Upload folder using huggingface_hub

README.md

@@ -0,0 +1,88 @@
+# Model Card · RTX 5090 Vision Encoder-Decoder Image Captioning
+
+## System
+
+`02-image-caption-generation-vision-encoder-decoder`
+
+## Status
+
+**Execution-ready, not yet result-complete.** This repository contains the complete training, evaluation, optimization, and deployment workflow. Fine-tuned weights, new metrics, ONNX parity results, and the final public URL must be generated on the target RTX 5090 system.
+
+## Architecture
+
+- Python checkpoint: `nlpconnect/vit-gpt2-image-captioning`
+- Vision encoder: ViT
+- Text decoder: GPT-2 with encoder cross-attention
+- Framework: PyTorch and Hugging Face `VisionEncoderDecoderModel`
+- Browser baseline: `Xenova/vit-gpt2-image-captioning`
+- Browser runtime: Transformers.js and ONNX Runtime Web
+
+## Training method
+
+1. Freeze the vision encoder and adapt the decoder/cross-modal path.
+2. Unfreeze the top six vision blocks with differential learning rates.
+3. Conservatively fine-tune the full architecture.
+4. Select checkpoints using generated-caption validation metrics, with validation-loss fallback.
+
+Implemented controls include BF16/FP16 handling, TF32, fused AdamW fallback, gradient accumulation, label smoothing, warm-up, cosine decay, minimum LR floor, gradient clipping, early stopping, resumable checkpoints, environment capture, and peak GPU-memory reporting.
+
+## Data
+
+Primary experiment: Flickr30k. Multiple references remain grouped for evaluation. The repository does not redistribute the complete dataset.
+
+## Inputs
+
+RGB JPG, PNG, WEBP, or BMP images after safe decoding, EXIF orientation correction, and checkpoint-native image processing.
+
+## Outputs
+
+- greedy caption;
+- ranked beam candidates;
+- concise model-generated alt-text draft;
+- token transition scores where available;
+- sequence score where available;
+- latency and generation settings;
+- optional genuine cross-attention output.
+
+## Evaluation
+
+Required final evidence:
+
+- BLEU-1 and BLEU-4;
+- METEOR;
+- ROUGE-L;
+- CIDEr;
+- CLIPScore;
+- distinct-1 and distinct-2;
+- unique-caption ratio and repetition rate;
+- greedy-versus-beam analysis;
+- model size and latency;
+- quality challenge performance;
+- manual failure analysis;
+- PyTorch-versus-ONNX parity.
+
+No new score is claimed until the corresponding artifact exists under `outputs/`.
+
+## Intended use
+
+Education, portfolio demonstration, image-captioning experimentation, draft metadata, human-reviewed visual documentation, and quality-reporting research prototypes.
+
+## Prohibited or unsuitable use
+
+Identity recognition, surveillance, sensitive-attribute inference, accessibility-critical publishing without review, product release, official quality decisions, medical, legal, financial, hiring, insurance, security, or safety-critical decisions.
+
+## Limitations
+
+The model can hallucinate, omit important objects, confuse colors and counts, produce generic descriptions, repeat phrases, and fail on poor-quality or out-of-distribution images. Flickr30k is not an industrial inspection dataset. Quality challenge results measure descriptive behavior only and do not establish defect-detection competence.
+
+## Confidence and interpretability
+
+Generated-token probability is not calibrated correctness. The browser's structural heuristic is not presented as model confidence. Token and attention views are interpretability aids and do not prove reasoning.
+
+## Privacy
+
+Do not upload personal, confidential, medical, proprietary, security-sensitive, or unlicensed images to a public Space. Public samples must be synthetic, public-domain, or properly licensed.
+
+## Deployment
+
+The fine-tuned Python checkpoint must be converted to a Transformers.js-compatible ONNX repository and pass parity/browser checks before `fine_tuned_model_id` is enabled in `web/metadata.json`.

config.json

@@ -0,0 +1,77 @@
+{
+  "architectures": [
+    "VisionEncoderDecoderModel"
+  ],
+  "bos_token_id": 50256,
+  "decoder": {
+    "activation_function": "gelu_new",
+    "add_cross_attention": true,
+    "architectures": [
+      "GPT2LMHeadModel"
+    ],
+    "attn_pdrop": 0.1,
+    "decoder_start_token_id": 50256,
+    "dtype": "float32",
+    "embd_pdrop": 0.1,
+    "initializer_range": 0.02,
+    "is_decoder": true,
+    "layer_norm_epsilon": 1e-05,
+    "model_type": "gpt2",
+    "n_ctx": 1024,
+    "n_embd": 768,
+    "n_head": 12,
+    "n_inner": null,
+    "n_layer": 12,
+    "n_positions": 1024,
+    "pad_token_id": 50256,
+    "reorder_and_upcast_attn": false,
+    "resid_pdrop": 0.1,
+    "scale_attn_by_inverse_layer_idx": false,
+    "scale_attn_weights": true,
+    "summary_activation": null,
+    "summary_first_dropout": 0.1,
+    "summary_proj_to_labels": true,
+    "summary_type": "cls_index",
+    "summary_use_proj": true,
+    "task_specific_params": {
+      "text-generation": {
+        "do_sample": true,
+        "max_length": 50
+      }
+    },
+    "use_cache": true,
+    "vocab_size": 50257
+  },
+  "decoder_start_token_id": 50256,
+  "dtype": "float32",
+  "encoder": {
+    "architectures": [
+      "ViTModel"
+    ],
+    "attention_probs_dropout_prob": 0.0,
+    "dtype": "float32",
+    "encoder_stride": 16,
+    "hidden_act": "gelu",
+    "hidden_dropout_prob": 0.0,
+    "hidden_size": 768,
+    "image_size": 224,
+    "initializer_range": 0.02,
+    "intermediate_size": 3072,
+    "layer_norm_eps": 1e-12,
+    "model_type": "vit",
+    "num_attention_heads": 12,
+    "num_channels": 3,
+    "num_hidden_layers": 12,
+    "patch_size": 16,
+    "pooler_act": "tanh",
+    "pooler_output_size": 768,
+    "qkv_bias": true
+  },
+  "eos_token_id": 50256,
+  "is_encoder_decoder": true,
+  "model_type": "vision-encoder-decoder",
+  "pad_token_id": 50256,
+  "tie_word_embeddings": false,
+  "transformers_version": "4.57.6",
+  "use_cache": true
+}

generation_config.json

@@ -0,0 +1,9 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 50256,
+  "decoder_start_token_id": 50256,
+  "eos_token_id": 50256,
+  "max_length": 40,
+  "pad_token_id": 50256,
+  "transformers_version": "4.57.6"
+}

onnx/decoder_model_merged_quantized.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36eb63aea2a27dadf5c1854bedaa28d0079ee90fe3906910d915d2a42f2aa9b7
+size 613593983

onnx/decoder_model_quantized.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:826c644798c966cbf8d3dd08d0fe97d2ae75ffcda200ea1af57fa93ae3a1b98d
+size 154654483

onnx/decoder_with_past_model_quantized.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae65f1b6543339d67804b01e8c600dc7aa3e448070b1b3473218857129730860
+size 154644876

onnx/encoder_model_quantized.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:836be7c5db45e728d9146bc884145ed2d1d5c694ceff980443aa4394cf8607d9
+size 86967767

preprocessor_config.json

@@ -0,0 +1,23 @@
+{
+  "do_convert_rgb": null,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "ViTImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 224,
+    "width": 224
+  }
+}

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,28 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "50256": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "extra_special_tokens": {},
+  "max_length": 32,
+  "model_max_length": 1024,
+  "pad_to_multiple_of": null,
+  "pad_token": "<|endoftext|>",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
+  "stride": 0,
+  "tokenizer_class": "GPT2Tokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
+  "unk_token": "<|endoftext|>"
+}