Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,127 @@
+---
+license: cc-by-nc-4.0
+datasets:
+- openbmb/VisRAG-Ret-Train-Synthetic-data
+- openbmb/VisRAG-Ret-Train-In-domain-data
+- Metric-AI/rag_docmatix_100k
+- vidore/colpali_train_set
+- llamaindex/vdr-multilingual-train
+language:
+- en
+- fr
+- es
+- it
+- de
+base_model:
+- Qwen/Qwen2.5-VL-3B-Instruct
+tags:
+- vidore
+- multimodal_embedding
+- multilingual_embedding
+- Text-to-Visual Document (T→VD) retrieval
+library_name: peft
+---
+# ColQwen2.5-3b-multilingual: Multilingual Visual Retriever based on Qwen2.5-VL-3B-Instruct with ColBERT strategy
+
+### This is the base version trained on 4xA100 80GB with per_device_batch_size=128 and gradient_accumulation_steps=2 for 5 epoch.
+
+ColQwen is a model based on a novel model architecture and training strategy based on Vision Language Models (VLMs) to efficiently index documents from their visual features.
+It is a [Qwen2.5-VL-3B](https://huggingface.co/Qwen/Qwen2.5-VL-3B-Instruct) extension that generates [ColBERT](https://arxiv.org/abs/2004.12832)- style multi-vector representations of text and images. 
+It was introduced in the paper [ColPali: Efficient Document Retrieval with Vision Language Models](https://arxiv.org/abs/2407.01449) and first released in [this repository](https://github.com/ManuelFay/colpali)
+
+<p align="center"><img width=800 src="https://github.com/illuin-tech/colpali/blob/main/assets/colpali_architecture.webp?raw=true"/></p>
+
+## Version specificity
+
+
+This model takes dynamic image resolutions in input and does not resize them, changing their aspect ratio as in ColPali.
+Maximal resolution is set so that 768 image patches are created at most. Experiments show clear improvements with larger amounts of image patches, at the cost of memory requirements.
+
+This version is trained with `colpali-engine==0.3.7`.
+
+## Data
+- **Synthetic data**: Selected and preprocessed from the `openbmb/VisRAG-Ret-Train-Synthetic-data` dataset.  
+- **In-domain VQA dataset**: Drawn from `openbmb/VisRAG-Ret-Train-In-domain-data`.  
+- **Docmatix dataset**: Extracted from the `Metric-AI/rag_docmatix_100k` dataset.  
+- **Colpali dataset**: Taken from `vidore/colpali_train_set`.
+- **Multilingual dataset**: Taken from `llamaindex/vdr-multilingual-train`.
+
+
+## Model Training
+
+### Parameters
+We train models  use low-rank adapters ([LoRA](https://arxiv.org/abs/2106.09685)) 
+with `alpha=128`  and `r=128` on the transformer layers from the language model, 
+as well as the final randomly initialized projection layer, and use a `paged_adamw_8bit` optimizer. 
+We train on an 4xA100 GPU setup with distributed data parallelism (via accelerate), a learning rate of 2e-4 with linear decay with 1% warmup steps, batch size per device is 128, gradient accumulation steps are 2, in `bfloat16` format
+
+## Usage
+
+Make sure `colpali-engine` is installed from source or with a version superior to 0.3.1.
+`transformers` version must be > 4.45.0.
+
+```bash
+pip install git+https://github.com/illuin-tech/colpali
+```
+
+```python
+import torch
+from PIL import Image
+
+from colpali_engine.models import ColQwen2_5, ColQwen2_5_Processor
+
+model = ColQwen2_5.from_pretrained(
+        "Metric-AI/colqwen2.5-3b-multilingual",
+        torch_dtype=torch.bfloat16,
+        device_map="cuda:0",  # or "mps" if on Apple Silicon
+    ).eval()
+processor = ColQwen2_5_Processor.from_pretrained("Metric-AI/colqwen2.5-3b-multilingual")
+
+# Your inputs
+images = [
+    Image.new("RGB", (32, 32), color="white"),
+    Image.new("RGB", (16, 16), color="black"),
+]
+queries = [
+    "Is attention really all you need?",
+    "What is the amount of bananas farmed in Salvador?",
+]
+
+# Process the inputs
+batch_images = processor.process_images(images).to(model.device)
+batch_queries = processor.process_queries(queries).to(model.device)
+
+# Forward pass
+with torch.no_grad():
+    image_embeddings = model(**batch_images)
+    query_embeddings = model(**batch_queries)
+
+scores = processor.score_multi_vector(query_embeddings, image_embeddings)
+```
+
+
+## Limitations
+
+ - **Focus**: The model primarily focuses on PDF-type documents and high-ressources languages, potentially limiting its generalization to other document types or less represented languages.
+ - **Support**: The model relies on multi-vector retreiving derived from the ColBERT late interaction mechanism, which may require engineering efforts to adapt to widely used vector retrieval frameworks that lack native multi-vector support.
+
+## License
+
+ColQwen2.5's vision language backbone model (Qwen2.5-VL) is under `apache2.0` license. The adapters attached to the model are under MIT license.
+
+
+## Citation
+
+If you use this models from this organization in your research, please cite the original paper as follows:
+
+```bibtex
+@misc{faysse2024colpaliefficientdocumentretrieval,
+  title={ColPali: Efficient Document Retrieval with Vision Language Models}, 
+  author={Manuel Faysse and Hugues Sibille and Tony Wu and Bilel Omrani and Gautier Viaud and Céline Hudelot and Pierre Colombo},
+  year={2024},
+  eprint={2407.01449},
+  archivePrefix={arXiv},
+  primaryClass={cs.IR},
+  url={https://arxiv.org/abs/2407.01449}, 
+}
+```

adapter_config.json

@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-VL-3B-Instruct",
+  "bias": "none",
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 128,
+  "lora_bias": false,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 128,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "(.*(model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)",
+  "task_type": "FEATURE_EXTRACTION",
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2571d83e0aefef4fc18a8d8b5d35e9c1812dfa3ed8d8c0b00c221ae3580606c1
+size 959057112

added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

chat_template.json

@@ -0,0 +1,3 @@
+{
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
+}

checkpoint-1800/.gitattributes

@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

checkpoint-1800/README.md

@@ -0,0 +1,127 @@
+---
+license: cc-by-nc-4.0
+datasets:
+- openbmb/VisRAG-Ret-Train-Synthetic-data
+- openbmb/VisRAG-Ret-Train-In-domain-data
+- Metric-AI/rag_docmatix_100k
+- vidore/colpali_train_set
+- llamaindex/vdr-multilingual-train
+language:
+- en
+- fr
+- es
+- it
+- de
+base_model:
+- Qwen/Qwen2.5-VL-3B-Instruct
+tags:
+- vidore
+- multimodal_embedding
+- multilingual_embedding
+- Text-to-Visual Document (T→VD) retrieval
+library_name: peft
+---
+# ColQwen2.5-3b-multilingual: Multilingual Visual Retriever based on Qwen2.5-VL-3B-Instruct with ColBERT strategy
+
+### This is the base version trained on 4xA100 80GB with per_device_batch_size=128 and gradient_accumulation_steps=2 for 5 epoch.
+
+ColQwen is a model based on a novel model architecture and training strategy based on Vision Language Models (VLMs) to efficiently index documents from their visual features.
+It is a [Qwen2.5-VL-3B](https://huggingface.co/Qwen/Qwen2.5-VL-3B-Instruct) extension that generates [ColBERT](https://arxiv.org/abs/2004.12832)- style multi-vector representations of text and images. 
+It was introduced in the paper [ColPali: Efficient Document Retrieval with Vision Language Models](https://arxiv.org/abs/2407.01449) and first released in [this repository](https://github.com/ManuelFay/colpali)
+
+<p align="center"><img width=800 src="https://github.com/illuin-tech/colpali/blob/main/assets/colpali_architecture.webp?raw=true"/></p>
+
+## Version specificity
+
+
+This model takes dynamic image resolutions in input and does not resize them, changing their aspect ratio as in ColPali.
+Maximal resolution is set so that 768 image patches are created at most. Experiments show clear improvements with larger amounts of image patches, at the cost of memory requirements.
+
+This version is trained with `colpali-engine==0.3.7`.
+
+## Data
+- **Synthetic data**: Selected and preprocessed from the `openbmb/VisRAG-Ret-Train-Synthetic-data` dataset.  
+- **In-domain VQA dataset**: Drawn from `openbmb/VisRAG-Ret-Train-In-domain-data`.  
+- **Docmatix dataset**: Extracted from the `Metric-AI/rag_docmatix_100k` dataset.  
+- **Colpali dataset**: Taken from `vidore/colpali_train_set`.
+- **Multilingual dataset**: Taken from `llamaindex/vdr-multilingual-train`.
+
+
+## Model Training
+
+### Parameters
+We train models  use low-rank adapters ([LoRA](https://arxiv.org/abs/2106.09685)) 
+with `alpha=128`  and `r=128` on the transformer layers from the language model, 
+as well as the final randomly initialized projection layer, and use a `paged_adamw_8bit` optimizer. 
+We train on an 4xA100 GPU setup with distributed data parallelism (via accelerate), a learning rate of 2e-4 with linear decay with 1% warmup steps, batch size per device is 128, gradient accumulation steps are 2, in `bfloat16` format
+
+## Usage
+
+Make sure `colpali-engine` is installed from source or with a version superior to 0.3.1.
+`transformers` version must be > 4.45.0.
+
+```bash
+pip install git+https://github.com/illuin-tech/colpali
+```
+
+```python
+import torch
+from PIL import Image
+
+from colpali_engine.models import ColQwen2_5, ColQwen2_5_Processor
+
+model = ColQwen2_5.from_pretrained(
+        "Metric-AI/colqwen2.5-3b-multilingual",
+        torch_dtype=torch.bfloat16,
+        device_map="cuda:0",  # or "mps" if on Apple Silicon
+    ).eval()
+processor = ColQwen2_5_Processor.from_pretrained("Metric-AI/colqwen2.5-3b-multilingual")
+
+# Your inputs
+images = [
+    Image.new("RGB", (32, 32), color="white"),
+    Image.new("RGB", (16, 16), color="black"),
+]
+queries = [
+    "Is attention really all you need?",
+    "What is the amount of bananas farmed in Salvador?",
+]
+
+# Process the inputs
+batch_images = processor.process_images(images).to(model.device)
+batch_queries = processor.process_queries(queries).to(model.device)
+
+# Forward pass
+with torch.no_grad():
+    image_embeddings = model(**batch_images)
+    query_embeddings = model(**batch_queries)
+
+scores = processor.score_multi_vector(query_embeddings, image_embeddings)
+```
+
+
+## Limitations
+
+ - **Focus**: The model primarily focuses on PDF-type documents and high-ressources languages, potentially limiting its generalization to other document types or less represented languages.
+ - **Support**: The model relies on multi-vector retreiving derived from the ColBERT late interaction mechanism, which may require engineering efforts to adapt to widely used vector retrieval frameworks that lack native multi-vector support.
+
+## License
+
+ColQwen2.5's vision language backbone model (Qwen2.5-VL) is under `apache2.0` license. The adapters attached to the model are under MIT license.
+
+
+## Citation
+
+If you use this models from this organization in your research, please cite the original paper as follows:
+
+```bibtex
+@misc{faysse2024colpaliefficientdocumentretrieval,
+  title={ColPali: Efficient Document Retrieval with Vision Language Models}, 
+  author={Manuel Faysse and Hugues Sibille and Tony Wu and Bilel Omrani and Gautier Viaud and Céline Hudelot and Pierre Colombo},
+  year={2024},
+  eprint={2407.01449},
+  archivePrefix={arXiv},
+  primaryClass={cs.IR},
+  url={https://arxiv.org/abs/2407.01449}, 
+}
+```

checkpoint-1800/adapter_config.json

@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-VL-3B-Instruct",
+  "bias": "none",
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 128,
+  "lora_bias": false,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 128,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "(.*(model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)",
+  "task_type": "FEATURE_EXTRACTION",
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-1800/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f51b1f9f6e343dffcc47e839d0c5957fd9758f1892dc4cacf39ce8c04e2d1bde
+size 959057112

checkpoint-1800/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoint-1800/chat_template.json

@@ -0,0 +1,3 @@
+{
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
+}

checkpoint-1800/git_hash.txt

@@ -0,0 +1 @@
+59e94a92790b67bd60507608c3115a2e48f83a07

checkpoint-1800/preprocessor_config.json

@@ -0,0 +1,29 @@
+{
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "Qwen2_5_VLImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "max_pixels": 12845056,
+  "merge_size": 2,
+  "min_pixels": 3136,
+  "patch_size": 14,
+  "processor_class": "ColQwen2_5Processor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "max_pixels": 12845056,
+    "min_pixels": 3136
+  },
+  "temporal_patch_size": 2
+}

checkpoint-1800/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-1800/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

checkpoint-1800/tokenizer_config.json

@@ -0,0 +1,209 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "processor_class": "ColQwen2_5Processor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-1800/training_config.yml

@@ -0,0 +1,72 @@
+config:
+  (): colpali_engine.trainer.colmodel_training.ColModelTrainingConfig
+  output_dir: !path ../../../models/colqwen2_5-3b-mixed-multiL-5ep-l-128
+  processor:
+    (): colpali_engine.utils.transformers_wrappers.AllPurposeWrapper
+    class_to_instanciate: !ext colpali_engine.models.ColQwen2_5Processor
+    pretrained_model_name_or_path:  "Qwen/Qwen2.5-VL-3B-Instruct" # "./models/paligemma-3b-mix-448"
+    # num_image_tokens: 2048
+    # max_length: 50
+
+  model:
+    (): colpali_engine.utils.transformers_wrappers.AllPurposeWrapper
+    class_to_instanciate: !ext colpali_engine.models.ColQwen2_5
+    pretrained_model_name_or_path: "Qwen/Qwen2.5-VL-3B-Instruct"
+    torch_dtype:  !ext torch.bfloat16
+    use_cache: false
+    attn_implementation: "flash_attention_2"
+#    device_map: "auto"
+#    quantization_config:
+#      (): transformers.BitsAndBytesConfig
+#      load_in_4bit: true
+#      bnb_4bit_quant_type: "nf4"
+#      bnb_4bit_compute_dtype:  "bfloat16"
+#      bnb_4bit_use_double_quant: true
+
+  dataset_loading_func: !ext colpali_engine.utils.dataset_transformation.load_mixed_multiL_train_set
+  eval_dataset_loader: !import ../data/test_data.yaml
+
+  # max_length: 50
+  run_eval: true
+  loss_func:
+    (): colpali_engine.loss.late_interaction_losses.ColbertPairwiseCELoss
+  tr_args:
+    (): transformers.training_args.TrainingArguments
+    output_dir: null
+    overwrite_output_dir: true
+    num_train_epochs: 1
+    per_device_train_batch_size: 128
+    gradient_checkpointing: true
+    gradient_checkpointing_kwargs: { "use_reentrant": false }
+    # gradient_checkpointing: true
+    # 6 x 8 gpus = 48 batch size
+    gradient_accumulation_steps: 2
+    per_device_eval_batch_size: 32
+    eval_strategy: "steps"
+    dataloader_num_workers: 8
+    # bf16: true
+    save_steps: 100
+    logging_steps: 1
+    eval_steps: 180
+    warmup_steps: 19
+    learning_rate: 2e-4
+    save_total_limit: 15
+    resume_from_checkpoint: false
+    optim: "paged_adamw_8bit"
+    # wandb logging
+    # wandb_project: "dolma"
+    run_name: "colqwen2_5-3b-mixed-multiL-5ep-l-128"
+    report_to: wandb
+
+
+  peft_config:
+    (): peft.LoraConfig
+    r: 128
+    lora_alpha: 128
+    lora_dropout: 0.1
+    init_lora_weights: "gaussian"
+    bias: "none"
+    task_type: "FEATURE_EXTRACTION"
+    target_modules: '(.*(model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)'
+    # target_modules: '(.*(language_model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)'
+

git_hash.txt

@@ -0,0 +1 @@
+59e94a92790b67bd60507608c3115a2e48f83a07

preprocessor_config.json

@@ -0,0 +1,29 @@
+{
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "Qwen2_5_VLImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "max_pixels": 12845056,
+  "merge_size": 2,
+  "min_pixels": 3136,
+  "patch_size": 14,
+  "processor_class": "ColQwen2_5Processor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "max_pixels": 12845056,
+    "min_pixels": 3136
+  },
+  "temporal_patch_size": 2
+}

results.json

@@ -0,0 +1 @@
+{"arxivqa_test_subsampled": {"ndcg_at_1": 0.86, "ndcg_at_3": 0.89602, "ndcg_at_5": 0.90738, "ndcg_at_10": 0.91247, "ndcg_at_20": 0.91649, "ndcg_at_100": 0.91954, "ndcg_at_1000": 0.9201, "map_at_1": 0.86, "map_at_3": 0.88767, "map_at_5": 0.89387, "map_at_10": 0.89592, "map_at_20": 0.89701, "map_at_100": 0.89746, "map_at_1000": 0.89748, "recall_at_1": 0.86, "recall_at_3": 0.92, "recall_at_5": 0.948, "recall_at_10": 0.964, "recall_at_20": 0.98, "recall_at_100": 0.996, "recall_at_1000": 1.0, "precision_at_1": 0.86, "precision_at_3": 0.30667, "precision_at_5": 0.1896, "precision_at_10": 0.0964, "precision_at_20": 0.049, "precision_at_100": 0.00996, "precision_at_1000": 0.001, "mrr_at_1": 0.86, "mrr_at_3": 0.8889999999999999, "mrr_at_5": 0.8934999999999997, "mrr_at_10": 0.8961936507936505, "mrr_at_20": 0.8972816403754481, "mrr_at_100": 0.8977296741827361, "mrr_at_1000": 0.8977586086890899, "naucs_at_1_max": 0.722793404461687, "naucs_at_1_std": -0.038146044062629184, "naucs_at_1_diff1": 0.9546556741028124, "naucs_at_3_max": 0.7859710550887037, "naucs_at_3_std": 0.01360877684407265, "naucs_at_3_diff1": 0.9129901960784345, "naucs_at_5_max": 0.8335667600373489, "naucs_at_5_std": 0.04666738490267592, "naucs_at_5_diff1": 0.9340120663650082, "naucs_at_10_max": 0.8512553169415892, "naucs_at_10_std": 0.07505965349101841, "naucs_at_10_diff1": 0.9192084241103815, "naucs_at_20_max": 0.9738562091503187, "naucs_at_20_std": 0.3808590102707724, "naucs_at_20_diff1": 0.9084967320261399, "naucs_at_100_max": 1.0, "naucs_at_100_std": 1.0, "naucs_at_100_diff1": 0.8692810457516423, "naucs_at_1000_max": 1.0, "naucs_at_1000_std": 1.0, "naucs_at_1000_diff1": 1.0}, "docvqa_test_subsampled": {"ndcg_at_1": 0.55654, "ndcg_at_3": 0.63486, "ndcg_at_5": 0.65355, "ndcg_at_10": 0.66767, "ndcg_at_20": 0.68004, "ndcg_at_100": 0.69372, "ndcg_at_1000": 0.70593, "map_at_1": 0.55654, "map_at_3": 0.61456, "map_at_5": 0.62465, "map_at_10": 0.63035, "map_at_20": 0.63377, "map_at_100": 0.63572, "map_at_1000": 0.63622, "recall_at_1": 0.55654, "recall_at_3": 0.69401, "recall_at_5": 0.74058, "recall_at_10": 0.78492, "recall_at_20": 0.8337, "recall_at_100": 0.90687, "recall_at_1000": 1.0, "precision_at_1": 0.55654, "precision_at_3": 0.23134, "precision_at_5": 0.14812, "precision_at_10": 0.07849, "precision_at_20": 0.04169, "precision_at_100": 0.00907, "precision_at_1000": 0.001, "mrr_at_1": 0.5565410199556541, "mrr_at_3": 0.6127124907612713, "mrr_at_5": 0.6236881005173688, "mrr_at_10": 0.6300408263822898, "mrr_at_20": 0.6333156432311732, "mrr_at_100": 0.6353411227279873, "mrr_at_1000": 0.6358054759704315, "naucs_at_1_max": 0.5257722632439136, "naucs_at_1_std": 0.10050303275375208, "naucs_at_1_diff1": 0.8849570154601619, "naucs_at_3_max": 0.4612022712381768, "naucs_at_3_std": 0.15996734473763619, "naucs_at_3_diff1": 0.7968064775356001, "naucs_at_5_max": 0.4464095203043038, "naucs_at_5_std": 0.32716928956958097, "naucs_at_5_diff1": 0.7844569699129116, "naucs_at_10_max": 0.37102681160936635, "naucs_at_10_std": 0.44024504471037584, "naucs_at_10_diff1": 0.7856507094113239, "naucs_at_20_max": 0.3234215889454963, "naucs_at_20_std": 0.4899776595937942, "naucs_at_20_diff1": 0.7677297231158269, "naucs_at_100_max": 0.15381107905278507, "naucs_at_100_std": 0.8716361071120375, "naucs_at_100_diff1": 0.7936512540332907, "naucs_at_1000_max": 1.0, "naucs_at_1000_std": 1.0, "naucs_at_1000_diff1": 1.0}, "infovqa_test_subsampled": {"ndcg_at_1": 0.89069, "ndcg_at_3": 0.92332, "ndcg_at_5": 0.92933, "ndcg_at_10": 0.93387, "ndcg_at_20": 0.93494, "ndcg_at_100": 0.93797, "ndcg_at_1000": 0.93875, "map_at_1": 0.89069, "map_at_3": 0.91565, "map_at_5": 0.9191, "map_at_10": 0.92094, "map_at_20": 0.92126, "map_at_100": 0.92169, "map_at_1000": 0.92171, "recall_at_1": 0.89069, "recall_at_3": 0.94534, "recall_at_5": 0.95951, "recall_at_10": 0.97368, "recall_at_20": 0.97773, "recall_at_100": 0.99393, "recall_at_1000": 1.0, "precision_at_1": 0.89069, "precision_at_3": 0.31511, "precision_at_5": 0.1919, "precision_at_10": 0.09737, "precision_at_20": 0.04889, "precision_at_100": 0.00994, "precision_at_1000": 0.001, "mrr_at_1": 0.8906882591093117, "mrr_at_3": 0.9153171390013495, "mrr_at_5": 0.9191632928475032, "mrr_at_10": 0.9206750851487693, "mrr_at_20": 0.920994064415117, "mrr_at_100": 0.9214189037498206, "mrr_at_1000": 0.9214460799655682, "naucs_at_1_max": 0.6266617765209551, "naucs_at_1_std": -0.19467734849749174, "naucs_at_1_diff1": 0.9490969427703365, "naucs_at_3_max": 0.7646193541044001, "naucs_at_3_std": -0.049253260481305516, "naucs_at_3_diff1": 0.9441833971070621, "naucs_at_5_max": 0.8436958283279918, "naucs_at_5_std": 0.11750221148894235, "naucs_at_5_diff1": 0.9469328268926711, "naucs_at_10_max": 0.8223529961696429, "naucs_at_10_std": 0.1964949763377099, "naucs_at_10_diff1": 0.9284044609167826, "naucs_at_20_max": 0.8964958869122784, "naucs_at_20_std": 0.4419658243478526, "naucs_at_20_diff1": 0.927259949634811, "naucs_at_100_max": 0.9564661819784259, "naucs_at_100_std": 0.39174816052979483, "naucs_at_100_diff1": 0.9564661819784259, "naucs_at_1000_max": 1.0, "naucs_at_1000_std": 1.0, "naucs_at_1000_diff1": 1.0}, "tabfquad_test_subsampled": {"ndcg_at_1": 0.86071, "ndcg_at_3": 0.89133, "ndcg_at_5": 0.89594, "ndcg_at_10": 0.90153, "ndcg_at_20": 0.90908, "ndcg_at_100": 0.91536, "ndcg_at_1000": 0.91536, "map_at_1": 0.86071, "map_at_3": 0.88452, "map_at_5": 0.8872, "map_at_10": 0.8894, "map_at_20": 0.89164, "map_at_100": 0.89262, "map_at_1000": 0.89262, "recall_at_1": 0.86071, "recall_at_3": 0.91071, "recall_at_5": 0.92143, "recall_at_10": 0.93929, "recall_at_20": 0.96786, "recall_at_100": 1.0, "recall_at_1000": 1.0, "precision_at_1": 0.86071, "precision_at_3": 0.30357, "precision_at_5": 0.18429, "precision_at_10": 0.09393, "precision_at_20": 0.04839, "precision_at_100": 0.01, "precision_at_1000": 0.001, "mrr_at_1": 0.8571428571428571, "mrr_at_3": 0.8839285714285714, "mrr_at_5": 0.8864285714285713, "mrr_at_10": 0.8889172335600906, "mrr_at_20": 0.890549232513518, "mrr_at_100": 0.8915407957626755, "mrr_at_1000": 0.8915407957626755, "naucs_at_1_max": 0.565388152825077, "naucs_at_1_std": 0.40889656715304873, "naucs_at_1_diff1": 0.8467772005269835, "naucs_at_3_max": 0.6566573295985063, "naucs_at_3_std": 0.5545284780578894, "naucs_at_3_diff1": 0.8447245564892628, "naucs_at_5_max": 0.6497113997114, "naucs_at_5_std": 0.5355869620575493, "naucs_at_5_diff1": 0.8421186656480768, "naucs_at_10_max": 0.7744548799912124, "naucs_at_10_std": 0.6476080628329782, "naucs_at_10_diff1": 0.8545888943812839, "naucs_at_20_max": 0.8451602863367568, "naucs_at_20_std": 0.7453574022201472, "naucs_at_20_diff1": 0.8306359580869404, "naucs_at_100_max": 1.0, "naucs_at_100_std": 1.0, "naucs_at_100_diff1": 1.0, "naucs_at_1000_max": NaN, "naucs_at_1000_std": NaN, "naucs_at_1000_diff1": NaN}, "tatdqa_test": {"ndcg_at_1": 0.68955, "ndcg_at_3": 0.78416, "ndcg_at_5": 0.80618, "ndcg_at_10": 0.81969, "ndcg_at_20": 0.82389, "ndcg_at_100": 0.82898, "ndcg_at_1000": 0.83022, "map_at_1": 0.68955, "map_at_3": 0.76063, "map_at_5": 0.773, "map_at_10": 0.77866, "map_at_20": 0.77984, "map_at_100": 0.7805, "map_at_1000": 0.78056, "recall_at_1": 0.68955, "recall_at_3": 0.85237, "recall_at_5": 0.90522, "recall_at_10": 0.94654, "recall_at_20": 0.96294, "recall_at_100": 0.99089, "recall_at_1000": 1.0, "precision_at_1": 0.68955, "precision_at_3": 0.28412, "precision_at_5": 0.18104, "precision_at_10": 0.09465, "precision_at_20": 0.04815, "precision_at_100": 0.00991, "precision_at_1000": 0.001, "mrr_at_1": 0.6816524908869988, "mrr_at_3": 0.7562778452814904, "mrr_at_5": 0.7686715269339816, "mrr_at_10": 0.7745462786167524, "mrr_at_20": 0.7758200056081633, "mrr_at_100": 0.7764618378499478, "mrr_at_1000": 0.7765242843809894, "naucs_at_1_max": 0.24837787265702854, "naucs_at_1_std": -0.19241555267499547, "naucs_at_1_diff1": 0.8236282079241498, "naucs_at_3_max": 0.24346304749013487, "naucs_at_3_std": -0.18927905384673094, "naucs_at_3_diff1": 0.7202539222925216, "naucs_at_5_max": 0.2964781524481732, "naucs_at_5_std": -0.07115963375478136, "naucs_at_5_diff1": 0.6729800878199006, "naucs_at_10_max": 0.4000649649660007, "naucs_at_10_std": 0.18029921295145243, "naucs_at_10_diff1": 0.6525400665529622, "naucs_at_20_max": 0.4809424079608138, "naucs_at_20_std": 0.28896008671282786, "naucs_at_20_diff1": 0.6872334498348703, "naucs_at_100_max": 0.4792973536271862, "naucs_at_100_std": 0.42059668900329966, "naucs_at_100_diff1": 0.6931508305328492, "naucs_at_1000_max": NaN, "naucs_at_1000_std": NaN, "naucs_at_1000_diff1": NaN}, "shiftproject_test": {"ndcg_at_1": 0.73, "ndcg_at_3": 0.84202, "ndcg_at_5": 0.85837, "ndcg_at_10": 0.86769, "ndcg_at_20": 0.86769, "ndcg_at_100": 0.86945, "ndcg_at_1000": 0.86945, "map_at_1": 0.73, "map_at_3": 0.815, "map_at_5": 0.824, "map_at_10": 0.82761, "map_at_20": 0.82761, "map_at_100": 0.82781, "map_at_1000": 0.82781, "recall_at_1": 0.73, "recall_at_3": 0.92, "recall_at_5": 0.96, "recall_at_10": 0.99, "recall_at_20": 0.99, "recall_at_100": 1.0, "recall_at_1000": 1.0, "precision_at_1": 0.73, "precision_at_3": 0.30667, "precision_at_5": 0.192, "precision_at_10": 0.099, "precision_at_20": 0.0495, "precision_at_100": 0.01, "precision_at_1000": 0.001, "mrr_at_1": 0.73, "mrr_at_3": 0.82, "mrr_at_5": 0.8320000000000001, "mrr_at_10": 0.8344999999999999, "mrr_at_20": 0.8344999999999999, "mrr_at_100": 0.8346960784313725, "mrr_at_1000": 0.8346960784313725, "naucs_at_1_max": -0.029605457154652116, "naucs_at_1_std": -0.4257209695492341, "naucs_at_1_diff1": 0.8361870687452088, "naucs_at_3_max": 0.3197945845004683, "naucs_at_3_std": -0.3085901027077499, "naucs_at_3_diff1": 0.7829131652661058, "naucs_at_5_max": 0.6050420168067251, "naucs_at_5_std": -0.029178338001862987, "naucs_at_5_diff1": 0.8651960784313721, "naucs_at_10_max": 0.7222222222222276, "naucs_at_10_std": 0.35807656395891135, "naucs_at_10_diff1": 0.8692810457516413, "naucs_at_20_max": 0.7222222222222276, "naucs_at_20_std": 0.35807656395891135, "naucs_at_20_diff1": 0.8692810457516413, "naucs_at_100_max": NaN, "naucs_at_100_std": NaN, "naucs_at_100_diff1": NaN, "naucs_at_1000_max": NaN, "naucs_at_1000_std": NaN, "naucs_at_1000_diff1": NaN}, "syntheticDocQA_artificial_intelligence_test": {"ndcg_at_1": 0.99, "ndcg_at_3": 0.99, "ndcg_at_5": 0.99431, "ndcg_at_10": 0.99431, "ndcg_at_20": 0.99431, "ndcg_at_100": 0.99431, "ndcg_at_1000": 0.99431, "map_at_1": 0.99, "map_at_3": 0.99, "map_at_5": 0.9925, "map_at_10": 0.9925, "map_at_20": 0.9925, "map_at_100": 0.9925, "map_at_1000": 0.9925, "recall_at_1": 0.99, "recall_at_3": 0.99, "recall_at_5": 1.0, "recall_at_10": 1.0, "recall_at_20": 1.0, "recall_at_100": 1.0, "recall_at_1000": 1.0, "precision_at_1": 0.99, "precision_at_3": 0.33, "precision_at_5": 0.2, "precision_at_10": 0.1, "precision_at_20": 0.05, "precision_at_100": 0.01, "precision_at_1000": 0.001, "mrr_at_1": 0.99, "mrr_at_3": 0.99, "mrr_at_5": 0.9925, "mrr_at_10": 0.9925, "mrr_at_20": 0.9925, "mrr_at_100": 0.9925, "mrr_at_1000": 0.9925, "naucs_at_1_max": 1.0, "naucs_at_1_std": 0.8692810457516276, "naucs_at_1_diff1": 0.8692810457516276, "naucs_at_3_max": 1.0, "naucs_at_3_std": 0.8692810457516356, "naucs_at_3_diff1": 0.8692810457516356, "naucs_at_5_max": 1.0, "naucs_at_5_std": 1.0, "naucs_at_5_diff1": 1.0, "naucs_at_10_max": 1.0, "naucs_at_10_std": 1.0, "naucs_at_10_diff1": 1.0, "naucs_at_20_max": 1.0, "naucs_at_20_std": 1.0, "naucs_at_20_diff1": 1.0, "naucs_at_100_max": NaN, "naucs_at_100_std": NaN, "naucs_at_100_diff1": NaN, "naucs_at_1000_max": NaN, "naucs_at_1000_std": NaN, "naucs_at_1000_diff1": NaN}, "syntheticDocQA_energy_test": {"ndcg_at_1": 0.94, "ndcg_at_3": 0.95631, "ndcg_at_5": 0.96448, "ndcg_at_10": 0.96448, "ndcg_at_20": 0.96448, "ndcg_at_100": 0.96645, "ndcg_at_1000": 0.96645, "map_at_1": 0.94, "map_at_3": 0.95167, "map_at_5": 0.95617, "map_at_10": 0.95617, "map_at_20": 0.95617, "map_at_100": 0.95647, "map_at_1000": 0.95647, "recall_at_1": 0.94, "recall_at_3": 0.97, "recall_at_5": 0.99, "recall_at_10": 0.99, "recall_at_20": 0.99, "recall_at_100": 1.0, "recall_at_1000": 1.0, "precision_at_1": 0.94, "precision_at_3": 0.32333, "precision_at_5": 0.198, "precision_at_10": 0.099, "precision_at_20": 0.0495, "precision_at_100": 0.01, "precision_at_1000": 0.001, "mrr_at_1": 0.95, "mrr_at_3": 0.9583333333333335, "mrr_at_5": 0.9628333333333334, "mrr_at_10": 0.9628333333333334, "mrr_at_20": 0.9628333333333334, "mrr_at_100": 0.9631458333333335, "mrr_at_1000": 0.9631458333333335, "naucs_at_1_max": 0.6098661686896992, "naucs_at_1_std": -0.6484593837534979, "naucs_at_1_diff1": 0.9782135076252712, "naucs_at_3_max": 0.7587924058512326, "naucs_at_3_std": -0.7791783380018689, "naucs_at_3_diff1": 1.0, "naucs_at_5_max": 1.0, "naucs_at_5_std": -1.1517273576097316, "naucs_at_5_diff1": 1.0, "naucs_at_10_max": 1.0, "naucs_at_10_std": -1.1517273576097316, "naucs_at_10_diff1": 1.0, "naucs_at_20_max": 1.0, "naucs_at_20_std": -1.1517273576097316, "naucs_at_20_diff1": 1.0, "naucs_at_100_max": NaN, "naucs_at_100_std": NaN, "naucs_at_100_diff1": NaN, "naucs_at_1000_max": NaN, "naucs_at_1000_std": NaN, "naucs_at_1000_diff1": NaN}, "syntheticDocQA_government_reports_test": {"ndcg_at_1": 0.92, "ndcg_at_3": 0.96917, "ndcg_at_5": 0.96917, "ndcg_at_10": 0.96917, "ndcg_at_20": 0.96917, "ndcg_at_100": 0.96917, "ndcg_at_1000": 0.96917, "map_at_1": 0.92, "map_at_3": 0.95833, "map_at_5": 0.95833, "map_at_10": 0.95833, "map_at_20": 0.95833, "map_at_100": 0.95833, "map_at_1000": 0.95833, "recall_at_1": 0.92, "recall_at_3": 1.0, "recall_at_5": 1.0, "recall_at_10": 1.0, "recall_at_20": 1.0, "recall_at_100": 1.0, "recall_at_1000": 1.0, "precision_at_1": 0.92, "precision_at_3": 0.33333, "precision_at_5": 0.2, "precision_at_10": 0.1, "precision_at_20": 0.05, "precision_at_100": 0.01, "precision_at_1000": 0.001, "mrr_at_1": 0.95, "mrr_at_3": 0.9733333333333333, "mrr_at_5": 0.9733333333333333, "mrr_at_10": 0.9733333333333333, "mrr_at_20": 0.9733333333333333, "mrr_at_100": 0.9733333333333333, "mrr_at_1000": 0.9733333333333333, "naucs_at_1_max": 0.4804505135387461, "naucs_at_1_std": 0.05508870214752533, "naucs_at_1_diff1": 0.8952497665732956, "naucs_at_3_max": 1.0, "naucs_at_3_std": 1.0, "naucs_at_3_diff1": 1.0, "naucs_at_5_max": 1.0, "naucs_at_5_std": 1.0, "naucs_at_5_diff1": 1.0, "naucs_at_10_max": 1.0, "naucs_at_10_std": 1.0, "naucs_at_10_diff1": 1.0, "naucs_at_20_max": 1.0, "naucs_at_20_std": 1.0, "naucs_at_20_diff1": 1.0, "naucs_at_100_max": NaN, "naucs_at_100_std": NaN, "naucs_at_100_diff1": NaN, "naucs_at_1000_max": NaN, "naucs_at_1000_std": NaN, "naucs_at_1000_diff1": NaN}, "syntheticDocQA_healthcare_industry_test": {"ndcg_at_1": 0.96, "ndcg_at_3": 0.98393, "ndcg_at_5": 0.98393, "ndcg_at_10": 0.98393, "ndcg_at_20": 0.98393, "ndcg_at_100": 0.98393, "ndcg_at_1000": 0.98393, "map_at_1": 0.96, "map_at_3": 0.97833, "map_at_5": 0.97833, "map_at_10": 0.97833, "map_at_20": 0.97833, "map_at_100": 0.97833, "map_at_1000": 0.97833, "recall_at_1": 0.96, "recall_at_3": 1.0, "recall_at_5": 1.0, "recall_at_10": 1.0, "recall_at_20": 1.0, "recall_at_100": 1.0, "recall_at_1000": 1.0, "precision_at_1": 0.96, "precision_at_3": 0.33333, "precision_at_5": 0.2, "precision_at_10": 0.1, "precision_at_20": 0.05, "precision_at_100": 0.01, "precision_at_1000": 0.001, "mrr_at_1": 0.97, "mrr_at_3": 0.9833333333333334, "mrr_at_5": 0.9833333333333334, "mrr_at_10": 0.9833333333333334, "mrr_at_20": 0.9833333333333334, "mrr_at_100": 0.9833333333333334, "mrr_at_1000": 0.9833333333333334, "naucs_at_1_max": 0.74439775910364, "naucs_at_1_std": -0.053688141923436454, "naucs_at_1_diff1": 0.96732026143791, "naucs_at_3_max": 1.0, "naucs_at_3_std": 1.0, "naucs_at_3_diff1": 1.0, "naucs_at_5_max": 1.0, "naucs_at_5_std": 1.0, "naucs_at_5_diff1": 1.0, "naucs_at_10_max": 1.0, "naucs_at_10_std": 1.0, "naucs_at_10_diff1": 1.0, "naucs_at_20_max": 1.0, "naucs_at_20_std": 1.0, "naucs_at_20_diff1": 1.0, "naucs_at_100_max": NaN, "naucs_at_100_std": NaN, "naucs_at_100_diff1": NaN, "naucs_at_1000_max": NaN, "naucs_at_1000_std": NaN, "naucs_at_1000_diff1": NaN}}

special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

tokenizer_config.json

@@ -0,0 +1,209 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "processor_class": "ColQwen2_5Processor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

training_config.yml

@@ -0,0 +1,72 @@
+config:
+  (): colpali_engine.trainer.colmodel_training.ColModelTrainingConfig
+  output_dir: !path ../../../models/colqwen2_5-3b-mixed-multiL-5ep-l-128
+  processor:
+    (): colpali_engine.utils.transformers_wrappers.AllPurposeWrapper
+    class_to_instanciate: !ext colpali_engine.models.ColQwen2_5Processor
+    pretrained_model_name_or_path:  "Qwen/Qwen2.5-VL-3B-Instruct" # "./models/paligemma-3b-mix-448"
+    # num_image_tokens: 2048
+    # max_length: 50
+
+  model:
+    (): colpali_engine.utils.transformers_wrappers.AllPurposeWrapper
+    class_to_instanciate: !ext colpali_engine.models.ColQwen2_5
+    pretrained_model_name_or_path: "Qwen/Qwen2.5-VL-3B-Instruct"
+    torch_dtype:  !ext torch.bfloat16
+    use_cache: false
+    attn_implementation: "flash_attention_2"
+#    device_map: "auto"
+#    quantization_config:
+#      (): transformers.BitsAndBytesConfig
+#      load_in_4bit: true
+#      bnb_4bit_quant_type: "nf4"
+#      bnb_4bit_compute_dtype:  "bfloat16"
+#      bnb_4bit_use_double_quant: true
+
+  dataset_loading_func: !ext colpali_engine.utils.dataset_transformation.load_mixed_multiL_train_set
+  eval_dataset_loader: !import ../data/test_data.yaml
+
+  # max_length: 50
+  run_eval: true
+  loss_func:
+    (): colpali_engine.loss.late_interaction_losses.ColbertPairwiseCELoss
+  tr_args:
+    (): transformers.training_args.TrainingArguments
+    output_dir: null
+    overwrite_output_dir: true
+    num_train_epochs: 1
+    per_device_train_batch_size: 128
+    gradient_checkpointing: true
+    gradient_checkpointing_kwargs: { "use_reentrant": false }
+    # gradient_checkpointing: true
+    # 6 x 8 gpus = 48 batch size
+    gradient_accumulation_steps: 2
+    per_device_eval_batch_size: 32
+    eval_strategy: "steps"
+    dataloader_num_workers: 8
+    # bf16: true
+    save_steps: 100
+    logging_steps: 1
+    eval_steps: 180
+    warmup_steps: 19
+    learning_rate: 2e-4
+    save_total_limit: 15
+    resume_from_checkpoint: false
+    optim: "paged_adamw_8bit"
+    # wandb logging
+    # wandb_project: "dolma"
+    run_name: "colqwen2_5-3b-mixed-multiL-5ep-l-128"
+    report_to: wandb
+
+
+  peft_config:
+    (): peft.LoraConfig
+    r: 128
+    lora_alpha: 128
+    lora_dropout: 0.1
+    init_lora_weights: "gaussian"
+    bias: "none"
+    task_type: "FEATURE_EXTRACTION"
+    target_modules: '(.*(model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)'
+    # target_modules: '(.*(language_model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)'
+