqzFY9vU1vbC4OWqRREo9liJhSaYTgJkB/ZHTOWEomdvF4H0G8UMwBQwNzukCoqJkri1EyuMMYMAAAUC8uyup+zODvykHuJxK6cXoesSGS1JCkYZL8MSpPHU6lAYLIkxpkB6D+KGQAocOyZyR3bzhxQzJxhOA0AAMDROcUMe2byx+cbmdPrkNVzYoZiBqUrFMqOM4tEKGZyzbbTam19Tnv2/FKtrc8xThIlx2M6AADgyILBWdq373ecmMmBzs5lSiab5XKFVFV1kuk4AAAAR+UUM52dbyiZ3Cevd6jhRKUnHJ4ry/LKtpOHvcbvH6uamlMGMVVxs+2MotE1krIj+YBSFQ7P1969T3JiJseam59SY+OXeo2Z9PvHaMqU+1Rff4HBZEDucGIGAAqcc2KGYubYOZ80rak5VS6Xz3AaAACAo/P7RyoYnCXJVmvrn0zHKTmZTFwrV154QCljHfK6QGDaYf8MB4vFtiiT6ZJl+VRRMcl0HCBvwuHsiZnOzqVmg5SQ5uantHLlRQft/orHd2jlyovU3PyUoWRAblHMAECBO7CYsW3bcJri1tqanc3OGDMAAFBMhgzJPndhnFlu2XZGq1dfqra25+R2V2rSpO/I7x/d6xqPZ4gkS21tS7Ru3Wdl2xkzYYuMM8YsGJwul4thLShd4fB8SVI0ulqZTNxsmBJg22k1Nn5J0qHe+8je1th4FWPNUBIoZgCgwAWDDZIspVL7lEw2mY5TtNLpmNrb/yKp580NAACAYtCzZ2ax4SSlw7ZtNTZeo+bmJ2RZXs2a9ZTGjfuKTjpps+bN+7NmzHhM8+b9We95T5NmzHhUkku7dv1Y69Z9jnKmDyKR1ZIYY4bS5/ePkcdTK9tOKRJZZTpO0Wtre+GgkzK92YrHt6mt7YVBywTkC8UMABQ4tzvYffyfJ3oD19HxkjKZmHy+kQoGWUAKAACKR3X1+2RZHsVim9TVtdF0nJKwbdtd2rHjPklSQ8MjGjIkW35Zllu1tadp+PCPqbb2NFmWW8OHf1wNDY8oW878SOvWXUE5cxTOiZlQiOfdKG2WZR0wzow9M8cqkdiV0+uAQkYxAwBFgD0zx65njNkCWRbzwQEAQPHweCpVVXWyJE7N5MLu3Y9q48ZrJUmTJ9+t4cM/dtTvGTHi/71dzljates/tW7d5ylnjiAa5cQMyoczziwSoZg5Vj7fyJxeBxQyihkAKAJOMRONUswMVEsL+2UAAEDx6hlnxp6ZY9HS8n9au/YySdKYMV/W2LHX9Pl7e5czP9T69V+gnDkE27a7T/pzUh3lIBRyTswsNRukJBz9Q5R+/1jV1JwyCFmA/KKYAYAiwImZY5NM7lNn5xuSet7UAAAAKCbOh0taW//E0uMB6uj4u1asuEC2ndKwYf+syZPv7Pd9jBjxie5yZufOBylnDiGR2Kl0ukOSW8HgVNNxgLxzTsx0dr4l2z7U0nr0RTS6TitXXnDALYcuaUaP/qIsyz04oYA8opgBgCIQDPYUMzzR67/W1j9JshUMzpLfz5FnAABQfCorT5DbXaVUqkX7979pOk7R6eraoOXLz1YmE1Ft7QI1NDwsyxrYWyLZcuan6ilnruQ5+gEikewYs0Bgslwuv+E0QP6FQjNkWR6lUq1HWVyPw0kkmrVs2UKlUi2qrHyXZsx4TH7/6F7XWFb298n27fcoFttqIiaQUxQzAFAEgsHpklxKpVqUSOwxHafoOLPYhwxhjBkAAChOLpdHNTXvl8Q4s/5KJJr01ltnKplsUjg8X7NmPSmXy3dM9zlixKVqaHhY2XLmB2+fnKGckaRoNDvGLBRijBnKg8vl796nxDiz/kunu7RixbmKxTaoomKi5sz5rYYP/5hOOmmz5s37s2bMeEzz5v1Z7373ToVCs5VI7Nby5WcrlWo3HR04JhQzAFAE3O6AAoFJktgzMxDOmxfslwEAAMWsZ8/MYsNJikcq1anly88+4A2/P8rjqcrJfY8Y8UlNn/6QesoZTs5IYr8MytKB48zQd7ad0Zo1l6qj42V5PDWaM+cP8vmGS5Isy63a2tM0fPjHVFt7mrzeIZoz5/fy+UYqElmhlSsvUiaTNJofOBYUMwBQJA4cZ4a+6+raoFhskyzLq+rq95mOAwAAMGDO6d/29heVTkcNpyl8mUxSK1depP37/y6vt05z5z4jv39ETh9j5MhPafr0nyhbzvyH1q//YtmXM9FodpSZc4IAKAfh8DxJnJjpr40bv6bm5t/IsryaPftphUINR7y+omKc5sz5nVyukFpbn9W6dVeU/e9cFC+KGQAoEqEQxcxAtLRkP1FaVXWyPJ6w4TQAAAADFwhMk98/RradUHv7i6bjFDTbtrV27afV2vq/crmCmjPndwoGp+XlsUaOXHRAOfNA2ZczjDJDOXJOzEQinJjpqx07fqBt274jSWpoeFg1Naf26fsqK9+hmTN/Jcml3bt/oq1bb8tjSiB/KGYAoEhQzAyMM+qDMWYAAKDYWZZ1wDgz9swcycaN12nPnkcluTVr1q9VVXViXh8vW878WE4509j4r2VZziQSzUom90py9mQC5SEUyp6Y6eraoFRqv+E0hW/fvj9o/forJUkTJnxTw4d/vF/fX1f3T5o69XuSpE2bvq49e36Z84xAvlHMAECRcIqZaHRlWb7IGwjbTqut7U+SemayAwAAFDPnwybsmTm87du/p23b7pAkTZ/+Yw0dunBQHnfkyMveLmekHTu+r8bGL5Xd83ZnjFlFxQS53SHDaYDB4/PVyecbJclWJLLcdJyCtn//m1q58iOSMhoxYpHGj//6gO5n9OgvaMyYayRJa9Z8Sm1tL+QwJZB/FDMAUCQCgemSXEql2pRI7DIdpyjs3/+6Uqk2ud3Vqqx8p+k4AAAAx6y29gOSsnsMEolmw2kKT1PTE2psvEqSNHHitzVy5KcG9fF7lzP3q7HxqrIqZyKR7BizYJAxZig/zjizzk7GmR1OLLZNy5efrUwmotraBZo27YeyLGvA9zd58ndUV3e+bDuhFSvOUzS6LodpgfyimAGAIuF2VygQmCKJcWZ91TPG7HS5XB7DaQAAAI6dzzdcodBcSeo+GYys1tbntHr1JyTZGjXq8xo37jojOUaO/LSmTfuRJGnHju+psfHqsilnnBMzweAMw0mAwRcOZ8eZdXYuNRukQKVS7Vq+fKESiV0KBmdp1qzfyOXyHtN9WpZLM2b8XJWV71Iq1aJlyxbyoQUUDYoZACgiPePMVhlOUhyc2euMMQMAAKXEeW7T0sI4M0dn5zKtWHGubDuhuroLNHXq947pU9jHatSofzmgnLlPGzZcUxbljPM6JRTixAzKDydmDi+TSWrlyosViayQzzdCc+f+QR5PdU7u2+0Oas6c36qiYoJisQ1aseJcpdNdOblvIJ8oZgCgiDjFDCdmji6djqi9/a+SemaxAwAAlIID98yUw5v9RxOLbdGyZWcpne5QdfUpmjHjF7Ist+lYb5cz/ylJ2r793rIoZ3pGmXFiBuXHOTETiSyXbacNpykctm1r3bor1Nq6WC5XSHPm/F4VFeNy+hg+33DNmfMHeTw16uh4WWvWfFK2ncnpYwC5RjEDAEUkGKSY6au2tr/ItpPy+8d3j4ADAAAoBTU1p8iyvIrHt6qrq9F0HKOSyX1atuxDSiR2Khicpdmz/1tud4XpWN1GjfqMpk37oSSnnPlyyZYzqVS7EomdkihmUJ4CgSlyuQLKZKJl/7v5QFu33qbdu38iyaWZM3+lysp35OVxQqEZmjXrv2RZXjU3/1obN16fl8cBcoViBgCKyIEnZkr1BV2uHDjGzOQYCwAAgFxzu0Oqqnq3pJ7nPOUonY5q+fJzFI2ukd8/RnPnPiOvt9Z0rIOMGnW5pk17UJK0fft3tWHDV0ryuXwkkt0v4/ONktdbYzYMYIBlubt3gDHOLGvPnse0adPXJUlTp96vurp/yuvj1daepunTfyJJ2rbtDu3c+Z95fTzgWFDMAEARCQanSXIrne75NBoOrbU1O3N9yBDGmAEAgNLjPMdxnvOUm0wmpVWrPqaOjpfl8dRo7txnVFExxnSswxo16rMHlDP3aMOGr5ZcORONZosZTsugnDnjzDo7l5oNUgDa2v6iNWsWSZLGjPmyRo/+/KA87ogRn9CECbdIktat+7z27XtmUB4X6C+KGQAoIi6XX8HgVEmMMzuSeHy3IpHlkqSamg8YTgMAAJB7tbULJEmtrX8qu10Gtm1r/frPa9++38rlqtDs2f/TfbK8kI0a9VlNnfoDSdL27Xdr48ZrS6qciUaz+2VCoZmGkwDm9BQz5X1iJhpdqxUrzpNtJ1RXd6EmT75zUB9//PibNHz4pZLSWrXq4rL/7wOFiWIGAIoMe2aOrq1tiSQpHD5OPl+d4TQAAAC5V1n5Trnd1Uqn27V//+um4wyqLVtu1a5dP5Lk0owZj6mm5r2mI/XZ6NGf09Sp/yFJ2rbtrpIqZyKRbDETDFLMoHyFw/MllXcxk0g0a9myhUqlWlVVdZJmzHhUljW4b0FblqXp03+kmpr3K53u1LJlZyse3zGoGYCjoZgBgCLjfAKNYubwWlqyIz1qaxljBgAASpNluVVbe7qk8hpntnPnf2rz5lskSVOnPqD6+vPNBhqA0aOv0NSpD0hyypl/K4lyxhllFgoxygzlKxSaI0lKJHYokdhrOM3gS6e7tGLFhxWLbVRFxSTNnv3fcrsDRrK4XD7NmvWkgsEZSiR2aNmys5VK7TeSBTgUihkAKDLOmIZolGLmUGzb7n5zgmIGAACUMue5Tmvrs4aTDI69e3+rdeuukCSNH3+jRo/+nOFEAzd69Oc1der3JUnbtn1HGzd+rajLmXQ6olhssyROzKC8eTyVCgSmSJIikfI6NWPbGa1e/Ql1dLwij6dWc+f+QT7fMKOZvN5azZnzB3m9wxSJvKVVqz6qTCZlNBPgoJgBgCLTM8psVVG/eMuXaHSNEomdcrkqVF1dPGMtAAAA+svZM9Pe/lel0xHDafKrvf0lrVr1UUkZjRjxaU2Y8A3TkY7Z6NFfOKCcuVMbN15XtM/vo9G1kmx5PEPl89WbjgMYFQo5e2aWmg0yyDZu/Dft3fukLMun2bOfVjA43XQkSVIgMEFz5vyPXK6AWlr+qMbGLxbt71qUFooZACgyweA0WZZH6XSH4vHtpuMUHOe0THX1e+V2VxhOAwAAkD+BwBT5/eNk20m1tb1gOk7eRCKrtXz5PymTiWno0H/StGkPyrIs07Fyonc5c4c2bbq+KN8w7BljxmkZIBx2ipnyOTGzY8d/aNu2uyRJDQ0Pq6bmfYYT9VZV9S7NmPGYJEs7dz6obdvuNh0JoJgBgGLjcvkUCEyVxJ6ZQ2GMGQAAKBeWZR0wzqw098zE4zu0bNmHlEq1qrLyRM2c+Su5XB7TsXJq9OgvaMqU+yVJW7feXpTlTCSyShJjzABJCofnSyqfYmbfvt9r/fovSpImTvyWhg//Z8OJDq2+/jxNnnyPJGnjxq+qqek3hhOh3FHMAEARYs/MoWUySbW1PSeJYgYAAJQHZ5xZKe6ZSSbbtGzZWYrHtyoQmKY5c34ntztkOlZejBlzpaZM+Z4kp5z5elGVMz0nZmYYTgKY55yYiUZXKZOJG06TX/v3v6GVK50xk5dp3LjrTUc6ojFjvqTRo6+UJK1Z8wm1t79sOBHKGcUMABShnj0zFDMH6uh4Vel0p7zeuu4nwwAAAKWstvYDkqRIZJkSiT2G0+ROOh3TihXnKRJZLp9vhObO/V/5fHWmY+XVmDFf1JQp90mStm69TZs23VA05QwnZoAefv9YeTy1su2UIpHVpuPkTSy29e0xkxHV1i4oijGTlmVpypR7NXToOcpkYlqx4sPq6tpgOhbKFMUMABQh58SM8wIIWc4Ij5qaD8iy+CcOAACUPp+vvntsTmvrErNhcsS2M1qz5lK1tz8vt7tSc+b8UYHABNOxBsWYMf96QDnz79q06caCL2cymYS6uholsWMGkLJv/vfsmVlqNkyepFLtWr78bCUSuxQKzdasWb+Ry+U1HatPLMutGTMeUzj8DiWTe7Vs2UIlky2mY6EM8a4VABShnlFmqwr+hdpgcoqZIUMYYwYAAMpHz56Z4h9nZtu2GhuvUnPzr2VZXs2e/bQqK+ebjjWosuXMvZKkrVu/rc2bbyro5/xdXeslpeV2V8rnG2U6DlAQQqFsMROJlN6emUwmqZUrL1IkskI+30jNmfN7eTzVpmP1i8cT1pw5v5PfP1ZdXeu0YsV5JT92DoWHYgYAilAgMFWW5VU6vV/x+DbTcQpCKtWujo7XJPXMWgcAACgHPXtmFhf0G/h9sW3bndqx435JUkPDz1Rbe7rhRGaMGfMlTZ78XUnSli3f0ubNNxfsf7cHjjEr9DFGwGBxTjJ2dpZWMWPbttat+5xaW5+VyxXSnDm/U0XFONOxBsTvH6k5c/4gt7tK7e0vaM2aywr29yxKU0EUMw888IAmTJigiooKnXjiiXrttdcOe20ymdStt96qyZMnq6KiQvPmzdMzzzzT65oJEybIsqyDvr7whS90XxOLxfSFL3xBQ4cOVTgc1oUXXqg9e0pnHi+A0uZyeRUITJPEnhlHW9tzktIKBKaqomK86TgAAACDprr6FFmWX/H4dnV1rTMdZ8B27/6ZNm78miRp8uTvavjwSwwnMmvs2Ks0efI9kqQtW75ZsOVMNJrdoREKzTCcBCgcB44yK8T/vx2orVv/Xbt3PyTJpVmzHldl5TtMRzom4fBszZr1pCzLo6amx7R5802mI6GMGC9mHn/8cV1zzTW6+eab9cYbb2jevHk688wz1dTUdMjrb7jhBv3whz/U/fffr1WrVulzn/uczj//fL355pvd1/ztb3/Trl27ur8WL86Otrn44ou7r7n66qv1P//zP/r1r3+t559/Xjt37tQFF1yQ3x8WAHKoZ88MxYwktbRkf9c7ozwAAADKhdsdUHX1eyT1PCcqNvv2PaO1az8tSRo79isaO/Yqs4EKxNixV2vy5LslOeXMLWYDHcKBJ2YAZIVCM2VZHqVSrYrHt5uOkxN79vxCmzbdIEmaOvV+DR16tuFEuTFkyAJNm/ZDSdkTirt2PWw4EcqF8WLmnnvu0Wc+8xktWrRIM2fO1IMPPqhgMKiHHnrokNc/+uijuv7667Vw4UJNmjRJV1xxhRYuXKi77767+5r6+nqNGDGi++t3v/udJk+erFNPPVWS1N7erp/85Ce65557dPrpp+v444/Xww8/rJdeekmvvPLKoPzcAHCsevbMUMxIPTPVGWMGAADKUc84s+LbM9PR8TetXHmRbDulYcM+rkmT7jAdqaCMHXvNAeXMrdq06Razgf5BNJotZkIhihnA4XL5FQxmT5F1di41GyYH2tqe15o1l0nKluejR3/ecKLcGjnyMo0b93VJ0rp1l6ulpfj+LUXxMVrMJBIJvf7661qwoOdNNJfLpQULFujll18+5PfE43FVVFT0ui0QCOjFF1887GP8/Oc/12WXXdY96/T1119XMpns9bgNDQ0aN27cER+3o6Oj1xcAmOR8Io0TM1Istk1dXWsluVRT837TcQAAAAadc2q4re3PymRShtP0XTTaqOXLz1YmE1Ft7RlqaHhIlmX8M6QFJ1vO3CVJ2rLlG9q8+RuGE2VlMilFo9nxec6b0ACyesaZFfeemUhkjVasOF+2nVBd3YUlW55PnPhNDRv2Mdl2SitXXqjOzhWmI6HEGX22s3fvXqXTaQ0fPrzX7cOHD9fu3bsP+T1nnnmm7rnnHq1fv16ZTEaLFy/WU089pV27dh3y+qefflptbW361Kc+1X3b7t275fP5VFNT0+fHve2221RdXd39NXbs2L7/oACQBz2jzFbJtjOG05jV2pod2VFV9S55vTVmwwAAABhQWXmcPJ5apdMd2r//b6bj9EkisUfLlp2pZLJZ4fA7NGvWk3K5fKZjFayxY7+sSZO+I0navPmWgihnYrFNsu24XK4Aex6BfxAOz5ckRSLFW8wkEk1avnyhUqlWVVWdpBkzHi3Z8tyyLDU0PKzq6lOUTndo+fKzFY8f+v1mIBeK7v+T7rvvPk2dOlUNDQ3y+Xy68sortWjRIrlch/5RfvKTn+iss87SqFGjjulxr7vuOrW3t3d/bdu27ZjuDwCOVSAwRZblVSYTUSy21XQcoxhjBgAAyp1luVVb+wFJxTHOLJXar2XLFioW26iKikmaO/cP8ngqTccqeOPGfUWTJt0pySlnbjWaxxljFgw2yLLcRrMAhSYUck7MLDUbZIDS6aiWL/+wYrFNqqiYpNmzfyu3O2A6Vl65XH7Nnv1fCgSmKh7fquXLz1E6HTEdCyXKaDFTV1cnt9utPXv29Lp9z549GjFixCG/p76+Xk8//bQikYi2bNmiNWvWKBwOa9KkSQddu2XLFj377LP6l3/5l163jxgxQolEQm1tbX1+XL/fr6qqql5fAGCSy+VVMDhdUnnvmbHtzAHFzBmG0wAAAJjTs2dmseEkR5bJJLRy5UXq7HxDXm+d5s59Rj7f8KN/IyRJ48Z9tXuU0ObNN2vz5m8ayxKJrJbEGDPgUJxRZl1dG5RK7Tecpn9sO63Vq/+f9u9/VR7PEM2d+0f5fPWmYw0Kr3eo5sz5g7zeOnV2vq5Vqz4m206bjoUSZLSY8fl8Ov7447VkyZLu2zKZjJYsWaKTTz75iN9bUVGh0aNHK5VK6cknn9S555570DUPP/ywhg0bprPPPrvX7ccff7y8Xm+vx127dq22bt161McFgEISDDrjzMq3mOnsXKZkslkuV0hVVSeZjgMAAGCM8yGVjo6XlUp1Gk5zaLad0dq1n1Zr6//J5Qpqzpw/KBicajpW0Rk37toDypmbtHnzt4zkcE7MhEIzjTw+UMh8vnr5fKMk2YpElpuO0y8bNlyrvXv/S5bl0+zZTysYnGY60qAKBqdo9uz/lmX5tW/f/6ix8RrTkVCCjI8yu+aaa/SjH/1IjzzyiFavXq0rrrhCkUhEixYtkiRdeumluu6667qvf/XVV/XUU09p48aNeuGFF/ShD31ImUxG1157ba/7zWQyevjhh/XJT35SHo+n159VV1fr05/+tK655hr9+c9/1uuvv65Fixbp5JNP1kkn8aYegOLRs2emfIsZ57RMTc2pzCQHAABlLRCYpIqKibLtlNrbnzcd55A2brxOe/b8XJJbs2b9RlVVJ5iOVLSy5cztkqTNm2/Uli3fHvQM0SgnZoAjcU7NdHYWz56Z7du/r+3b75EkNTQ8rJqaUwwnMqO6+t2aMeNRSdKOHd/T9u33GU6EUuM5+iX59dGPflTNzc266aabtHv3bs2fP1/PPPOMhg/PHmPeunVrr/0xsVhMN9xwgzZu3KhwOKyFCxfq0UcfVU1NTa/7ffbZZ7V161Zddtllh3zc7373u3K5XLrwwgsVj8d15pln6j/+4z/y9nMCQD5QzPSM6mCMGQAAQHac2a5dP1Jr67MaOvTso3/DINq+/T5t25bdj9LQ8BMNHXqW4UTFb9y4f5Nt29q06Tpt2nSDJEvjx18/KI9t25kDRplxYgY4lHB4vlpa/lg0xczevf+jxsYvSZImTvy2hg//Z8OJzBo27GLFYndo48Z/U2Pj1aqomKC6uoOnNgEDYdm2bZsOUYw6OjpUXV2t9vZ29s0AMCYaXavXXmuQyxXUKafsl2UZPwg5qNLpmP7611plMjGdcMKK7qIKAJB7PP9Ff/G/GTOamn6tVas+olBotk44oXBG5zQ1Pa5Vqz4mydbEif+u8eOvO+r3oO+2bLlNmzZlC5mJE789KOVMLLZFr7wyQZbl1SmnRORyefP+mECxyf7uu0SVlSfq+ONfMR3niPbvf11vvvk+ZTJRjRjxaU2f/iNZlmU6lnG2bWvduiu0a9cP5XIFNH/+85z2xBH19Tlweb2DBwAlpqJisizLp0wmqlhsi+k4g66j4yVlMjH5fCP5lB4AAICk2trTJVmKRFYoHt9lOo4kqbX1T1q9+lJJtkaPvlLjxn3NdKSSM378dZo4MTvKbNOmr2vLltvy/pjOaZlAYCqlDHAY4fB8SVIksrygF8jHYlu1fPk/KZOJqrb2DE2b9gNKmbdZlqWpU7+vIUM+pEymS8uXn6Ours2mY6EEUMwAQBFzuTwKBhsklec4s54xZgt40ggAACDJ6x2qcPgdkqTW1iWG02T3KqxYcZ5sO6H6+os0Zcq9PG/Lk/Hjrz+gnLleW7bcntfHi0ZXSZJCIT4gBRxOIDBFLldAmUxUXV2NpuMcUirVrmXLFiqR2K1QaI5mzfo1Zes/cLk8mjnzCYVC85RM7tHy5QuVTLaZjoUiRzEDAEXOGd8VjZZfMdPSwn4ZAACAf1Rbu0BSz4dYTOnq2qxly85SOr1f1dWnqqHhUVmW22imUpctZ74lSdq06Tpt3XpH3h4rEskWM8HgjLw9BlDsLMutUGiOJBXknplMJqEVKy5UNLpSPt9IzZnze3k81aZjFSSPp1Jz5vxOPt8oRaOrtXLlhcpkEqZjoYhRzABAkXOKmXI7MZNM7lNn5xuSet58AAAAgDRkSPZDK62tz8rUWtlkcp+WLfuQEoldCoVma/bsp+V2VxjJUm7Gj/+6Jkz4piRp48avaevWO/PyONFodpQZI4WBI3PGmRVaMZPdnfI5tbUtkcsV0pw5v1dFxVjTsQpaRcUYzZnze7ndYbW1/Ulr115u7N9ZFD+KGQAocsFgeRYzra1/kmQrGJwlv3+k6TgAAAAFo6rqPXK5KpRI7Ox+83wwpdNRLV/+T+rqWiu/f6zmzn1GXm/NoOcoZxMm3KAJE26VJG3c+G85L2ds22aUGdBH4fA8SVJn51KzQf7Bli3f1u7dD0tyadasJ1RZeZzpSEWhsnK+Zs58QpJbe/Y8oi1bvmU6EooUxQwAFLmeUWarZdsZw2kGjzOaw/lEKAAAALLc7gpVV79XUvbUzGDKZFJateqj6uh4RR5PrebOfUZ+/+hBzYCsCRNu/Idy5js5u+9EYo9SqTZJLgUC03J2v0ApKsQTM7t3/1ybN98oSZo69QENHbrQcKLiMnToWZo27QFJ0ubNN2n37p8bToRiRDEDAEUuEJgky/Irk+lSLLbJdJxB47zJwH4ZAACAgznPkQZzz4xt21q//grt2/c7uVwVmjPnfzhNYVi2nPmGJGnjxmu1detdOblf57RMIDCJEXXAUTg7ZhKJHUok9hpOI7W1Pa+1ay+TJI0d+1WNHv05w4mK06hRn9XYsV+VJK1de5na2p43nAjFhmIGAIqcZbkVDDZIKp9xZl1dGxSLbZJleVVd/T7TcQAAAAqOs4Ovre05ZTLJQXnMzZtv0a5dP5bk0syZv1J19XsG5XFxZBMm3KQJE26RJG3c+FVt23b3Md9nJJItZtgvAxydx1OpiorJkqRIxOypmUhktVasOE+2nVR9/UWaNOl2o3mK3aRJt6u+/iLZdlIrVpynSGSN6UgoIhQzAFACnHFm5VLMtLRkP/lZVXWyPJ6w4TQAAACFJxyeL49nqNLpTu3f/1reH2/nzh9qy5bs2Kxp0/5DdXXn5v0x0XcTJtys8eNvliRt2PAVbdt2zzHdn7O7KBiccczZgHJQCOPMEok9Wr58oVKpNlVVnayGhp/Jsnhr+FhYlksNDT9TVdXJSqXatHz5QiUSTaZjoUjw/30AUALKrZhxRnIwxgwAAODQLMul2toPSOr5UEu+NDc/rXXrPi9JGj/+Jo0a9dm8Ph4GZuLEWw4oZ758TOWMM8qMUXVA34TD8yRJnZ1LjTx+Oh3V8uUfViy2WRUVkzV79n/L7Q4YyVJq3O6AZs/+b1VUTFIstknLl39Y6XSX6VgoAhQzAFACnGImGi39Ysa202pr+5OknhEdAAAAOFjPnpln8/YY7e1/1erVH5OU0ciR/9I9MguFKVvO3CTJKWe+O6D76RllxokZoC9Mnpix7bRWr/5/2r//NXk8QzR37h/k89UPeo5S5vPVa+7cP8jjqdX+/a9q9er/J9vOmI6FAkcxAwAlIBh0ipk1su204TT5tX//60ql2uR2V6uy8p2m4wAAABQs50MsHR2vKJXqyPn9RyKrtHz5OcpkYho69BxNnfoDWZaV88dBbk2YcIvGj79RkrRhwzX9LmeSyX1KJrOjepxdlwCOzDkxE42uViaTGNTH3rDhq9q7979kWT7Nnv3fCganDerjl4tgcLpmz35aluXT3r1PacOGa01HQoGjmAGAEhAITJTLVaFMJqauro2m4+RVzxiz0+VyeQynAQAAKFyBwIS3F06n1db2fE7vOxbbrmXLPqRUqlVVVSdp5sxf8dysSFiWpQkTvqHx42+Q5JQz9/b5+yOR7H4Zv3+cPJ7KfEQESo7fP1YeT41sO9l94mwwbN9+v7Zvz5avDQ2PqKbmvYP22OWopuZ9amh4WJK0ffvd2rHjPwwnQiGjmAGAEmBZ7u4xAs6851LljOJgjBkAAMDRDRnijDPL3Z6ZZLJNy5efpXh8mwKB6Zoz53dyu4M5u3/kX7acuVXjxn1dkrRhw9Xavv2+Pn1vNJotZhhjBvSdZVnd48wikcEZZ7Z372/V2HiVJGnixH/X8OGXDMrjlrvhw/9ZEyd+S5K0fv0XtW/f7w0nQqGimAGAEuHsmYlESnfPTDodUXv7XyX1zEwHAADA4TkfZsnVnpl0OqYVK85VJLJCPt9IzZv3v/J6h+bkvjG4LMvSxInf7C5nGhuv0vbt3zNbqecAAHOtSURBVDvq9zkfBAuFZuY1H1BqQqHsOLPOzqV5f6yOjr9r1aqe/V/jxn0t74+JHuPGXa8RIy6TlNHKlR/V/v1vmo6EAkQxAwAlwtkzU8rFTFvbX2TbSfn94xUITDEdBwAAoODV1JwuyVI0ulrx+I5jui9ngXR7+1/kdldp7tw/qqJifG6CwoiecuZ6SVJj45e0ffv9R/weZwxTMEgxA/SHc2KmszO/J2ZisS1avvyflMlEVVv7QU2d+h/s/xpklmVp2rQHVVu7QJlMRMuXn61YbJvpWCgwFDMAUCLK4cTMgWPMeGIJAABwdF5vrSor3ynp2E7N2Lat9eu/pL17n3x7gfTT3cusUdyy5cy3NG7cdZKkxsZ/PWI544wyC4UYZQb0h/M7s7PzLdm2nZfHSCbbtGzZQiWTexQKzdGsWb+Wy+XNy2PhyFwur2bN+o2CwVlKJHZp+fKzlUp1mI6FAkIxAwAlwilmotE1su204TT54cxGd2alAwAA4OicEbDHUsxs3Xq7du58QJKlGTMeVW3t+3OUDoUgW858u3vcUbac+f5B16VS+xWPZz/1zY4ZoH9CoZmyLI9SqRbF49tzfv+ZTEIrV16oaHSVfL5RmjPn9/J4qnL+OOg7j6dac+f+Xj7fCEUiy7Vy5cXKZJKmY6FAUMwAQImoqJgglysg246rq2uD6Tg5F4/vViSyXJKlmpoPmI4DAABQNA7cMzOQT2nv2vVTbdqUHXU1Zcp3NWzYR3KaD4UhW878u8aO/TdJUmPjF7VjxwO9rolG10iSvN7h8nqHDHpGoJi5XP7uQjPX48xs29batZerre1PcrvDmjPn96qoGJvTx8DAVFSM15w5v5PLFVRr6/9p/frP5+3EFIoLxQwAlAjLcnU/ySvFcWZtbUskSeHwcfL56gynAQAAKB7V1e+WyxVQIrG7388T9+37o9au/RdJ0tix12rMmC/lIyIKhGVZmjTptu5yZv36K7Vjx39Iyu4Yam5+UpLk840s2VP6QD71jDNbmtP73bLlm9qz5xFJbs2c+YQqK+fn9P5xbCorj9fMmb+S5NKuXT/W1q13mI6EAkAxAwAlpJT3zLS0ZMeYOZ/4BAAAQN+4XH5VV79PUs9o2L7o6HhNK1deJCmt4cM/oUmTbstTQhSSnnLmWknS+vVf0Nq1l+uVVyZo27bsm4mRyFK98soENTc/ZTIqUHTC4fmSpEgkdydmdu/+mTZvvlmSNG3aAxo69Kyc3Tdyp67uHE2Zcq8kadOm69TU9LjZQDCOYgYASkjPnpnSKmZs2+5+E8GZkQ4AAIC+O3CcWV9Eo+u1fPnZymSiqq39oKZP/4ksi7cQykW2nLldY8d+VZK0a9ePDtqJEY/v0MqVF1HOAP0QCjknZnJTzLS2PtfrVOOoUZ/Nyf0iP8aM+aLGjLlKkrR69SfV3v7X/9/encdFUf9/AH/NLrDcNwIKyqkiIqaYV5Q3aHmheaSlZtbXr5ZmpllmlqVd9oMuLfMsb0PTSk1JDe8TQ8UbBJFT7muB3fn9wZfVFVBA2GHh9eyxD3dnPjPz3m2s+cx7Pu+PtAGRpHhVRUTUiJiaNs4RMwUFl1FcfAcymTGsrJ6SOhwiIiIivWNrW/ZwS1bWIajVxQ9tq1Qm499/g1BSkg5z887w9d0GmcxQF2FSA1I+54xcbl5Fi7I5Eq5fn8myZkTVVF7KrLDwOkpL8x5rX/n5Mbh4cThEsQQODs9zVKOe8PT8EnZ2QyGKSkRHD0VBwTWpQyKJMDFDRNSI3BsxcwVqdanE0dSd8tEyVlZPQS43ljgaIiIiIv1jZuYHQ0MHqNX5yMk5XmW70tJcREcPQlFRLIyNPdGhwx8wMLDQYaTUkGRnH4ZK9bCbxyKUygRkZUXqLCYifWZk5AAjo+YAROTnR9d6P8XFKYiOHoTS0ixYWvZA27ZrOapRTwiCHO3arYeFRReUlt5FdPQgFBenSx0WSYB/Y4mIGhFj41aQyUwhisUoLLwudTh1hmXMiIiIiB6PIMgeWc5MrS7GxYshyMs7B0NDB3TosAdGRo66DJMamOLipDptR0T3Rs3k5UXVanuVqgDR0YNRVBQHExMvtG//G+RykzqMkOqbXG4GP79dUChaobDwOi5cGAaVqkjqsEjHmJghImpEBEEGM7N2ABrPPDNqdQmysg4CYGKGiIiI6HHcS8zsq7BOFNW4fPllZGbuh0xmBj+/P2Fq6qXrEKmBMTJyrtN2RASYm3cEULt5ZkRRhZiYccjNPQUDAzv4+f0JIyP7Oo6QdMHIyBEdOvwJudwKOTlHcPnyRIiiWuqwSIeYmCEiamQa2zwzOTknoFLlwdDQXvNkERERERHVXHliJifnJEpLs7XW3bw5F6mp6yEIBmjf/ldYWgZIESI1MNbWgVAoXAAIVbQQoFC4wto6UJdhEem18n5tfn7NEzM3bsxGevoOCIICfn6/wdTUu67DIx0yM2uH9u3DIQgGSEvbjNjY+VKHRDrExAwRUSNTPs9Mfv4liSOpG+VPdFpb92XNXCIiIqLHYGzcEiYmrQGocfv2N0hJ2YjMzIOIj1+KhIQvAQBt2qyCrW2QtIFSgyEIcnh5hZV/enAtAMDLKxSCINdpXET6zMysvJTZvxBFVbW3u337a9y+HQoA8PFZCyurnvURHumYjU0ftGnzEwAgPn4J7txZIXFEpCu8w0VE1MiUJ2YaSymz8hrotrYsY0ZERPQw3333Hdzc3GBsbIyuXbvi5MmTD22/detWtG3bFsbGxvDz88Off/6ptX7ixIkQBEHrFRwcXJ9fgXTA2NgdABAX9z5iYl7A+fO9cfPmbACAh8encHJ6UcrwqAFycAiBr+82KBQttJYrFC7w9d0GB4cQiSIj0k+mpt6QyUygVhegsPBGtbZJT/8N16/PBAC4uy9Bs2aj6zFC0jUnpwlo1WoBAODq1anIyPhL4ohIF5iYISJqZO4lZq5ArS6ROJrHU1qajZycEwDuld4gIiKiijZv3oxZs2bhgw8+wNmzZ+Hv74+goCCkpqZW2v7o0aMYO3YsJk+ejHPnzmHYsGEYNmwYLly4oNUuODgYSUlJmtfGjRt18XWonqSlhSMzc2+V642NOacMVc7BIQTdusXB3/8AfHw2wN//ALp1i2VShqgWBEEOMzM/AEBeXtQj2+fknMKlS2MBiHB2noKWLefWb4AkCTe3hXB0HA9AhYsXRyIv71+pQ6J6xsQMEVEjo1C0hFxuDlEsQWHhdanDeSxZWQcBqGBi4g1j41ZSh0NERNRgffXVV5gyZQomTZqEdu3aYfny5TA1NcWqVasqbR8WFobg4GC8/fbb8PHxwaJFi9CpUyd8++23Wu0UCgWcnJw0LxsbG118HaoHoqjC9eszHtJCwI0bb9aorA41LYIgh41NLzg6joWNTS+WLyN6DOXzzOTlPXyemcLCOERHD4ZaXQgbmyB4e38HQahqzifSZ4IgoE2bn2Bl9QxUqlxERz8LpfKO1GFRPWJihoiokREEAaam7QAA+fn6Xc4sI6NsfhkbG5YxIyIiqkpxcTHOnDmDfv3ujS6VyWTo168fjh07Vuk2x44d02oPAEFBQRXaHzx4EM2aNUObNm0wdepU3L17t8o4lEolcnJytF7UcGRlRUKpvP2QFiKUygRkZUXqLCYioqbK3LwjgIcnZkpKshAdPQglJSkwM+sAX98tkMkMdRQhSUEmU6B9++0wMWkDpfI2oqOfQ2lpntRhUT1hYoaIqBFqLPPMlM8vwzJmREREVUtPT4dKpYKjo6PWckdHRyQnJ1e6TXJy8iPbBwcHY926dYiIiMBnn32GQ4cOYeDAgVCpKh9RsWTJElhZWWlerq6uj/nNqC4VFyfVaTsiIqq9eyNmoipdr1YX4+LFEBQUxMDIqDn8/P6AgYGlDiMkqRga2qBDhz9haOiAvLxzuHRpDNTqUqnDonrAxAwRUSPUGEbMFBUloLDwCgAZrK17Sx0OERFRkzNmzBgMGTIEfn5+GDZsGH7//XecOnUKBw8erLT9vHnzkJ2drXklJCToNmB6KCMj5zptR0REtWdm1gEAUFyciOLidK11oijiypUpyMo6ALncHH5+f8DY2EWKMEkiJiYe8PPbBZnMGBkZf+D69RkQRVHqsKiOMTFDRNQIlY+Y0efETGZmWRkzS8snYWhoLW0wREREDZi9vT3kcjlSUlK0lqekpMDJyanSbZycnGrUHgA8PDxgb2+P69crn8NOoVDA0tJS60UNh7V1IBQKFwBVzU0gQKFwhbV1oC7DIiJqkgwMLGBs7AkAyM/XLmd269ZHSElZB0COdu22wMKio+4DJMlZWnaFj88vAATcufM9bt/+P6lDojrGxAwRUSNUnpgpLLwKtbpY4mhqh2XMiIiIqsfIyAidO3dGRESEZplarUZERAS6d+9e6Tbdu3fXag8A+/btq7I9ANy+fRt3796FszNHVOgjQZDDyyus/NODawEAXl6hnNCdiEhH7pUzu5eYSU5ei7i4hQCA1q2/h53dQClCowbCwWEEPD2/AADcuDEbaWm/ShwR1SUmZoiIGiGFwhVyuQVEsRSFhdekDqfGRFF9X2Kmv8TREBERNXyzZs3CihUrsHbtWsTExGDq1KnIz8/HpEmTAAAvvfQS5s2bp2k/Y8YM7NmzB0uXLsXly5excOFCnD59GtOnTwcA5OXl4e2338bx48cRFxeHiIgIDB06FF5eXggKCpLkO9Ljc3AIga/vNigULbSWKxQu8PXdBgeHEIkiIyJqesrLmaWl7UBm5kFkZOzDlSuvAABcXeeiefNXpQyPGggXl1lo3vy/AETExIxHTs4JiKIKmZkHkZKyEZmZByGKlc//Rw2bgdQBEBFR3RMEAaam7ZCbewL5+Rc1I2j0RV7evygpSYNMZgZLy25Sh0NERNTgjR49GmlpaViwYAGSk5PRsWNH7NmzB46OjgCA+Ph4yGT3nsvr0aMHNmzYgPnz5+Pdd9+Ft7c3duzYgfbt2wMA5HI5/v33X6xduxZZWVlo3rw5BgwYgEWLFkGhUEjyHaluODiEwN5+KLKyIlFcnAQjI2dYWwdypAwRkQ6lpYUjMfEbAEBOTiTOn++NstGLIhwcRsHDY7Gk8VHDIQgCvLzCUFQUh4yMP3H+/ADI5aYoLk7WtFEoXODlFcYHLPSMIHLmoFrJycmBlZUVsrOzWTuZiBqky5cnIzl5FVq1WgB39w+lDqdG4uO/xM2bb8PW9ll06PC71OEQERF4/Us1x3OGiIioorS0cFy8OBJA5bdkfXw2wtFxjG6DogavtDQPp093QFFRbCVry0qScvRrw1Dda2CWMiMiaqTKR8nk51+UOJKay8zcB4DzyxAREREREVHjIYoqXL8+A1UlZQABN2/OYWkqqkAuN4FaXVTF2rLz6fr1mTx39AhLmRERNVLliZmCgksSR1IzKlURsrP/AQDY2nJ+GSIiIiIiImocsrIioVTefkgLEUplArKyImFj00tHUZE+KC9BWrWycyc6eijMzNrD0NAOhoa2MDCwe+C9LWQyI53FTVVjYoaIqJEyNS1LzBQWXoNaXaw3/+PNyTkKtboIRkbOMDVtJ3U4RERERERERHXi4TfWa96Omo7qnhMZGX8gI+OPh7aRy801SRpDQ7tqvrfWi/noRFGlN/PoMTFDRNRIKRQtIJdbQqXKQUHBVZibt5c6pGq5v4yZIAgSR0NERERERERUN4yMnOu0HTUd1T0nHB0nwcDAEqWlGSgpuYuSkrua96WlmQBEqFR5UKnyoFTeqkEEAgwMrP+XpLH93yice+8rW2ZoaAe53EJn93bS0sJx/foMrVFpCoULvLzCGuTcO0zMEBE1UoIgwMzMFzk5x1BQcFFvEjMZGeWJGZYxIyIiIiIiosbD2joQCoULlMpEVD7PjACFwgXW1oG6Do0auOqeO23brqhyhIgoqlFamqWVrNF+n4HS0rsV3qtUuQBElJZm/i+5U32CYPC/RE1lZdW0S6zdP0pHLjet0XHS0sJx8eLICr+NUpmIixdHwtd3W4NLzjAxQ0TUiJUnZvLzL0odSrWUlNxFXt5ZAGUjZoiIiIiIiIgaC0GQw8sr7H83kAVo30QuG1Xg5RXaYEsvkXTq4twRBNn/EiC2NTq2Wl1SSfKmfBROVe8zoFYXQhRLUVKSipKS1BodUyYzfkhZNe3ROQYG1rh2bToqT1iJAARcvz4T9vZDG9TfLSZmiIgasfJ5ZvQlMZOZ+TcAEaamvlAoOHSbiIiIiIiIGhcHhxD4+m6rouRSaIN7qp8aDqnOHZnMEEZGjjAycqzRdipVYZWjcyofsVP2XhRLoVYXobg4EcXFiXXwDUQolQnIyoqEjU2vOthf3WBihoioETMzawdAnxIzZWXMbG1ZxoyIiIiIiIgaJweHENjbD9WbScqp4dCnc0cuN4Fc7gLApdrbiKIIlSq3yrJqlb1XKpOgVuc/ct/FxUmP8W3qHhMzRESNmJlZ2YiZwsLrUKuVkMkUEkf0cJmZ+wFwfhkiIiIiIiJq3ARB3qCe3if90ZjPHUEQYGBgCQMDSwBu1domM/Mgzp/v/ch2RkYNqzKLTOoAiIio/hgZNYdcbgVAhYKCK1KH81CFhTdQVBQLQTCEldXTUodDREREREREREQNnLV1IBQKF5TPtVORAIXCFdbWgboM65GYmCEiasQEQdCMmmno5cwyMsrKmFladoeBgbnE0RARERERERERUUMnCHJ4eYWVf3pwLQDAyyu0wZV7Y2KGiKiR05fEDMuYERERERERERFRTTk4hMDXdxsUihZayxUKF/j6boODQ4hEkVWNc8wQETVy5YmZgoKGm5gRRRWysv4GANjY9JM4GiIiIiIiIiIi0icODiGwtx+KrKxIFBcnwcjIGdbWgQ1upEw5JmaIiBo5U9OGP2ImN/cMSkszIZdbwcIiQOpwiIiIiIiIiIhIzwiCHDY2vaQOo1pYyoyIqJErHzFTWHgDKlWRxNFULjOzbH4ZG5s+kMn4zAARERERERERETVeTMwQETVyRkZOMDCwAaBGYeEVqcOp1L35ZVjGjIiIiIiIiIiIGjcmZoiIGjlBEDSjZhpiOTOVKh/Z2UcAADY2/SWOhoiIiBoDlVqFg3EHsTF6Iw7GHYRKrZI6JCIiIiIiDdaLISJqAkxNfZGdfbhBJmaysv6BKJZAoWgFExMvqcMhIiIiPRceE44Ze2bgds5tzTIXSxeEBYchxCdEwsiIiIiIHp9KrUJkfCSScpPgbOGMwJaBkMsa5gT3VDUmZoiImoCGPGKmvIyZrW1/CIIgcTRERESkz8JjwjFyy0iIELWWJ+YkYuSWkdg2ahuTM0RERKS3+ABK48FSZkRETUB5YqagoCEmZvYB4PwyRERE9HhUahVm7JlRISkDQLNs5p6ZLGtGREREeqn8AZT7kzLAvQdQwmPCJYqMaoOJGSKiJqA8MVNYeAMqVaHE0dyjVCYjPz8agABr675Sh0NERER6LDI+ssKNivuJEJGQk4DI+EgdRkVERET0+PgASuPDxAwRURNgaNgMBga2AEQUFFyWOhyNrKwIAIC5+RMwMrKXOBoiIiLSZ0m5SdVqdyf3Tj1HQkRERFS3+ABK4yN5Yua7776Dm5sbjI2N0bVrV5w8ebLKtiUlJfjoo4/g6ekJY2Nj+Pv7Y8+ePRXaJSYmYvz48bCzs4OJiQn8/Pxw+vRpzfqJEydCEAStV3BwcL18PyKihkAQhAY5z0xGBsuYERERUd1wtnCuVruP//kYO6/shFpU13NERERERHWjug+gVLcdSU/SxMzmzZsxa9YsfPDBBzh79iz8/f0RFBSE1NTUStvPnz8fP/zwA7755htcunQJ//nPfzB8+HCcO3dO0yYzMxM9e/aEoaEhdu/ejUuXLmHp0qWwsbHR2ldwcDCSkpI0r40bN9brdyUiklpDm2dGFMX75pfpL3E0REREpO8CWwbCxdIFAoSHtotJj8HQTUPhv9wfG6I3oFRdqqMIiYiIiGrn6t2r1WpX3QdVSHqSJma++uorTJkyBZMmTUK7du2wfPlymJqaYtWqVZW2//nnn/Huu+9i0KBB8PDwwNSpUzFo0CAsXbpU0+azzz6Dq6srVq9ejSeffBLu7u4YMGAAPD09tfalUCjg5OSkeT2YuHmQUqlETk6O1ouISJ+YmjasETMFBZdRXHwHMpkxrKyekjocIiIi0nNymRxhwWEAUCE5I/zvn5VDVuKdnu/AwsgCF1IvYFz4OLT5tg1+PPMjlKVKKcImIiIiqpJKrcK7Ee9i4aGFj2zraumKwJaB9R8U1QnJEjPFxcU4c+YM+vW7V75GJpOhX79+OHbsWKXbKJVKGBsbay0zMTHB4cOHNZ937tyJgIAAPP/882jWrBmeeOIJrFixosK+Dh48iGbNmqFNmzaYOnUq7t69+9B4lyxZAisrK83L1dW1Jl+XiEhyDa2UWfloGSurpyCXGz+iNREREdGjhfiEYNuobWhh2UJruYulC7aN2oaXn3gZS/otQfyb8fi498ewN7XHzcybeO331+DxtQe+OvYV8orzJIqeiIiI6J6Mwgw8u+FZLDm8BAAwuPVgzcMmlZn8xGTIZXJdhkiPQbLETHp6OlQqFRwdHbWWOzo6Ijk5udJtgoKC8NVXX+HatWtQq9XYt28fwsPDkZR0r3bezZs3sWzZMnh7e2Pv3r2YOnUq3njjDaxdu1bTJjg4GOvWrUNERAQ+++wzHDp0CAMHDoRKpaoy3nnz5iE7O1vzSkhIeMxfgIhIt8oTM0VFsVCpCiSOBixjRkRERPUixCcEcTPicGDCAWwI2YADEw4gdkYsQnxCNG2sja3x3tPvIW5GHEKDQtHCogXu5N7BW3+9BbdQNyw6tAiZhZkSfgsiIiJqyv5N+RddVnTB3ht7YWJggvUh67Fz7M5KH0AxNTQFAISdCMP1jOtShEu1IIiiKEpx4Dt37qBFixY4evQounfvrlk+Z84cHDp0CCdOnKiwTVpaGqZMmYJdu3ZBEAR4enqiX79+WLVqFQoLCwEARkZGCAgIwNGjRzXbvfHGGzh16lSVI3Fu3rwJT09P7N+/H3379q1W/Dk5ObCyskJ2djYsLS1r8tWJiCRz5IgDSkrS0bnzaVhYdJYsDrW6BEeO2EKlykPnzmdhYfGEZLEQEVH18PqXakqfzhllqRK//PsLPj3yqeaGhrmROf4b8F+82f1NOJk7SRwhERERNRWbLmzC5J2TUVBSADdrN2wfvR0dnTpq1qvUKkTGRyIpNwnOFs4IcA5A35/74mTiSfjY++DY5GOwMraS7gs0cdW9BpZsxIy9vT3kcjlSUlK0lqekpMDJqfKLXgcHB+zYsQP5+fm4desWLl++DHNzc3h4eGjaODs7o127dlrb+fj4ID4+vspYPDw8YG9vj+vXmVEkosbt3jwzlySNIyfnBFSqPBga2sPc3F/SWIiIiIgUBgpM7jQZl6ddxqYRm9DBsQPyivPw+dHP4Rbqhml/TENcVpzUYRIREVEjVqouxdt/vY2xv45FQUkB+nv0x+kpp7WSMkDZvHq93HphrN9Y9HLrBXOFOXaM3oEWFi0Qkx6DF8JfgEpddWUoahgkS8wYGRmhc+fOiIiI0CxTq9WIiIjQGkFTGWNjY7Ro0QKlpaX49ddfMXToUM26nj174sqVK1rtr169ilatWlW5v9u3b+Pu3btwdnau5bchItIPDWWemfIyZtbWfSEIkv2viIiIiEiLXCbH6PajEfVaFHaN3YVuLt2gVCnx/env4fW1FybsmICYtBipwyQiIqJGJr0gHcG/BOPLY18CAOb2nIvd43bDztSuWts7WzjjtzG/wcTABH9e+xPv7H+nPsOlOiDp3bBZs2ZhxYoVWLt2LWJiYjB16lTk5+dj0qRJAICXXnoJ8+bN07Q/ceIEwsPDcfPmTURGRiI4OBhqtRpz5szRtHnzzTdx/PhxLF68GNevX8eGDRvw448/Ytq0aQCAvLw8vP322zh+/Dji4uIQERGBoUOHwsvLC0FBQbr9AYiIdKw8MVNQIHViZj8AwNaW88sQERFRwyMIAp5r/RyOvnwUByYcQH+P/lCJKqw7vw6+3/tixJYROH3ntNRhEhERUSNwNuksAn4MQERsBMwMzbB55GZ82u9TyGXyGu2nc/POWD10NQDgy2NfYm3U2kdsQVKSNDEzevRofPnll1iwYAE6duyIqKgo7NmzB46OjgCA+Ph4JCUladoXFRVh/vz5aNeuHYYPH44WLVrg8OHDsLa21rTp0qULtm/fjo0bN6J9+/ZYtGgRQkNDMW7cOACAXC7Hv//+iyFDhqB169aYPHkyOnfujMjISCgUCp1+fyIiXWsII2ZKS7ORk1M2j5iNTT/J4iAiIiJ6FEEQ0MutF/568S+cfOUkhrcdDhEiwmPC0WVFFwT9EoRDcYcg0dStREREpOd+Pv8zeq7qiVvZt+Bp44njrxzHKN9Rtd7f6Paj8f7T7wMAXv39VRxNOPqILUgqgsgryFrRp4ksiYjKFRen4+hRBwBAYGAe5HIznceQnv4bLlwYBhMTb3TtelXnxyciotrh9S/VVGM9Zy6mXsRnRz7DhugNUIll9dt7uPbAu0+9i0HegyAIgsQREhERUUNXoirB2/veRtiJMADAQK+BWB+yHjYmNo+9b7WoxvNbn0d4TDiamTXDqSmn0NKq5WPvl6qnutfALOxPRNSEGBnZw9CwGQAgP1+a+ujlZcxsbFjGjIiIiPSPbzNfrBu+Dtdev4apAVOhkCtwNOEontv4HJ744QlsvrCZE+4SERFRlVLzU9H/5/6apMz8wPnYNXZXnSRlAEAmyLBu2Dr4O/ojNT8VQzYOQV5xXp3sm+oOEzNERE2MmVk7ANLNM5ORsQ8Ay5gRERGRfnO3ccf3z36P2BmxeLvH2zA3Msf5lPMY8+sY+Hzng5VnV6JYVSx1mERERNSAnEo8hc4/dsahW4dgbmSO8FHhWNRnUY3nk3kUMyMz/DbmNzQza4bzKecxYccEqEV1nR6DHg8TM0RETYypqXTzzBQVJaCw8AoAGayte+v8+ERERER1zdnCGZ/3/xy3Zt7Ch70+hK2JLa5lXMMru16B59ee+PrE1ygoKZA6TCIiIpLY6nOrEbg6ELdzbqONXZuy+et8htfb8VpZt8L20dthJDdCeEw4Pjz4Yb0di2qOiRkioibGzEy6xExmZtloGUvLJ2FoaK3z4xMRERHVF1sTWyx4ZgFuzbyFpQOWwtncGbdzbmPGnhloFdoKiyMXI6soS+owiYiISMeKVcWY9sc0vLzzZShVSgxpMwQnXjkBHwefej92D9ce+OG5HwAAH/3zETZf2Fzvx6TqYWKGiKiJkTYxUz6/DMuYERERUeNkbmSOWd1n4eaMm/jhuR/gYeOB9IJ0vPf3e2gV2grvRryL1PxUqcMkIiIiHUjOS0aftX3w/envAQAf9voQ20dvh5Wxlc5imNhxIt7q/lbZ+98m4sydMzo7NlWNiRkioiamPDGjVN5CaanuJn8TRfV9iZn+OjsuERERkRSMDYzxaudXcWX6Ffwy/Bf4OvgiR5mDJYeXoFVoK7yx+w3EZ8dLHSYRERHVk2MJx9Dph044knAElgpL7Bq7CwueWQCZoPtb8p/1+wwDvQaiqLQIQzcNRVJuks5jIG1MzBARNTGGhnYwNHQEABQUXNLZcfPy/kVJSRpkMjNYWnbT2XGJiIiIpGQgM8C4DuPw79R/sWP0DjzZ4kkUlRbhm5PfwPNrT7z828u4kn5F6jCJiIioDv145kc8s+YZJOUloZ1DO5yacgrPtX5OsnjkMjk2jtgIH3sfJOYmYtjmYSgsKZQsHmJihoioSZKinFn5aBlr616QyYx0dlwiIiKihkAmyDC07VAcn3wc+1/cjz7ufVCqLsXqqNXw+c4Ho7aOwrmkc1KHSURERI9BWarEq7texWu/v4YSdQlCfEJwfPJxtLZrLXVosDK2ws6xO2FjbIOTiScxZdcUiKIodVhNFhMzRERNkDSJmX0AOL8MERERNW2CIKCvR19EvBSBY5OPYUibIRAhYuulrej0YycMWj8Ih+MPSx0mERER1VBiTiKeWfMMVpxdAQECFvdZjG3Pb4OFwkLq0DS8bL2wbdQ2yAU51kevx+dHPpc6pCaLiRkioiaoPDGjq1JmKlURsrP/AQDY2nJ+GSIiIiIA6ObSDb+N+Q3//udfvOD3AmSCDLuv70bg6kA8vfpp7Lm+h0+yEhER6YHIW5Ho/GNnnEg8AWtja/w57k/MC5wHQRCkDq2CPu598M3AbwAA8yLmYeeVnRJH1DQxMUNE1ASZmup2xExOzlGo1UUwMnKGqWk7nRyTiIiISF/4Ofphfch6XJ1+Fa92ehVGciNExkdi4PqBCFgRgF8v/Qq1qJY6TCIiInqAKIr47uR36LOuD1LyU+DXzA+np5xGsFew1KE91NQuUzE1YCpEiBgXPg7RKdFSh9TkMDFDRNQElY+YUSrjUVqaW+/Hu7+MWUN8WoSIiIioIfC09cQPg3/AzTduYla3WTA1NMXZpLMYuXUkfL/3xdqotShRlUgdJhEREQEoKi3CyztfxvTd01GqLsVo39E4NvkYPG09pQ6tWsKCw9DbrTfyivMwZNMQpOWnSR1Sk8LEDBFRE2RoaAMjI2cAuilnlpFRnphhGTMiIiKiR2lh2QJLg5bi1sxbWPD0AlgbW+Ny+mVM/G0ivL7xwncnv0NhSaHUYRIRETVZ8dnxCFwdiDVRayATZPii/xfYOGIjzIzMpA6t2gzlhtj6/FZ42ngiLisOI7eORLGqWOqwmgwmZoiImqjyUTP1Xc6spOQu8vLOAigbMUNERERE1WNvao8Pe3+IWzNv4fN+n8PRzBHx2fGYvns63MLc8Nnhz5CjzJE6TCIioiblYNxBdP6xM07fOQ07EzvsHb8Xs3vM1ssKIXamdtg5dicsjCzwz61/MO2PaZzfTkeYmCEiaqJ0Nc9MZubfAESYmvpCoXCu12MRERERNUaWCku83fNtxM6IxfeDvkcrq1ZIzU/FOxHvoOX/tcT7f7+P9IJ0qcMkIiJq1ERRROjxUPRb1w/pBeno6NQRp189jX4e+v0QajuHdtg0chMECPjp3E/45uQ3UofUJDAxQ0TURJmZtQOgi8RMWRkzW1uWMSMiIiJ6HCaGJpjaZSquvX4N64atg4+9D7KV2fg48mO0Cm2FN/e8ids5t6UOk4iIqNEpKCnA+O3j8ebeN6ESVRjfYTyOvHwEbtZuUodWJwZ5D8IX/b8AALy59038deMviSNq/JiYISJqospLmRUU1HdiZj8Azi9DREREVFcM5YZ40f9FXPjvBfw66ld0du6MgpIChJ4IhUeYB6bsnILrGdelDpOIiKhRiM2MRc9VPbEhegPkghxhwWFYN2wdTA1NpQ6tTs3qPgsTO06EWlRj1NZRuJJ+ReqQGjUmZoiImqjyUmZK5W2UlmbXyzEKC2+gqCgWgmAIK6un6+UYRERERE2VTJAhxCcEp6acwt7xe/FMq2dQoi7BT+d+Qptv22Dsr2Pxb8q/FbZTqVU4GHcQG6M34mDcQajUKgmiJyIiavj23diHgBUBiEqOgoOpA/a/tB9vdH1DL+eTeRRBELD82eXo4doD2cpsDNk0BJmFmVKH1WgxMUNE1EQZGlrDyKg5ACA//1K9HCMjo6yMmaVldxgYmNfLMYiIiIiaOkEQMMBzAA5OPIjDkw7jWe9noRbV2HRhE/yX+2PwxsE4lnAMABAeEw63MDf0XtsbL4S/gN5re8MtzA3hMeESfwsiIqKGQxRFfHHkCwSvD0ZGYQYCmgfgzKtn0Mutl9Sh1SuFgQLho8LhaumKq3evYsyvY1CqLpU6rEaJiRkioiasvJxZfc0zwzJmRERERLrVs2VP/P7C7zj32jmM9h0NAQJ+v/o7eqzqgfbft8eILSMqzEOTmJOIkVtGMjlDREQEIL84H2N+HYM5++dALaoxqeMkRE6KhKuVq9Sh6YSjuSN2jt0JU0NT/HXjL8z+a7bUITVKTMwQETVh9TnPjCiqkJX1NwDAxqZfne+fiIiIiKrW0akjNo3chMvTL2PyE5NhIBjgYlrl13wiRADAzD0zWdaMiIiatBsZN9B9ZXdsubgFBjIDfD/oe6wcshLGBsZSh6ZTHZ064ufhPwMAwk6EYcWZFRJH1PgwMUNE1ISVzzNTHyNmcnPPoLQ0E3K5FSwsAup8/0RERET0aK3tWuOnIT9hfcj6h7YTISIhJwEf//MxziefR35xvo4iJCIiahj2XN+DgBUBiE6NhqOZIw5MOICpXaY2yvlkqiPEJwQf9foIAPDfP/+Lf279I3FEjYuB1AEQEZF06rOUWWZm2fwyNjZ9IJPxfzdEREREUlKJ1RsJs/DQQiw8tBAA0NyiObxtvctedt5obdca3rbe8LT1bHJPDhMRUeMliiKWHF6C+X/PhwgR3Vy6Ydvz29DCsoXUoUlu/tPzcTHtIjZf3IwRW0bg5Csn4W7jLnVYjQLvlBERNWFmZu0AAMXFd1BSkgVDQ+s62/e9+WVYxoyIiIhIas4WztVq52Pvg5T8FGQUZuBO7h3cyb2DQ7cOabURIMDVylUraVP+p4eNB4zkRvXxFYiIiOpcrjIXE3+bqJln7bXOryEsOAwKA4XEkTUMgiBg1dBVuJ5xHWeSzmDIpiE4+vJRWCgspA5N7zExQ0TUhBkYWEGhcIFSeRsFBZdgZdWjTvarUuUjO/sIAMDGpn+d7JOIiIiIai+wZSBcLF2QmJOomVPmfgIEuFi6IHpqNOQyOTIKM3Dt7jVcy7h278//vc9WZiM+Ox7x2fGIiI3Q2o9MkMHN2q3SpI2btRsMOJKaiIgaiCvpVzB883DEpMfASG6Ebwd+iymdp0gdVoNjamiKHWN2oMuKLriQegHjt4/H9tHbIRM4S8rj4BUREVETZ2rqC6XyNvLzL9ZZYiYr6x+IYgkUilYwMfGqk30SERERUe3JZXKEBYdh5JaRECBoJWcElNXODw0OhVwmBwDYmtiiq0tXdHXpqrUfURSRVpBWZdImvyQfNzNv4mbmTey9sVdrWwOZATxsPCpN2rhaumqOTUREVN92XdmF8dvHI0eZg+YWzfHrqF/RzaWb1GE1WC6WLtgxegeeWfMMdl7Zifl/z8fivoulDkuvMTFDRNTEmZn5IjNzb53OM1NexszWtn+TnSSPiIiIqKEJ8QnBtlHbMGPPDNzOua1Z7mLpgtDgUIT4hDxyH4IgoJlZMzQza4aeLXtqrRNFEcl5ybiWcQ1X717VStpcz7iOotIiXL17FVfvXq2wX4VcAQ8bD808NvcnbZpbNOdTuUREVCfUohqLDi3SzKf2VMunsPX5rXAyd5I2MD3Q1aUrVg5ZifHbx2PJ4SXwdfDFuA7jpA5LbzExQ0TUxJmZ+QIACgrqMjGzDwDnlyEiIiJqaEJ8QjC0zVBExkciKTcJzhbOCGwZWCejVQRBgLOFM5wtnPF0q6e11qlFNRJzErVG2Vy9exXXMq7hZuZNKFVKxKTHICY9psJ+TQxM4GXrBW87b7S2ba2VtHE0c6zTB4FUalW9/DZERCS97KJsvLj9Rey6ugsAMK3LNHwV9BXnRquBcR3G4ULqBXx65FNM3jkZXrZeFUbXUvUwMUNE1MSVJ2bqasSMUpmM/PxoAAKsrfvWyT6JiIiIqO7IZXL0cuul02PKBBlcrVzhauWKPu59tNap1CrEZ8dXKI129e5VxGbGorC0ENGp0YhOja6wXwsjC03SxtvWW2vEjZ2JXY2SNuEx4ZWOJgoLDqvWaCIiImq4LqVdwvDNw3H17lUo5Aosf245JnacKHVYeumTvp/gYtpF7Lq6C8M2D8OpKafgYukidVh6RxBFseKsf/RIOTk5sLKyQnZ2NiwtLaUOh4io1kpLc3D4sBUAoGfPDBga2jzW/lJS1iMmZjzMzTshIOBMXYRIREQNAK9/qaZ4zlBdKFGVIC4rrtL5bG5l34JaVFe5rbWxtXZZtPLEjZ03rI2ttdqGx4Rj5JaRWnPvAPfm39k2ahuTM0REeio8JhwTdkxAXnEeXC1dET46HAHNA6QOS6/lKnPRY1UPXEi9gM7OnfHPpH9gamgqdVgNQnWvgTlihoioiTMwsIRC4QqlMgH5+Rdhbf3UY+0vI4NlzIiIiIiobhjKDcsSK3begLf2OmWpEjczb1aatEnISUBWURZO3TmFU3dOVdivvam9JmnjaeOJsBNhFZIyACBChAABM/fMxNA2Q1nWjIhIj6jUKnxw8AN8EvkJAKCXWy9sHrkZzcyaSRyZ/rNQWGDnmJ148qcncSbpDCb9NgmbRmziPMM1wMQMERHBzMwXSmUCCgoeLzEjiuJ988v0r6vwiIiIiIgqUBgo4OPgAx8HnwrrCksKcSPzRtk8Ng8kbZLykpBekI70gnQcu33skccRISIhJwGR8ZE6LwFHRES1k1mYiXHh47D7+m4AwJvd3sTn/T+HgYy3w+uKu407fh31K/qu64stF7egvUN7vP/M+1KHpTd4JhIREUxNfZGRseex55kpKLiM4uI7kMmMYWX1eCNviIiIiIhqy8TQBO2btUf7Zu0rrMsrzsP1jOuahM2e63sQGR/5yH3uurILnZw7wVLB0nxERA3ZhdQLGLZpGG5k3oCJgQlWDF6BcR3GSR1Wo/R0q6ex7NllmLJrChYcXADfZr4s/VlNTMwQERHMzHwB4LETM+WjZaysnoJcbvzYcRERERER1TVzI3N0dOqIjk4dAQA9XHug99rej9zuq+NfIexEGLq7dscAjwEI8gpCZ+fOLG9GRNSAbLm4BZN+m4SCkgK4Wbth++jtmv/eU/14pdMriE6Jxtcnv8aL21+Eh40Hf/NqkEkdABERSa+uEzMsY0ZERERE+iKwZSBcLF0goOq6+OZG5vC29YZKVOFw/GEsOLgAXX/qCocvHDBq6yj8dPYnJGQn6DBqIiK6X6m6FHP3zcXobaNRUFKAfh79cHrKaSYIdGRp0FL09+iPgpICDNk4BCl5KVKH1OAxMUNERDA1bQcAKClJQUnJ3VrtQ60uQVbWQQBMzBARERGR/pDL5AgLDgOACskZ4X//rB22Fldfv4qbb9zED8/9gBE+I2ClsEJmUSa2XtqKKbumoGVoS/h854MZu2fgz2t/Ir84X4qvQ0TU5NwtuIuB6wfi86OfAwDm9JiD3eN2w87UTuLImg4DmQE2j9yM1natkZCTgJAtIVCWKqUOq0ETRFEUpQ5CH+Xk5MDKygrZ2dmwtGR9WSLSf8eOuUGpvIWOHf+BtXVgjbfPyjqMqKhAGBrao0ePFAgCc/9ERI0Jr3+ppnjOkL4JjwnHjD0zcDvntmaZq6UrQoNDK62XX6ouxanEU/jrxl/46+ZfOH77ONSiWrPeUGaIp1o+hQGeAxDkGQR/J3/IeI1MRFSnopKjMHzzcMRlxcHU0BSrh67GKN9RUofVZF29exVdf+qKrKIsTOw4EauGrIIgVD0itTGq7jUwEzO1xE4GETU2//77LDIy/oS39zK0aPGfGm8fG/sBbt36CA4Oo+Hru6keIiQiIinx+pdqiucM6SOVWoXI+Egk5SbB2cIZgS0Dqz2HTFZRFv6O/Rt/3fgLe2/sRVxWnNZ6B1MH9PfsjyDPIPT36A9nC+d6+AZERE3HhugNeGXnKygsLYSnjSe2j94OP0c/qcNq8v668RcGrh8ItajGl/2/xFs93pI6JJ2q7jWwgQ5jIiKiBszMzBcZGX+ioKB288xkZu4HANjasowZEREREeknuUyOXm69arWttbE1QnxCEOITAlEUcT3juiZJcyDuANIK0rAhegM2RG8AAPg189OMpnmq5VMwMTSpw29CuvQ4CT0iqrlSdSnm7JuD/zv+fwCAgV4DsT5kPWxMbCSOjABggOcA/F/Q/2HGnhmYs38OfBx8MMh7kNRhNThMzBAREYCyxAwA5OfXPDFTWpqNnJwTAAAbm351GhcRERERkb4RBAHedt7wtvPGtCenoVhVjOO3j2sSNWfunEF0ajSiU6Ox9NhSGBsY4+lWTyPIMwgDPAfA18G3yZV+0VeVlcBzsXRBWHBYpSXwiOjxpOWnYdS2UTgYdxAA8F7ge/iw14dMhjYwrz/5Oi6kXsCKsysw9texODb5GNo5tJM6rAaFpcxqicPyiaixyck5jbNnu8DQsBl69kyp0bbp6b/hwoVhMDHxRteuV+spQiIikhKvf6mmeM4QVS29IB0RNyOw98Ze/HXjLyTmJmqtdzZ31oym6efRDw5mDhJFSg8THhOOkVtGQoT2rTUBZUm1baO2NfnkDEcTUW1Vdu6UzyeTkJMAcyNzrB22tsn/HWvIilXF6P9zf/xz6x942njixCsnYGdqJ3VY9Y6lzIiIqEbMzHwAACUlqSguToeRkX21ty0vY2ZjwzJmRERERESPYm9qj9HtR2N0+9EQRREx6THYe30v/rr5Fw7FHUJSXhLWnl+LtefXAgA6OXfCAI8BCPIKQg/XHjCSG0n8DahUVYrX/3y9QlIGgGbZlF1ToFarYaGwgKmhKUwMTWBqaKr1MjEwabSJCo4motqq7NyxMbZBbnEuStWlaG3XGttHb+cIjAbOSG6Ebc9vw5M/PYkbmTfw/NbnsXf8XhjKDaUOrUHgiJla4tNfRNQYHT/ujqKiOHTseBDW1s9Ue7sTJ9qisPAKfH3D4eAwvB4jJCIiqfD6l2qK5wxR7RSVFuFI/BHNaJrzKee11psZmqGXWy9N2bPWdq1Z9qye5BfnIzYrFrGZsYjNisXNzJuaz9cyrqGotKhOjqOQK6pM3NyfwKnNuvL96joBxNFEVFtVnTvlApwDsP+l/bAyttJxZFRb0SnR6LGqB/KK8/DfgP/iu2e/kzqkelXda2AmZmqJnQwiaoz+/fc5ZGT8AW/v79CixX+rtU1RUQKOH28JQIaePe/C0NC6XmMkIiJp8PqXaornDFHdSM5Lxr4b+/DXzb/w142/kJqfqrW+pVVLzWiavu59Ofl1DZSoShCfHa+VcInNitW8TytIe+xjeNt6w8zIDAUlBSgsKURBSUHZ+9LCOvgGNVOeAHowaVMhmWPwkHWPSAiZGJpAFEW4hblpjXa4nwABLpYuiJ0R22hHC1HtqNSqh547AOBq6cpzRw/tvLITwzYNgwgR3w36Dv/tUr17TvqIpcyIiKjGzMx8kZHxB/LzL1Z7m8zMfQAAS8snmZQhIiIiIqpjTuZOeNH/Rbzo/yLUohrRKdGa0TSR8ZGIz47HT+d+wk/nfoJMkKFL8y6a0TRdXbrCQNZ0b/2oRTWScpMqTbrEZsXids5tqEX1Q/dhbWwNDxsPuFu7l71syv5MK0jDhB0THhnDj4N/RC+3XpXGVlRadC9Rc1/S5sFXYWkt15UUaiWAlCollColMosya/xb1oShzBAl6pIq14sQkZCTgH9u/YPe7r3rNRbSL79f/f2hSRkASMhJQGR8ZKV/r6jhGtJmCBb3XYx5EfPwxu430Na+Lfq495E6LEk13f87ExFRBWZmvgBQw8QM55chIiIiItIFmSCDv5M//J38MafnHBSUFOCfW/9o5qe5lHYJJxJP4ETiCXz0z0ewVFiij3sfTaLGw8ZD6q9Qp0RRRGZR5r2kywPJl7isOChVyofuw9jAWCvhovXexh3WxtaVbqdSq/De3+8hMSex0pJL5aNCAlsGVrq9TJBpRpnUpwcTQPWVCLq/rNvDkjL3G7xxMDo5d0I7h3Zo59AOvg6+aOfQDk7mTizP14iJooi0gjRcTL2IS2mXyl7pZX8+OCKwKkm5SfUcJdWHuT3n4kLqBayPXo+RW0bi5JST8LL1kjosyTAxQ0REGuWJmYKC6iVmRFF9X2KmX73FRUREREREFZkamiLYKxjBXsEAgNs5t/HXjbKSZ/tu7kNGYQZ2XN6BHZd3AAA8bTwxwHMAgjyD0Nu9NywVDb/MYEFJAeKy4iokX8rLj+Uocx66vVyQw9XKtdKki7u1e62TAHKZHGHBYRi5ZSQECFrJmfJ5VEKDQyUvtyRFAijiZgTG/Drmkdvkl+QjMj4SkfGRWsttjG0qJGvaObRDc4vmTNjoEVEUkZyXjEtpl3Ax7b4kTNol3C28+1j7drZwrqMoSZcEQcBPQ37CtYxrOJl4EkM2DsGxycea7HxBnGOmllgvmYgaI5WqAJGR5gBE9OiRAiOjZg9tn5sbhTNnnoBMZoannsqATGakm0CJiEjneP1LNcVzhkhaKrUKZ5POliVqbv6FowlHUaou1ayXC3J0d+2uGU3T2bnzQ5MIKrUKkfGRSMpNgrOFMwJbBtZJ0qFUXYqE7ARN0kUz38v/PqfkpzxyH45mjhVGvJSXH3OxdIGh3PCx46xKeEw4ZuyZoVV+ydXSFaHBoU12cvvyeUIeNpqohWUL7Bi9A5fTL2tGTFxMvYgbmTeqLC9npbCqkKzxbeaLFhYtmLCRkCiKSMxNLEvAlI+C+d8ImKyirEq3ESDAw8ZD8++x/N+rt603fJf5PnIkGueY0W9JuUnosqILEnMTMdBrIHaN3dWo/n1W9xqYiZlaYieDiBqr48c9UVR0E/7+f8PG5uH1fuPjv8TNm2/D1vZZdOjwu44iJCIiKfD6l2qK5wxRw5KrzMWBuAOaETXXMq5prbcxtkE/j36aRI2rlatmXWXJBxdLF4QFhz0y+SCKIlLyU8oSLpWUHEvIToBKVD10H5YKS61RLpo5X2zc4WbtVu+jQR6lvpJW+iw8Jhwjt4wEgEpHE20bta3Sc6eotAhX0q9oRlaUj7S4nnG9yvPEUmFZdnPf/l6ypp1DO7haujJhU4fUohoJ2QmVjoDJLc6tdBuZIIOXrZfWv592Du3Q1r4tTAxNKt2mtucO6Zczd84gcHUgCksLMbv7bHwx4AupQ6ozTMzUM3YyiKixio4egrt3d8HL6xu4uEx/aNvz54OQmfkXPD3/D66uM3UTIFEDoFKpUFJSvdrZRPrE0NAQcnnlN5J4/Us1xXOGqGGLzYzVjKaJuBmBbGW21vq29m0R5BkEMyMzLIlcUuHp9ftvkvZx7/PQeV7un4C+MkZyI7hZu91LuDxQcszG2IY32PVQXY4mUpYqcfXu1QpJgat3r1aZsDE3Mq+0JFpLq5aQCbLH+m6NmVpUIy4r7t4ImP+NfolJi0F+SX6l2xjIDOBt6601AqadQzu0tmsNYwPjGsfAkWhNw5aLWzB622gAwOqhqzGx40RpA6ojTMzUM3YyiKixunlzHuLjP0Xz5lPRuvX3VbZTqYpw5IgN1OoidOlyQTM/DVFjJooikpOTkZWVJXUoRPXG2toaTk4V6+3z+pdqiucMkf4oVZfiZOJJzWiaE4knqiwn9aAH51epqo1mnpcHSo65W7vD2cKZN8obqfoeTVSsKsa1u9e0kjUX0y7i6t2rWqX77mdmaAYfBx/tkmgOvmhl3apJnYcqtQo3M29WSHZdTr9cZTLVUGaINvZtKoyA8bbzhpG8bkubcyRa07DgwAIs+mcRjORG+Pulv9GzZU+pQ3psTMzUM3YyiKixSk7+BZcvvwgrq6fxxBOHqmyXmfk3zp/vCyMjZ3Tvnsgn2KhJSEpKQlZWFpo1awZTU1Oe99SoiKKIgoICpKamwtraGs7O2pOq8vqXaornDJH+yizMxN+xf2NN1Br8fq16JYsdTB0qTbq427ijpVXLOr9pS/QwJaoSXMu4VqEk2pX0KyhRVz7y3cTARCthU/6nm7WbXicESlQluJF5o9LfQqlSVrqNkdwIbe3baiWv2jm0g6eNZ73O2URNj1pU4/mtzyM8JhzNzJrh1JRTaGnVUuqwHkt1r4ENdBgTERHpgfKRL/n5FyGKYpU3njMz9wEAbGz68eY0NQkqlUqTlLGzs5M6HKJ6YWJSVus7NTUVzZo1q7KsGRERNW42JjYY0W4EilXF1UrMrBqyCpOemKSDyIiqx1BuqEkm3O/+JMX9ZbrKR4mcTTqLs0lntbYxMTBBW/u2FUqiedh4PFbCpq5HhJSPHnowAXP17tUqk1HGBsbwsfcpm5fnvhEw7jbuMJDxtjHVP5kgw7ph63Aj4wbOp5zHkI1DcPjlwzA3Mpc6tHrHv2FERKTF1LQtABlKS++ipCQVRkaOlbbLyChPzPTXYXRE0imfU8bUVNrJZYnqW/k5XlJSwsQMEVET52zh/OhGANxt3Os5EqK6YSg3RFv7tmhr31ZrrpJSdSluZt4sS9akXcKl9LLETXnC5lzyOZxLPqe1L4VcUSFh49vMFx42Ho9MalQ2h4qLpQvCgsMeOYdKUWmRZr6d+xMw1+5eq3K+nfLybQ8ml1pZtdLr0UDUOJgZmWHn2J3osqILzqecx4QdE7D1+a2NvrQgEzNERKRFLjeBiYkHCguvIz//YqWJmZKSu8jLK3uKyMamn65DJJIUR4hRY8dznIiIygW2DISLpQsScxIrnUdGgAAXSxcEtgyUIDqiumMgM0Bru9Zobdcaw32Ga5bfPw/L/UmQmPQYFJUW4XzKeZxPOa+1LyO5EdrYtdGMQvFt5qtVBiw8Jhwjt4ys8HcqMScRI7eMxLZR2xDiE4KCkgJcSb9S4dg3Mm9UOQeUhZGF1rw55e9drVwb/U1u0m8trVpi++jt6L22N8JjwrHw4EJ81PsjqcOqV0zMEBFRBaamvprEjI1NnwrrMzP/BiDC1NQXCkX1nqIjIiIiIiL9IpfJERYchpFbRkKAoHUjWUBZIj80OJRP3FOjJZfJ4W3nDW87bwxtO1SzXKVWIS4rTithcjHtImLSYlBYWojo1GhEp0Zr7ctQZojWdq1xM/NmpYnO8mXjwsfB2dwZcVlxlbYDACuFlVb5sfLkTwuLFnzIhvRWD9ce+OG5HzDpt0lY9M8i+Dr4YnT70VKHVW+YmCEiogrMzNrh7t3fkJ9/sdL15fPL2NqyjBlRTdV1LWkpuLm5YebMmZg5c2a12h88eBC9e/dGZmYmrK2t6zU2IiIiqlshPiHYNmpbpWWXQoNDH1l2iagxksvk8LT1hKetJwa3GaxZrhbVuJV1S5OsuT9xU1BSgItplfex71dUWoTYrFgAgK2JrdbIl/L3TuZOTMBQozSx40RcTL2IL499iYm/TYSnrScCmgdIHVa9YGKGiIgqMDPzBQAUFFSVmNkPgPPLENXU49SSro1HddY++OADLFy4sMb7PXXqFMzMzKrdvkePHkhKSoKVlVWNj1Vbbdu2RWxsLG7dugUnJyedHZeIiKgxCvEJwdA2Q/X+4RKi+iYTZHC3cYe7jTuea/2cZrlaVCM+Ox7fn/oeXxz94pH7eS/wPbzR9Q04mDowAUNNzqf9PsWl9Ev489qfGLZpGE5NOVXtOc/0CYsLEhFRBaamZYmZ/PyLEEXtodOFhTdQVBQLQTCEldXTUoRHpJfKa0nfn5QB7tWSDo8Jr/NjJiUlaV6hoaGwtLTUWjZ79mxNW1EUUVpaWq39Ojg4aCaIrw4jIyM4Oenuqb7Dhw+jsLAQI0eOxNq1a3VyzIcpKSmROgQiIqLHJpfJ0cutF8b6jUUvt15MyhDVgEyQwc3aDYO8B1WrfT+Pfmhm1oxJGWqS5DI5NoRsgI+9DxJzEzFs8zAUlhRKHVadY2KGiIgqMDVtC0CG0tJMFBcna63LyCgrY2Zp2R0GBuYSREfUMIiiiPzi/Gq9copy8MbuNx5aS3rG7hnIKcqp1v4eTJhWxcnJSfOysrKCIAiaz5cvX4aFhQV2796Nzp07Q6FQ4PDhw7hx4waGDh0KR0dHmJubo0uXLti/f7/Wft3c3BAaGqr5LAgCfvrpJwwfPhympqbw9vbGzp07NesPHjwIQRCQlZUFAFizZg2sra2xd+9e+Pj4wNzcHMHBwUhKStJsU1paijfeeAPW1taws7PD3LlzMWHCBAwbNuyR33vlypV44YUX8OKLL2LVqlUV1t++fRtjx46Fra0tzMzMEBAQgBMnTmjW79q1C126dIGxsTHs7e0xfPi9CWAFQcCOHTu09mdtbY01a9YAAOLi4iAIAjZv3oxnnnkGxsbGWL9+Pe7evYuxY8eiRYsWMDU1hZ+fHzZu3Ki1H7Vajc8//xxeXl5QKBRo2bIlPvnkEwBAnz59MH36dK32aWlpMDIyQkRExCN/EyIiIiKSXmDLQLhYumjmaHqQAAGulq4IbBmo48iIGhYrYyvsHLsTtia2OJl4ElN2Tal2P1hfSJ6Y+e677+Dm5gZjY2N07doVJ0+erLJtSUkJPvroI3h6esLY2Bj+/v7Ys2dPhXaJiYkYP3487OzsYGJiAj8/P5w+fVqzXhRFLFiwAM7OzjAxMUG/fv1w7dq1evl+RET6SC43homJJwBUmGeGZcyIyhSUFMB8iXm1XlafWSExN7HKfYkQcTv3Nqw+s6rW/gpKCurse7zzzjv49NNPERMTgw4dOiAvLw+DBg1CREQEzp07h+DgYAwePBjx8fEP3c+HH36IUaNG4d9//8WgQYMwbtw4ZGRkVNm+oKAAX375JX7++Wf8888/iI+P1xrB89lnn2H9+vVYvXo1jhw5gpycnAoJkcrk5uZi69atGD9+PPr374/s7GxERkZq1ufl5eGZZ55BYmIidu7cifPnz2POnDlQq9UAgD/++APDhw/HoEGDcO7cOURERODJJ5985HEf9M4772DGjBmIiYlBUFAQioqK0LlzZ/zxxx+4cOECXn31Vbz44ota177z5s3Dp59+ivfffx+XLl3Chg0b4OjoCAB45ZVXsGHDBiiVSk37X375BS1atECfPn1qHB8RERER6Z5cJkdYcBgAVEjOlH8ODQ7liDQiAF62Xtj6/FbIBTnWR6/HZ0c+kzqkOiVpYmbz5s2YNWsWPvjgA5w9exb+/v4ICgpCampqpe3nz5+PH374Ad988w0uXbqE//znPxg+fDjOnTunaZOZmYmePXvC0NAQu3fvxqVLl7B06VLY2Nho2nz++ef4+uuvsXz5cpw4cQJmZmaaDjMREZWpbJ4ZUVQhK+tvAICNTT9J4iKiuvXRRx+hf//+8PT0hK2tLfz9/fHaa6+hffv28Pb2xqJFi+Dp6ak1AqYyEydOxNixY+Hl5YXFixcjLy/vkQ/cLF++HAEBAejUqROmT5+uNfLjm2++wbx58zB8+HC0bdsW3377LaytrR/5fTZt2gRvb2/4+vpCLpdjzJgxWLlypWb9hg0bkJaWhh07duCpp56Cl5cXRo0ahe7duwMAPvnkE4wZMwYffvghfHx84O/vj3nz5j3yuA+aOXMmQkJC4O7uDmdnZ7Ro0QKzZ89Gx44d4eHhgddffx3BwcHYsmULgLKEUlhYGD7//HNMmDABnp6eeOqpp/DKK68AAEJCyuYg+u233zTHWLNmDSZOnMgSF0RERER6JMQnBNtGbUMLyxZay10sXbBt1LZ6mXuSSF/1ce+DbwZ+AwB4N+Jd7Lzy8H6pPjGQ8uBfffUVpkyZgkmTJgEAli9fjj/++AOrVq3CO++8U6H9zz//jPfeew+DBpXVY5w6dSr279+PpUuX4pdffgFQ9nSlq6srVq9erdnO3d1d814URYSGhmL+/PkYOnQoAGDdunVwdHTEjh07MGbMmHr7vkRE+qRsnpkdWiNmcnPPoLQ0E3K5FSwsAqQLjqgBMDU0Rd68vGq1/efWPxi04dH1pP984U883erRczeZGlZ/fpdHCQjQ/rucl5eHhQsX4o8//kBSUhJKS0tRWFj4yBEzHTp00Lw3MzODpaVllQ/bAICpqSk8PT01n52dnTXts7OzkZKSojVSRS6Xo3PnzpqRLVVZtWoVxo8fr/k8fvx4PPPMM/jmm29gYWGBqKgoPPHEE7C1ta10+6ioKEyZMuWhx6iOB39XlUqFxYsXY8uWLUhMTERxcTGUSqVmrp6YmBgolUr07du30v0ZGxtrSrONGjUKZ8+exYULFx6ZMCMiIiKihifEJwRD2wxFZHwkknKT4GzhjMCWgRwpQ1SJqV2m4kLqBXx/+nuMCx+Hoy8fhZ+jn9RhPTbJRswUFxfjzJkz6Nfv3hPXMpkM/fr1w7FjxyrdRqlUwtjYWGuZiYkJDh8+rPm8c+dOBAQE4Pnnn0ezZs3wxBNPYMWKFZr1sbGxSE5O1jqulZUVunbtWuVxy4+dk5Oj9SIiaszKR8zk51/SLLtXxqwPZDJJc/tEkhMEAWZGZtV6DfAcUK1a0gM8B1Rrf3U5QsLMzEzr8+zZs7F9+3YsXrwYkZGRiIqKgp+fH4qLix+6H0NDQ+3vJAgPTaJU1v5xawZfunQJx48fx5w5c2BgYAADAwN069YNBQUF2LRpE4Cya8eHedT6yuIsKSmp0O7B3/WLL75AWFgY5s6diwMHDiAqKgpBQUGa3/VRxwXKypnt27cPt2/fxurVq9GnTx+0atXqkdsRERERUcMjl8nRy60XxvqNRS+3XkzKED1EaHAoerv1Rl5xHoZsGoK0/DSpQ3pskiVm0tPToVKpNHWzyzk6OiI5ObnSbYKCgvDVV1/h2rVrUKvV2LdvH8LDw7Umir158yaWLVsGb29v7N27F1OnTsUbb7yBtWvXAoBm3zU5LgAsWbIEVlZWmperq2utvjcRkb64v5RZ+U3IzMx9ADi/DFFN6VMt6SNHjmDixIkYPnw4/Pz84OTkhLi4OJ3GYGVlBUdHR5w6dUqzTKVS4ezZsw/dbuXKlXj66adx/vx5REVFaV6zZs3SlDPr0KEDoqKiqpz/pkOHDlol1R7k4OCgde157do1FBQ8es6fI0eOYOjQoRg/fjz8/f3h4eGBq1evatZ7e3vDxMTkocf28/NDQEAAVqxYgQ0bNuDll19+5HGJiIiIiIj0naHcEFuf3wpPG0/EZcVhxJYRKFY9/OHBhk7SOWZqKiwsDN7e3mjbti2MjIwwffp0TJo0CTLZva+hVqvRqVMnLF68GE888QReffVVTJkyBcuXL3+sY8+bNw/Z2dmaV0JCwuN+HSKiBs3UtA0AOUpLs1BcnASVKh/Z2UcAcH4ZotrQl1rS3t7eCA8PR1RUFM6fP48XXnjhkeXD6sPrr7+OJUuW4LfffsOVK1cwY8YMZGZmVjlaqKSkBD///DPGjh2L9u3ba71eeeUVnDhxAhcvXsTYsWPh5OSEYcOG4ciRI7h58yZ+/fVXzcjpDz74ABs3bsQHH3yAmJgYREdH47PP7k0y2adPH3z77bc4d+4cTp8+jf/85z8VRv9UxtvbG/v27cPRo0cRExOD1157DSkpKZr1xsbGmDt3LubMmYN169bhxo0bOH78uNb8OEDZqJlPP/0Uoihi+PDhtflpiYiIiIiI9I6dqR12jd0FS4UlIuMjMe2PaY9ddUFKkiVm7O3tIZfLtTqkAJCSkgInJ6dKt3FwcMCOHTuQn5+PW7du4fLlyzA3N4eHh4emjbOzM9q1a6e1nY+Pj6Yuevm+a3JcAFAoFLC0tNR6ERE1ZjKZAiYmXgCA/PyLyMr6B6JYAoWilWY5EdVMiE8I4mbE4cCEA9gQsgEHJhxA7IzYBpOUAcrmALSxsUGPHj0wePBgBAUFoVOnTjqPY+7cuRg7dixeeukldO/eHebm5ggKCqpQ1rbczp07cffu3UqTFT4+PvDx8cHKlSthZGSEv/76C82aNcOgQYPg5+eHTz/9FHJ52WilXr16YevWrdi5cyc6duyIPn364OTJk5p9LV26FK6urggMDMQLL7yA2bNna+aJeZj58+ejU6dOCAoKQq9evTTJofu9//77eOutt7BgwQL4+Phg9OjRFebpGTt2LAwMDDB27NgqfwsiIiIiIqLGyMfBB5tGbIJMkOGncz/hm5PfSB1SrQmihGmlrl274sknn8Q335T9gGq1Gi1btsT06dPxzjvvPHL7kpIS+Pj4YNSoUVi8eDEA4IUXXkBCQgIiIyM17d58802cOHECR48ehSiKaN68OWbPno233noLAJCTk4NmzZphzZo1GDNmTLViz8nJgZWVFbKzs5mkIaJG68KFEUhPD4en5/9BqUzA7dtfwdn5FbRps+LRGxM1MkVFRYiNjYW7uztviEtArVZrrvsWLVokdTiSiYuLg6enJ06dOlVvCbOqznVe/1JN8ZwhIiIiovqw9OhSzN43GzJBht3jdmOA5wCpQ9Ko7jWwpKXMZs2ahRUrVmDt2rWIiYnB1KlTkZ+fj0mTJgEAXnrpJcybN0/T/sSJEwgPD8fNmzcRGRmJ4OBgqNVqzJkzR9PmzTffxPHjx7F48WJcv34dGzZswI8//ohp06YBKJuwdebMmfj444+xc+dOREdH46WXXkLz5s0rPLVIRNTU3T/PzL35ZVjGjIjq361bt7BixQpcvXoV0dHRmDp1KmJjY/HCCy9IHZokSkpKkJycjPnz56Nbt26SjGIiIiIiIiJqCGZ1n4WJHSdCLaoxausoXEm/InVINWYg5cFHjx6NtLQ0LFiwAMnJyejYsSP27NkDR0dHAEB8fLzW/DFFRUWYP38+bt68CXNzcwwaNAg///wzrK2tNW26dOmC7du3Y968efjoo4/g7u6O0NBQjBs3TtNmzpw5yM/Px6uvvoqsrCw89dRT2LNnD59+JSJ6gKlpWWnIzMwDKCq6AUCAtXVfaYMioiZBJpNhzZo1mD17NkRRRPv27bF//374+PhIHZokjhw5gt69e6N169bYtm2b1OEQERERERFJRhAELH92Oa7evYqjCUcxZNMQHJ98HDYmNlKHVm2SljLTZxyWT0RNQV5eNE6f7qD5bGzsja5dYyAIcgmjIpIGS5lRU8FSZlRXeM4QERERUX1KyUtBlxVdkJCTgP4e/bFr7C4cu30MSblJcLZwRmDLQMhlur2HVd1rYElHzBARUcNWUHBJ63NR0TUcP+4GL68wODg0nMnKiYiIiIiIiIioaXE0d8TOsTvRc1VP7Lu5D/Zf2COvOE+z3sXSBWHBYQjxaXj3sCSdY4aIiBqutLRwXLo0tsJypTIRFy+ORFpauARRERERERERERERleno1BHTupTNL39/UgYAEnMSMXLLSITHNLx7WEzMEBFRBaKowvXrMwBUVu2ybNn16zMhiiqdxkVERERERERERFROpVZh44WNla4T/3cPa+aemVCpG9Y9LCZmiIiogqysSCiVtx/SQoRSmYCsrEidxURERERERERERHS/yPhI3M6p+h6WCBEJOQmIjG9Y97CYmCEiogqKi5PqtB0REREREREREVFdS8qt3r2p6rbTFSZmiIioAiMj5zptR0REREREREREVNecLap3b6q67XSFiRkiIqrA2joQCoULAKGKFgIUCldYWwfqMiyiRkEUVcjMPIiUlI3IzDyoF3M19erVCzNnztR8dnNzQ2ho6EO3EQQBO3bseOxj19V+iIiIiIiIqPEJbBkIF0sXCFXcwxIgwNXSFYEtG9Y9LCZmiIioAkGQw8srrPzTg2sBAF5eoRAEuU7jItJ3aWnhOH7cDefP90ZMzAs4f743jh93Q1paeL0cb/DgwQgODq50XWRkJARBwL///lvj/Z46dQqvvvrq44anZeHChejYsWOF5UlJSRg4cGCdHqsqhYWFsLW1hb29PZRKpU6OSURERERERLUnl8kRFlx2D+vB5Ez559DgUMhlDeseFhMzRERUKQeHEPj6boNC0UJruULhAl/fbXBwCJEoMiL9lJYWjosXR0Kp1J6UUKlMxMWLI+slOTN58mTs27cPt29XnAhx9erVCAgIQIcOHWq8XwcHB5iamtZFiI/k5OQEhUKhk2P9+uuv8PX1Rdu2bSUfpSOKIkpLSyWNgYiIiIiISB+E+IRg26htaGGpfQ/LxdIF20ZtQ4hPw7uHxcQMERFVycEhBN26xcHf/wB8fDbA3/8AunWLZVKGCGU3zlWq/Gq9SktzcO3aGwDEyvYEALh+fQZKS3OqtT9RrGw/FT333HNwcHDAmjVrtJbn5eVh69atmDx5Mu7evYuxY8eiRYsWMDU1hZ+fHzZu3PjQ/T5YyuzatWt4+umnYWxsjHbt2mHfvn0Vtpk7dy5at24NU1NTeHh44P3330dJSQkAYM2aNfjwww9x/vx5CIIAQRA0MT9Yyiw6Ohp9+vSBiYkJ7Ozs8OqrryIvL0+zfuLEiRg2bBi+/PJLODs7w87ODtOmTdMc62FWrlyJ8ePHY/z48Vi5cmWF9RcvXsRzzz0HS0tLWFhYIDAwEDdu3NCsX7VqFXx9faFQKODs7Izp06cDAOLi4iAIAqKiojRts7KyIAgCDh48CAA4ePAgBEHA7t270blzZygUChw+fBg3btzA0KFD4ejoCHNzc3Tp0gX79+/XikupVGLu3LlwdXWFQqGAl5cXVq5cCVEU4eXlhS+//FKrfVRUFARBwPXr1x/5mxAREREREemDEJ8QxM2Iw4EJB7AhZAMOTDiA2BmxDTIpAwAGUgdAREQNmyDIYWPTS+owiBoctboAkZHmdbQ3EUrlbRw+bFWt1oGBeZDLzR7ZzsDAAC+99BLWrFmD9957D4JQNox769atUKlUGDt2LPLy8tC5c2fMnTsXlpaW+OOPP/Diiy/C09MTTz755COPoVarERISAkdHR5w4cQLZ2dla89GUs7CwwJo1a9C8eXNER0djypQpsLCwwJw5czB69GhcuHABe/bs0SQdrKwq/hb5+fkICgpC9+7dcerUKaSmpuKVV17B9OnTtZJPBw4cgLOzMw4cOIDr169j9OjR6NixI6ZMmVLl97hx4waOHTuG8PBwiKKIN998E7du3UKrVq0AAImJiXj66afRq1cv/P3337C0tMSRI0c0o1qWLVuGWbNm4dNPP8XAgQORnZ2NI0eOPPL3e9A777yDL7/8Eh4eHrCxsUFCQgIGDRqETz75BAqFAuvWrcPgwYNx5coVtGzZEgDw0ksv4dixY/j666/h7++P2NhYpKenQxAEvPzyy1i9ejVmz56tOcbq1avx9NNPw8vLq8bxERERERERNVRymRy93HpJHUa1MDFDRERE1Ii9/PLL+OKLL3Do0CH06tULQNmN+REjRsDKygpWVlZaN+1ff/117N27F1u2bKlWYmb//v24fPky9u7di+bNmwMAFi9eXGFemPnz52veu7m5Yfbs2di0aRPmzJkDExMTmJubw8DAAE5OTlUea8OGDSgqKsK6detgZlaWmPr2228xePBgfPbZZ3B0dAQA2NjY4Ntvv4VcLkfbtm3x7LPPIiIi4qGJmVWrVmHgwIGwsbEBAAQFBWH16tVYuHAhAOC7776DlZUVNm3aBENDQwBA69atNdt//PHHeOuttzBjxgzNsi5dujzy93vQRx99hP79+2s+29rawt/fX/N50aJF2L59O3bu3Inp06fj6tWr2LJlC/bt24d+/foBADw8PDTtJ06ciAULFuDkyZN48sknUVJSgg0bNlQYRUNERERERES6w1JmRERERLUgk5kiMDCvWi8/vz+rtU8/vz+rtT+ZrPrzu7Rt2xY9evTAqlWrAADXr19HZGQkJk+eDABQqVRYtGgR/Pz8YGtrC3Nzc+zduxfx8fHV2n9MTAxcXV01SRkA6N69e4V2mzdvRs+ePeHk5ARzc3PMnz+/2se4/1j+/v6apAwA9OzZE2q1GleuXNEs8/X1hVx+b2JHZ2dnpKamVrlflUqFtWvXYvz48Zpl48ePx5o1a6BWqwGUlf8KDAzUJGXul5qaijt37qBv3741+j6VCQgI0Pqcl5eH2bNnw8fHB9bW1jA3N0dMTIzmt4uKioJcLsczzzxT6f6aN2+OZ599VvPvf9euXVAqlXj++ecfO1ZqeL777ju4ubnB2NgYXbt2xcmTJx/afuvWrWjbti2MjY3h5+eHP//U/m+VKIpYsGABnJ2dYWJign79+uHatWv1+RWIiIiIiJoEJmaIiIiIakEQBMjlZtV62doOgELhAkCoam9QKFxhazugWvsrL0lWXZMnT8avv/6K3NxcrF69Gp6enpob+V988QXCwsIwd+5cHDhwAFFRUQgKCkJxcfHj/UD3OXbsGMaNG4dBgwbh999/x7lz5/Dee+/V6THu92DyRBAETYKlMnv37kViYiJGjx4NAwMDGBgYYMyYMbh16xYiIiIAACYmJlVu/7B1ACCTlV1y3z83UFVz3tyfdAKA2bNnY/v27Vi8eDEiIyMRFRUFPz8/zW/3qGMDwCuvvIJNmzahsLAQq1evxujRo2FqWv3kHumHzZs3Y9asWfjggw9w9uxZ+Pv7IygoqMqk5NGjRzF27FhMnjwZ586dw7BhwzBs2DBcuHBB0+bzzz/H119/jeXLl+PEiRMwMzNDUFAQioqKdPW1iIiIiIgaJSZmiIiIiOqZIMjh5RVW/unBtQAAL69QCIIc9WHUqFGQyWTYsGED1q1bh5dfflmT3Dly5AiGDh2K8ePHw9/fHx4eHrh69Wq19+3j44OEhAQkJSVplh0/flyrzdGjR9GqVSu89957CAgIgLe3N27duqXVxsjICCqV6pHHOn/+PPLz8zXLjhw5AplMhjZt2lQ75getXLkSY8aMQVRUlNZrzJgxWLlyJQCgQ4cOiIyMrDShYmFhATc3N00S50EODg4AoPUbRUVFVSu2I0eOYOLEiRg+fDj8/Pzg5OSEuLg4zXo/Pz+o1WocOnSoyn0MGjQIZmZmWLZsGfbs2YOXX365Wscm/fLVV19hypQpmDRpEtq1a4fly5fD1NRUM1rqQWFhYQgODsbbb78NHx8fLFq0CJ06dcK3334LoCyRGBoaivnz52Po0KHo0KED1q1bhzt37mDHjh06/GZERERERI0PEzNEREREOuDgEAJf321QKFpoLVcoXODruw0ODiH1dmxzc3OMHj0a8+bNQ1JSEiZOnKhZ5+3tjX379uHo0aOIiYnBa6+9hpSUlGrvu1+/fmjdujUmTJiA8+fPIzIyEu+9955WG29vb8THx2PTpk24ceMGvv76a2zfvl2rjZubG2JjYxEVFYX09HQolcoKxxo3bhyMjY0xYcIEXLhwAQcOHMDrr7+OF198UTO/TE2lpaVh165dmDBhAtq3b6/1eumll7Bjxw5kZGRg+vTpyMnJwZgxY3D69Glcu3YNP//8s6aE2sKFC7F06VJ8/fXXuHbtGs6ePYtvvvkGQNmolm7duuHTTz9FTEwMDh06pDXnzsN4e3sjPDwcUVFROH/+PF544QWt0T9ubm6YMGECXn75ZezYsQOxsbE4ePAgtmzZomkjl8sxceJEzJs3D97e3pWWmiP9VlxcjDNnzmjmGQLKRmr169cPx44dq3SbY8eOabUHyuZWKm8fGxuL5ORkrTZWVlbo2rVrlftUKpXIycnRehERERERUUVMzBARERHpiINDCLp1i4O//wH4+GyAv/8BdOsWW69JmXKTJ09GZmYmgoKCtOaDmT9/Pjp16oSgoCD06tULTk5OGDZsWLX3K5PJsH37dhQWFuLJJ5/EK6+8gk8++USrzZAhQ/Dmm29i+vTp6NixI44ePYr3339fq82IESMQHByM3r17w8HBARs3bqxwLFNTU+zduxcZGRno0qULRo4cib59+2qe8K+NdevWwczMrNL5Yfr27QsTExP88ssvsLOzw99//428vDw888wz6Ny5M1asWKEpmzZhwgSEhobi+++/h6+vL5577jmtuThWrVqF0tJSdO7cGTNnzsTHH39crfi++uor2NjYoEePHhg8eDCCgoLQqVMnrTbLli3DyJEj8d///hdt27bFlClTtEYVAWX//ouLizFp0qSa/kSkB9LT06FSqSokKB0dHZGcnFzpNsnJyQ9tX/5nTfa5ZMkSWFlZaV6urq61+j5ERERERI2dIN5f7JqqLScnB1ZWVsjOzoalpaXU4RAREVE9KyoqQmxsLNzd3WFsbCx1OEQ1EhkZib59+yIhIeGRo4uqOtd5/dtw3blzBy1atMDRo0e1RkTNmTMHhw4dwokTJypsY2RkhLVr12Ls2LGaZd9//z0+/PBDpKSk4OjRo+jZsyfu3LkDZ2dnTZtRo0ZBEARs3ry5wj6VSqXWaLecnBy4urrynCEiIiKiJqO6/SYDHcZEREREREQ6pFQqkZaWhoULF+L555+vdck3atjs7e0hl8srlCFMSUmBk5NTpds4OTk9tH35nykpKVqJmZSUFHTs2LHSfSoUCigUitp+DSIiIiKiJoOlzIiIiIiIGqmNGzeiVatWyMrKwueffy51OFRPjIyM0LlzZ0RERGiWqdVqREREVDmnUPfu3bXaA8C+ffs07d3d3eHk5KTVJicnBydOnOA8RUREREREj4kjZoiIiIiIGqmJEydi4sSJUodBOjBr1ixMmDABAQEBePLJJxEaGor8/HzNvEIvvfQSWrRogSVLlgAAZsyYgWeeeQZLly7Fs88+i02bNuH06dP48ccfAQCCIGjmQ/L29oa7uzvef/99NG/evEbzUBERERERUUVMzBAREREREem50aNHIy0tDQsWLEBycjI6duyIPXv2aMrXxcfHQya7VzChR48e2LBhA+bPn493330X3t7e2LFjB9q3b69pM2fOHOTn5+PVV19FVlYWnnrqKezZs4fzbBERERERPSZBFEVR6iD0ESc/JSIialrKJ0R3c3ODiYmJ1OEQ1ZvCwkLExcXB3d1d6wY8r3+ppnjOEBEREVFTU91rYM4xQ0RERFQNhoaGAICCggKJIyGqX+XnePk5T0RERERERHWLpcyIiIiIqkEul8Pa2hqpqakAAFNTUwiCIHFURHVHFEUUFBQgNTUV1tbWkMvlUodERERERETUKDExQ0RERFRNTk5OAKBJzhA1RtbW1ppznYiIiIiIiOoeEzNERERE1SQIApydndGsWTOUlJRIHQ5RnTM0NORIGSIiIiIionrGxAwRERFRDcnlct68JiIiIiIiIqJakUkdABERERERERERERERUVPBxAwREREREREREREREZGOMDFDRERERERERERERESkI5xjppZEUQQA5OTkSBwJEREREVH9K7/uLb8OJnoU9pmIiIiIqKmpbr+JiZlays3NBQC4urpKHAkRERERke7k5ubCyspK6jBID7DPRERERERN1aP6TYLIR95qRa1W486dO7CwsIAgCFKHQ/fJycmBq6srEhISYGlpKXU4pCd43lBt8dyh2uB5Q7Uh9XkjiiJyc3PRvHlzyGSsiEyPxj5Twyb1f1NIP/G8odrgeUO1xXOHakPq86a6/SaOmKklmUwGFxcXqcOgh7C0tOR/tKnGeN5QbfHcodrgeUO1IeV5w5EyVBPsM+kH/r+IaoPnDdUGzxuqLZ47VBsNvd/ER92IiIiIiIiIiIiIiIh0hIkZIiIiIiIiIiIiIiIiHWFihhodhUKBDz74AAqFQupQSI/wvKHa4rlDtcHzhmqD5w0R1SX+N4Vqg+cN1QbPG6otnjtUG/py3giiKIpSB0FERERERERERERERNQUcMQMERERERERERERERGRjjAxQ0REREREREREREREpCNMzBAREREREREREREREekIEzNEREREREREREREREQ6wsQMNRpLlixBly5dYGFhgWbNmmHYsGG4cuWK1GGRnvn0008hCAJmzpwpdSjUwCUmJmL8+PGws7ODiYkJ/Pz8cPr0aanDogZOpVLh/fffh7u7O0xMTODp6YlFixZBFEWpQ6MG5J9//sHgwYPRvHlzCIKAHTt2aK0XRRELFiyAs7MzTExM0K9fP1y7dk2aYIlIr7DPRHWBfSaqCfabqKbYZ6Lq0vd+ExMz1GgcOnQI06ZNw/Hjx7Fv3z6UlJRgwIAByM/Plzo00hOnTp3CDz/8gA4dOkgdCjVwmZmZ6NmzJwwNDbF7925cunQJS5cuhY2NjdShUQP32WefYdmyZfj2228RExODzz77DJ9//jm++eYbqUOjBiQ/Px/+/v747rvvKl3/+eef4+uvv8by5ctx4sQJmJmZISgoCEVFRTqOlIj0DftM9LjYZ6KaYL+JaoN9Jqoufe83CSLTjdRIpaWloVmzZjh06BCefvppqcOhBi4vLw+dOnXC999/j48//hgdO3ZEaGio1GFRA/XOO+/gyJEjiIyMlDoU0jPPPfccHB0dsXLlSs2yESNGwMTEBL/88ouEkVFDJQgCtm/fjmHDhgEoe+qrefPmeOuttzB79mwAQHZ2NhwdHbFmzRqMGTNGwmiJSN+wz0Q1wT4T1RT7TVQb7DNRbehjv4kjZqjRys7OBgDY2tpKHAnpg2nTpuHZZ59Fv379pA6F9MDOnTsREBCA559/Hs2aNcMTTzyBFStWSB0W6YEePXogIiICV69eBQCcP38ehw8fxsCBAyWOjPRFbGwskpOTtf5/ZWVlha5du+LYsWMSRkZE+oh9JqoJ9pmopthvotpgn4nqgj70mwykDoCoPqjVasycORM9e/ZE+/btpQ6HGrhNmzbh7NmzOHXqlNShkJ64efMmli1bhlmzZuHdd9/FqVOn8MYbb8DIyAgTJkyQOjxqwN555x3k5OSgbdu2kMvlUKlU+OSTTzBu3DipQyM9kZycDABwdHTUWu7o6KhZR0RUHewzUU2wz0S1wX4T1Qb7TFQX9KHfxMQMNUrTpk3DhQsXcPjwYalDoQYuISEBM2bMwL59+2BsbCx1OKQn1Go1AgICsHjxYgDAE088gQsXLmD58uXsYNBDbdmyBevXr8eGDRvg6+uLqKgozJw5E82bN+e5Q0REOsU+E1UX+0xUW+w3UW2wz0RNBUuZUaMzffp0/P777zhw4ABcXFykDocauDNnziA1NRWdOnWCgYEBDAwMcOjQIXz99dcwMDCASqWSOkRqgJydndGuXTutZT4+PoiPj5coItIXb7/9Nt555x2MGTMGfn5+ePHFF/Hmm29iyZIlUodGesLJyQkAkJKSorU8JSVFs46I6FHYZ6KaYJ+Jaov9JqoN9pmoLuhDv4mJGWo0RFHE9OnTsX37dvz9999wd3eXOiTSA3379kV0dDSioqI0r4CAAIwbNw5RUVGQy+VSh0gNUM+ePXHlyhWtZVevXkWrVq0kioj0RUFBAWQy7csvuVwOtVotUUSkb9zd3eHk5ISIiAjNspycHJw4cQLdu3eXMDIi0gfsM1FtsM9EtcV+E9UG+0xUF/Sh38RSZtRoTJs2DRs2bMBvv/0GCwsLTb1AKysrmJiYSBwdNVQWFhYVamqbmZnBzs6OtbapSm+++SZ69OiBxYsXY9SoUTh58iR+/PFH/Pjjj1KHRg3c4MGD8cknn6Bly5bw9fXFuXPn8NVXX+Hll1+WOjRqQPLy8nD9+nXN59jYWERFRcHW1hYtW7bEzJkz8fHHH8Pb2xvu7u54//330bx5cwwbNky6oIlIL7DPRLXBPhPVFvtNVBvsM1F16Xu/SRBFUZQ6CKK6IAhCpctXr16NiRMn6jYY0mu9evVCx44dERoaKnUo1ID9/vvvmDdvHq5duwZ3d3fMmjULU6ZMkTosauByc3Px/vvvY/v27UhNTUXz5s0xduxYLFiwAEZGRlKHRw3EwYMH0bt37wrLJ0yYgDVr1kAURXzwwQf48ccfkZWVhaeeegrff/89WrduLUG0RKRP2GeiusI+E1UX+01UU+wzUXXpe7+JiRkiIiIiIiIiIiIiIiId4RwzREREREREREREREREOsLEDBERERERERERERERkY4wMUNERERERERERERERKQjTMwQERERERERERERERHpCBMzREREREREREREREREOsLEDBERERERERERERERkY4wMUNERERERERERERERKQjTMwQERERERERERERERHpCBMzRETUaAmCgB07dkgdBhERERERUYPEPhMRkTSYmCEionoxceJECIJQ4RUcHCx1aERERERERJJjn4mIqOkykDoAIiJqvIKDg7F69WqtZQqFQqJoiIiIiIiIGhb2mYiImiaOmCEionqjUCjg5OSk9bKxsQFQNmR+2bJlGDhwIExMTODh4YFt27ZpbR8dHY0+ffrAxMQEdnZ2ePXVV5GXl6fVZtWqVfD19YVCoYCzszOmT5+utT49PR3Dhw+HqakpvL29sXPnTs26zMxMjBs3Dg4ODjAxMYG3t3eFThEREREREVF9YZ+JiKhpYmKGiIgk8/7772PEiBE4f/48xo0bhzFjxiAmJgYAkJ+fj6CgINjY2ODUqVPYunUr9u/fr9WJWLZsGaZNm4ZXX30V0dHR2LlzJ7y8vLSO8eGHH2LUqFH4999/MWjQIIwbNw4ZGRma41+6dAm7d+9GTEwMli1bBnt7e939AERERERERA/BPhMRUeMkiKIoSh0EERE1PhMnTsQvv/wCY2NjreXvvvsu3n33XQiCgP/85z9YtmyZZl23bt3QqVMnfP/991ixYgXmzp2LhIQEmJmZAQD+/PNPDB48GHfu3IGjoyNatGiBSZMm4eOPP640BkEQMH/+fCxatAhAWcfF3Nwcu3fvRnBwMIYMGQJ7e3usWrWqnn4FIiIiIiKiyrHPRETUdHGOGSIiqje9e/fW6kQAgK2treZ99+7dtdZ1794dUVFRAICYmBj4+/trOhgA0LNnT6jValy5cgWCIODOnTvo27fvQ2Po0KGD5r2ZmRksLS2RmpoKAJg6dSpGjBiBs2fPYsCAARg2bBh69OhRq+9KRERERERUU+wzERE1TUzMEBFRvTEzM6swTL6umJiYVKudoaGh1mdBEKBWqwEAAwcOxK1bt/Dnn39i37596Nu3L6ZNm4Yvv/yyzuMlIiIiIiJ6EPtMRERNE+eYISIiyRw/frzCZx8fHwCAj48Pzp8/j/z8fM36I0eOQCaToU2bNrCwsICbmxsiIiIeKwYHBwdMmDABv/zyC0JDQ/Hjjz8+1v6IiIiIiIjqCvtMRESNE0fMEBFRvVEqlUhOTtZaZmBgoJkscuvWrQgICMBTTz2F9evX4+TJk1i5ciUAYNy4cfjggw8wYcIELFy4EGlpaXj99dfx4osvwtHREQCwcOFC/Oc//0GzZs0wcOBA5Obm4siRI3j99derFd+CBQvQuXNn+Pr6QqlU4vfff9d0coiIiIiIiOob+0xERE0TEzNERFRv9uzZA2dnZ61lbdq0weXLlwEAH374ITZt2oT//ve/cHZ2xsaNG9GuXTsAgKmpKfbu3YsZM2agS5cuMDU1xYgRI/DVV19p9jVhwgQUFRXh//7v/zB79mzY29tj5MiR1Y7PyMgI8+bNQ1xcHExMTBAYGIhNmzbVwTcnIiIiIiJ6NPaZiIiaJkEURVHqIIiIqOkRBAHbt2/HsGHDpA6FiIiIiIiowWGfiYio8eIcM0RERERERERERERERDrCxAwREREREREREREREZGOsJQZERERERERERERERGRjnDEDBERERERERERERERkY4wMUNERERERERERERERKQjTMwQERERERERERERERHpCBMzREREREREREREREREOsLEDBERERERERERERERkY4wMUNERERERERERERERKQjTMwQERERERERERERERHpCBMzREREREREREREREREOvL/PwokB++uloMAAAAASUVORK5CYII=\n"
+ },
+ "metadata": {}
+ }
+ ]
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "true_labels = []\n",
+ "for _, labels in test_ds:\n",
+ " true_labels.extend(labels.numpy())\n",
+ "\n",
+ "# Predict with the model\n",
+ "pred_probs = model.predict(test_ds)\n",
+ "pred_labels = np.argmax(pred_probs, axis=1)\n",
+ "\n",
+ "# Compute confusion matrix\n",
+ "cm = confusion_matrix(true_labels, pred_labels)\n",
+ "\n",
+ "# Display\n",
+ "cm_display = ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=le.classes_)\n",
+ "cm_display.plot(cmap='Blues', values_format='d')\n",
+ "plt.show()"
+ ],
+ "metadata": {
+ "id": "oGYLbAjGcuA6",
+ "colab": {
+ "base_uri": "https://localhost:8080/",
+ "height": 472
+ },
+ "outputId": "869a0994-31f6-4323-f418-e6fe3964fb4c"
+ },
+ "execution_count": 27,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "\u001b[1m10/10\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 815ms/step\n"
+ ]
+ },
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ ""
+ ],
+ "image/png": "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\n"
+ },
+ "metadata": {}
+ }
+ ]
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "# Evaluate on test dataset\n",
+ "test_loss, test_accuracy = model.evaluate(test_ds, verbose=1)\n",
+ "print(f\"Test Accuracy: {test_accuracy:.4f}\")\n",
+ "\n",
+ "# Predict probabilities\n",
+ "y_pred_probs = model.predict(test_ds)\n",
+ "y_pred = np.argmax(y_pred_probs, axis=1)\n",
+ "\n",
+ "# True labels (same order as test_ds batching)\n",
+ "y_true = np.concatenate([y for x, y in test_ds], axis=0)\n",
+ "\n",
+ "# Metrics\n",
+ "precision = precision_score(y_true, y_pred, average='macro')\n",
+ "recall = recall_score(y_true, y_pred, average='macro')\n",
+ "f1 = f1_score(y_true, y_pred, average='macro')\n",
+ "\n",
+ "print(f\"Precision: {precision:.4f}, Recall: {recall:.4f}, F1-score: {f1:.4f}\")\n",
+ "\n",
+ "# detailed report per class\n",
+ "print(\"\\nClassification Report:\")\n",
+ "print(classification_report(y_true, y_pred, target_names=le.classes_))"
+ ],
+ "metadata": {
+ "id": "oEO1pt4gcwWp",
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "outputId": "7fa01030-c39b-4f2f-c4e1-3cd00b9d90ce"
+ },
+ "execution_count": 28,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "\u001b[1m10/10\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 83ms/step - accuracy: 0.9826 - loss: 0.3217\n",
+ "Test Accuracy: 0.9900\n",
+ "\u001b[1m10/10\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 82ms/step\n",
+ "Precision: 0.9836, Recall: 0.9833, F1-score: 0.9833\n",
+ "\n",
+ "Classification Report:\n",
+ " precision recall f1-score support\n",
+ "\n",
+ " cats 0.99 0.96 0.97 100\n",
+ " dogs 0.96 0.99 0.98 100\n",
+ " snakes 1.00 1.00 1.00 100\n",
+ "\n",
+ " accuracy 0.98 300\n",
+ " macro avg 0.98 0.98 0.98 300\n",
+ "weighted avg 0.98 0.98 0.98 300\n",
+ "\n"
+ ]
+ }
+ ]
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "# Evaluate model\n",
+ "test_loss, test_accuracy = model.evaluate(test_ds, verbose=1)\n",
+ "print(f\"Test Accuracy: {test_accuracy:.4f}\")\n",
+ "\n",
+ "# Predictions\n",
+ "y_probs = model.predict(test_ds) # shape: (num_samples, num_classes)\n",
+ "y_pred = np.argmax(y_probs, axis=1)\n",
+ "\n",
+ "# True labels (extract from test_ds)\n",
+ "y_true = np.concatenate([y for _, y in test_ds], axis=0)\n",
+ "\n",
+ "# Classification report\n",
+ "print(\"\\nClassification Report:\")\n",
+ "print(classification_report(y_true, y_pred, target_names=le.classes_))\n",
+ "\n",
+ "# Confusion matrix\n",
+ "cm = confusion_matrix(y_true, y_pred)\n",
+ "plt.figure(figsize=(10,8))\n",
+ "sns.heatmap(cm, annot=True, fmt=\"d\", cmap=\"Blues\",\n",
+ " xticklabels=le.classes_, yticklabels=le.classes_)\n",
+ "plt.xlabel(\"Predicted\")\n",
+ "plt.ylabel(\"True\")\n",
+ "plt.title(\"Confusion Matrix\")\n",
+ "plt.show()\n",
+ "\n",
+ "# ROC curve (multi-class, one-vs-rest)\n",
+ "y_true_bin = label_binarize(y_true, classes=np.arange(len(le.classes_))) # binarized true labels\n",
+ "\n",
+ "plt.figure(figsize=(10,8))\n",
+ "for i in range(len(le.classes_)):\n",
+ " fpr, tpr, _ = roc_curve(y_true_bin[:, i], y_probs[:, i])\n",
+ " plt.plot(fpr, tpr, label=f\"{le.classes_[i]}\")\n",
+ "plt.plot([0,1],[0,1],'k--', label='Random')\n",
+ "plt.xlabel(\"False Positive Rate\")\n",
+ "plt.ylabel(\"True Positive Rate\")\n",
+ "plt.title(\"ROC Curves (One-vs-Rest)\")\n",
+ "plt.legend()\n",
+ "plt.show()"
+ ],
+ "metadata": {
+ "id": "FWC_3N3KcxvO",
+ "colab": {
+ "base_uri": "https://localhost:8080/",
+ "height": 1000
+ },
+ "outputId": "50e25ae9-d4f5-4a15-ce7a-70e17f8ff777"
+ },
+ "execution_count": 29,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "\u001b[1m10/10\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 106ms/step - accuracy: 0.9826 - loss: 0.3082\n",
+ "Test Accuracy: 0.9900\n",
+ "\u001b[1m10/10\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 93ms/step\n",
+ "\n",
+ "Classification Report:\n",
+ " precision recall f1-score support\n",
+ "\n",
+ " cats 0.99 0.97 0.98 100\n",
+ " dogs 0.97 0.99 0.98 100\n",
+ " snakes 1.00 1.00 1.00 100\n",
+ "\n",
+ " accuracy 0.99 300\n",
+ " macro avg 0.99 0.99 0.99 300\n",
+ "weighted avg 0.99 0.99 0.99 300\n",
+ "\n"
+ ]
+ },
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ ""
+ ],
+ "image/png": "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\n"
+ },
+ "metadata": {}
+ },
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ ""
+ ],
+ "image/png": "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\n"
+ },
+ "metadata": {}
+ }
+ ]
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "# Predictions\n",
+ "y_probs = model.predict(test_ds)\n",
+ "y_pred = np.argmax(y_probs, axis=1)\n",
+ "\n",
+ "# True labels\n",
+ "y_true = np.concatenate([y for _, y in test_ds], axis=0)\n",
+ "\n",
+ "# Metrics per class\n",
+ "precision, recall, f1, support = precision_recall_fscore_support(\n",
+ " y_true, y_pred, average=None, labels=np.arange(len(le.classes_))\n",
+ ")\n",
+ "\n",
+ "df_metrics = pd.DataFrame({\n",
+ " 'Class': le.classes_, # use actual class names instead of 0,1,2,3\n",
+ " 'Precision': precision,\n",
+ " 'Recall': recall,\n",
+ " 'F1-score': f1,\n",
+ " 'Support': support\n",
+ "})\n",
+ "\n",
+ "# Sort by F1-score ascending\n",
+ "df_metrics_sorted = df_metrics.sort_values(by='F1-score')\n",
+ "print(df_metrics_sorted)\n",
+ "\n",
+ "# Macro averages\n",
+ "precision_macro, recall_macro, f1_macro, _ = precision_recall_fscore_support(\n",
+ " y_true, y_pred, average='macro'\n",
+ ")\n",
+ "print(f\"\\nMacro Avg -> Precision: {precision_macro:.4f}, Recall: {recall_macro:.4f}, F1-score: {f1_macro:.4f}\")"
+ ],
+ "metadata": {
+ "id": "dl7p4hjaczdg",
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "outputId": "6ca2cce9-2b2a-4a6f-8da0-ac4c72118c8f"
+ },
+ "execution_count": 30,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "\u001b[1m10/10\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 94ms/step\n",
+ " Class Precision Recall F1-score Support\n",
+ "0 cats 0.989583 0.95 0.969388 100\n",
+ "1 dogs 0.951923 0.99 0.970588 100\n",
+ "2 snakes 1.000000 1.00 1.000000 100\n",
+ "\n",
+ "Macro Avg -> Precision: 0.9805, Recall: 0.9800, F1-score: 0.9800\n"
+ ]
+ }
+ ]
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "# Confusion matrix (no annotations, just intensity heatmap)\n",
+ "cm = confusion_matrix(y_true, y_pred)\n",
+ "\n",
+ "plt.figure(figsize=(15,12))\n",
+ "sns.heatmap(cm, annot=False, fmt='d', cmap='Blues',\n",
+ " xticklabels=le.classes_, yticklabels=le.classes_)\n",
+ "plt.xlabel(\"Predicted Class\")\n",
+ "plt.ylabel(\"True Class\")\n",
+ "plt.title(\"Confusion Matrix Heatmap\")\n",
+ "plt.show()"
+ ],
+ "metadata": {
+ "id": "2pVkJmljc4WZ",
+ "colab": {
+ "base_uri": "https://localhost:8080/",
+ "height": 1000
+ },
+ "outputId": "c1d4d3c8-12b6-4553-f3c5-994700c34dc7"
+ },
+ "execution_count": 31,
+ "outputs": [
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ ""
+ ],
+ "image/png": "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\n"
+ },
+ "metadata": {}
+ }
+ ]
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "# Binarize true labels\n",
+ "y_test_bin = label_binarize(y_true, classes=np.arange(len(le.classes_)))\n",
+ "\n",
+ "# Predict class probabilities\n",
+ "y_probs = model.predict(test_ds)\n",
+ "\n",
+ "# Compute macro-average ROC\n",
+ "all_fpr = np.linspace(0, 1, 100)\n",
+ "mean_tpr = 0\n",
+ "\n",
+ "for i in range(len(le.classes_)):\n",
+ " fpr, tpr, _ = roc_curve(y_test_bin[:, i], y_probs[:, i])\n",
+ " mean_tpr += np.interp(all_fpr, fpr, tpr)\n",
+ "\n",
+ "mean_tpr /= len(le.classes_)\n",
+ "roc_auc = auc(all_fpr, mean_tpr)\n",
+ "\n",
+ "# Plot\n",
+ "plt.figure(figsize=(10,6))\n",
+ "plt.plot(all_fpr, mean_tpr, color='b',\n",
+ " label=f'Macro-average ROC (AUC = {roc_auc:.4f})')\n",
+ "plt.plot([0,1],[0,1],'k--', label='Random')\n",
+ "plt.xlabel('False Positive Rate')\n",
+ "plt.ylabel('True Positive Rate')\n",
+ "plt.title('Macro-average ROC Curve')\n",
+ "plt.legend()\n",
+ "plt.show()"
+ ],
+ "metadata": {
+ "id": "csx4x9Kic51r",
+ "colab": {
+ "base_uri": "https://localhost:8080/",
+ "height": 581
+ },
+ "outputId": "6b6aea72-b1bc-422d-c30f-457bc33f7baf"
+ },
+ "execution_count": 32,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "\u001b[1m10/10\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 81ms/step\n"
+ ]
+ },
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ ""
+ ],
+ "image/png": "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\n"
+ },
+ "metadata": {}
+ }
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "source": [
+ "# **Saving the Model**"
+ ],
+ "metadata": {
+ "id": "Yhzt1hPHwfJa"
+ }
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "model.save(\"Simple_CNN_Classification.h5\")\n",
+ "files.download(\"Simple_CNN_Classification.h5\")"
+ ],
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/",
+ "height": 54
+ },
+ "id": "p5TmIB7M4AXN",
+ "outputId": "0afe668a-a49e-4e20-9110-2db6ba4500fc"
+ },
+ "execution_count": 33,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stderr",
+ "text": [
+ "WARNING:absl:You are saving your model as an HDF5 file via `model.save()` or `keras.saving.save_model(model)`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')` or `keras.saving.save_model(model, 'my_model.keras')`. \n"
+ ]
+ },
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ ""
+ ],
+ "application/javascript": [
+ "\n",
+ " async function download(id, filename, size) {\n",
+ " if (!google.colab.kernel.accessAllowed) {\n",
+ " return;\n",
+ " }\n",
+ " const div = document.createElement('div');\n",
+ " const label = document.createElement('label');\n",
+ " label.textContent = `Downloading \"${filename}\": `;\n",
+ " div.appendChild(label);\n",
+ " const progress = document.createElement('progress');\n",
+ " progress.max = size;\n",
+ " div.appendChild(progress);\n",
+ " document.body.appendChild(div);\n",
+ "\n",
+ " const buffers = [];\n",
+ " let downloaded = 0;\n",
+ "\n",
+ " const channel = await google.colab.kernel.comms.open(id);\n",
+ " // Send a message to notify the kernel that we're ready.\n",
+ " channel.send({})\n",
+ "\n",
+ " for await (const message of channel.messages) {\n",
+ " // Send a message to notify the kernel that we're ready.\n",
+ " channel.send({})\n",
+ " if (message.buffers) {\n",
+ " for (const buffer of message.buffers) {\n",
+ " buffers.push(buffer);\n",
+ " downloaded += buffer.byteLength;\n",
+ " progress.value = downloaded;\n",
+ " }\n",
+ " }\n",
+ " }\n",
+ " const blob = new Blob(buffers, {type: 'application/binary'});\n",
+ " const a = document.createElement('a');\n",
+ " a.href = window.URL.createObjectURL(blob);\n",
+ " a.download = filename;\n",
+ " div.appendChild(a);\n",
+ " a.click();\n",
+ " div.remove();\n",
+ " }\n",
+ " "
+ ]
+ },
+ "metadata": {}
+ },
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ ""
+ ],
+ "application/javascript": [
+ "download(\"download_a7ffc351-6c0e-4800-8ded-561c84c82e72\", \"Simple_CNN_Classification.h5\", 142072584)"
+ ]
+ },
+ "metadata": {}
+ }
+ ]
+ }
+ ]
+}
\ No newline at end of file