Update tiny models for VideomtForUniversalSegmentation

config.json

@@ -0,0 +1,50 @@
+{
+  "architectures": [
+    "VideomtForUniversalSegmentation"
+  ],
+  "attention_dropout": 0.0,
+  "class_weight": 2.0,
+  "dice_weight": 5.0,
+  "drop_path_rate": 0.0,
+  "dtype": "float32",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 8,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3"
+  },
+  "image_size": 40,
+  "importance_sample_ratio": 0.75,
+  "initializer_range": 0.02,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3
+  },
+  "layer_norm_eps": 1e-06,
+  "layerscale_value": 1.0,
+  "mask_weight": 5.0,
+  "mlp_ratio": 4,
+  "model_type": "videomt",
+  "no_object_weight": 0.1,
+  "num_attention_heads": 2,
+  "num_blocks": 1,
+  "num_channels": 3,
+  "num_hidden_layers": 2,
+  "num_queries": 5,
+  "num_register_tokens": 19,
+  "num_upscale_blocks": 2,
+  "oversample_ratio": 3.0,
+  "patch_size": 2,
+  "rope_parameters": {
+    "rope_theta": 100.0,
+    "rope_type": "default"
+  },
+  "train_num_points": 12544,
+  "transformers_version": "5.16.0.dev0",
+  "use_swiglu_ffn": false
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72e93e3fe135488b51ce63b4df1c05fc0ecc672ffb3ca90774d66a7115cc674c
+size 31276

video_preprocessor_config.json

@@ -0,0 +1,30 @@
+{
+  "crop_size": {
+    "height": 40,
+    "width": 40
+  },
+  "do_center_crop": false,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "do_sample_frames": false,
+  "image_mean": [
+    0.485,
+    0.456,
+    0.406
+  ],
+  "image_std": [
+    0.229,
+    0.224,
+    0.225
+  ],
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "return_metadata": false,
+  "size": {
+    "height": 40,
+    "width": 40
+  },
+  "video_processor_type": "VideomtVideoProcessor"
+}