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 +1,202 @@
+---
+base_model: vidore/colpaligemma-3b-pt-448-base
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.11.1

adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "./models/colpaligemma-3b-pt-448-base",
+  "bias": "none",
+  "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": 32,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "(.*(language_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:caed65068cae6d50e572d984914324a7d8a9360cdd7f4263ea82f1792614391f
+size 78625112

checkpoint-18000/README.md

@@ -0,0 +1,202 @@
+---
+base_model: ./models/colpaligemma-3b-pt-448-base
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.11.1

checkpoint-18000/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "./models/colpaligemma-3b-pt-448-base",
+  "bias": "none",
+  "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": 32,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "(.*(language_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-18000/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da7776b2c26c1a62f0bbc5421c7201d2e30f7a5b05c1d4623e0e1b0f75ae3f60
+size 78625112

checkpoint-18000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d104d7b059bb1f146e59e053261c664c659adb9ca34a3dc82dcd25d491adb0ea
+size 157385722

checkpoint-18000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83c65b52f653f8c152a0e5f0fe6aaf87b7eef99af937d84679e6fdfa74192435
+size 14244

checkpoint-18000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:099cb1a8ae8c67ac7d0a23b3d3641d0efb236ad5bf903c5a94993ec5b5058192
+size 1064

checkpoint-18000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5caac953cc790faad6d9fc47d5b6516882f63815bf7f5962c459a1fa37b07e8
+size 5048

git_hash.txt

@@ -0,0 +1 @@
+69293a9893f663429e172e512ee5908af0cffef5

preprocessor_config.json

@@ -0,0 +1,40 @@
+{
+  "_valid_processor_keys": [
+    "images",
+    "do_resize",
+    "size",
+    "resample",
+    "do_rescale",
+    "rescale_factor",
+    "do_normalize",
+    "image_mean",
+    "image_std",
+    "return_tensors",
+    "data_format",
+    "input_data_format",
+    "do_convert_rgb"
+  ],
+  "do_convert_rgb": null,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "SiglipImageProcessor",
+  "image_seq_length": 1024,
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "processor_class": "PaliGemmaProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 448,
+    "width": 448
+  }
+}

results.json

@@ -0,0 +1 @@
+{"validation_set": {"ndcg_at_1": 0.822, "ndcg_at_3": 0.86002, "ndcg_at_5": 0.87224, "ndcg_at_10": 0.88182, "ndcg_at_20": 0.88399, "ndcg_at_100": 0.89029, "ndcg_at_1000": 0.89199, "map_at_1": 0.822, "map_at_3": 0.851, "map_at_5": 0.8577, "map_at_10": 0.86159, "map_at_20": 0.86226, "map_at_100": 0.86313, "map_at_1000": 0.86322, "recall_at_1": 0.822, "recall_at_3": 0.886, "recall_at_5": 0.916, "recall_at_10": 0.946, "recall_at_20": 0.954, "recall_at_100": 0.988, "recall_at_1000": 1.0, "precision_at_1": 0.822, "precision_at_3": 0.29533, "precision_at_5": 0.1832, "precision_at_10": 0.0946, "precision_at_20": 0.0477, "precision_at_100": 0.00988, "precision_at_1000": 0.001, "mrr_at_1": 0.818, "mrr_at_3": 0.849, "mrr_at_5": 0.8554999999999998, "mrr_at_10": 0.8591857142857141, "mrr_at_20": 0.8603808764335078, "mrr_at_100": 0.8610268592085819, "mrr_at_1000": 0.8611122735411757, "naucs_at_1_max": 0.2153985688690058, "naucs_at_1_std": -0.8663183431820963, "naucs_at_1_diff1": 0.9298345185234652, "naucs_at_3_max": 0.10585679324754012, "naucs_at_3_std": -1.0737759501889552, "naucs_at_3_diff1": 0.9178001231957661, "naucs_at_5_max": 0.024365301676223024, "naucs_at_5_std": -1.1983460050687003, "naucs_at_5_diff1": 0.9034613845538216, "naucs_at_10_max": -0.10167029774873008, "naucs_at_10_std": -1.3703184977694771, "naucs_at_10_diff1": 0.9105197634609385, "naucs_at_20_max": -0.1524174887346353, "naucs_at_20_std": -1.4552226687776701, "naucs_at_20_diff1": 0.9264198432996398, "naucs_at_100_max": -0.33800186741361815, "naucs_at_100_std": -1.7399626517273559, "naucs_at_100_diff1": 0.931917211328972, "naucs_at_1000_max": 1.0, "naucs_at_1000_std": 1.0, "naucs_at_1000_diff1": 1.0}, "syntheticDocQA_energy": {"ndcg_at_1": 0.89, "ndcg_at_3": 0.93917, "ndcg_at_5": 0.93917, "ndcg_at_10": 0.93917, "ndcg_at_20": 0.94156, "ndcg_at_100": 0.94521, "ndcg_at_1000": 0.94521, "map_at_1": 0.89, "map_at_3": 0.92833, "map_at_5": 0.92833, "map_at_10": 0.92833, "map_at_20": 0.92892, "map_at_100": 0.92938, "map_at_1000": 0.92938, "recall_at_1": 0.89, "recall_at_3": 0.97, "recall_at_5": 0.97, "recall_at_10": 0.97, "recall_at_20": 0.98, "recall_at_100": 1.0, "recall_at_1000": 1.0, "precision_at_1": 0.89, "precision_at_3": 0.32333, "precision_at_5": 0.194, "precision_at_10": 0.097, "precision_at_20": 0.049, "precision_at_100": 0.01, "precision_at_1000": 0.001, "mrr_at_1": 0.9, "mrr_at_3": 0.9333333333333332, "mrr_at_5": 0.9333333333333332, "mrr_at_10": 0.9333333333333332, "mrr_at_20": 0.9338888888888889, "mrr_at_100": 0.9343481252149982, "mrr_at_1000": 0.9343481252149982, "naucs_at_1_max": 0.28399622025599114, "naucs_at_1_std": -0.7361051456060473, "naucs_at_1_diff1": 0.9759470835838844, "naucs_at_3_max": 0.7424525365701778, "naucs_at_3_std": -1.3478057889822568, "naucs_at_3_diff1": 1.0, "naucs_at_5_max": 0.742452536570183, "naucs_at_5_std": -1.3478057889822557, "naucs_at_5_diff1": 1.0, "naucs_at_10_max": 0.742452536570183, "naucs_at_10_std": -1.3478057889822557, "naucs_at_10_diff1": 1.0, "naucs_at_20_max": 0.9346405228758136, "naucs_at_20_std": -1.1517273576097127, "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": {"ndcg_at_1": 0.92, "ndcg_at_3": 0.95024, "ndcg_at_5": 0.95454, "ndcg_at_10": 0.96077, "ndcg_at_20": 0.96077, "ndcg_at_100": 0.96077, "ndcg_at_1000": 0.96077, "map_at_1": 0.92, "map_at_3": 0.94333, "map_at_5": 0.94583, "map_at_10": 0.94826, "map_at_20": 0.94826, "map_at_100": 0.94826, "map_at_1000": 0.94826, "recall_at_1": 0.92, "recall_at_3": 0.97, "recall_at_5": 0.98, "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": 0.32333, "precision_at_5": 0.196, "precision_at_10": 0.1, "precision_at_20": 0.05, "precision_at_100": 0.01, "precision_at_1000": 0.001, "mrr_at_1": 0.93, "mrr_at_3": 0.9483333333333333, "mrr_at_5": 0.9508333333333333, "mrr_at_10": 0.9533730158730158, "mrr_at_20": 0.9533730158730158, "mrr_at_100": 0.9533730158730158, "mrr_at_1000": 0.9533730158730158, "naucs_at_1_max": 0.5720121381886089, "naucs_at_1_std": -0.3297152194211025, "naucs_at_1_diff1": 0.9509803921568633, "naucs_at_3_max": 0.33022097727980126, "naucs_at_3_std": -0.9192343604108318, "naucs_at_3_diff1": 0.9564270152505466, "naucs_at_5_max": 0.21825396825397442, "naucs_at_5_std": -1.7399626517273414, "naucs_at_5_diff1": 0.9346405228758136, "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_artificial_intelligence_test": {"ndcg_at_1": 0.95, "ndcg_at_3": 0.97393, "ndcg_at_5": 0.97393, "ndcg_at_10": 0.97393, "ndcg_at_20": 0.97393, "ndcg_at_100": 0.97617, "ndcg_at_1000": 0.97617, "map_at_1": 0.95, "map_at_3": 0.96833, "map_at_5": 0.96833, "map_at_10": 0.96833, "map_at_20": 0.96833, "map_at_100": 0.96881, "map_at_1000": 0.96881, "recall_at_1": 0.95, "recall_at_3": 0.99, "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": 0.95, "precision_at_3": 0.33, "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": 0.9683333333333333, "mrr_at_5": 0.9683333333333333, "mrr_at_10": 0.9683333333333333, "mrr_at_20": 0.968859649122807, "mrr_at_100": 0.968859649122807, "mrr_at_1000": 0.968859649122807, "naucs_at_1_max": 0.5066293183940227, "naucs_at_1_std": -0.6262371615312773, "naucs_at_1_diff1": 0.9444444444444438, "naucs_at_3_max": 0.8692810457516356, "naucs_at_3_std": -1.7399626517274398, "naucs_at_3_diff1": 1.0, "naucs_at_5_max": 0.8692810457516413, "naucs_at_5_std": -1.7399626517273863, "naucs_at_5_diff1": 1.0, "naucs_at_10_max": 0.8692810457516413, "naucs_at_10_std": -1.7399626517273863, "naucs_at_10_diff1": 1.0, "naucs_at_20_max": 0.8692810457516413, "naucs_at_20_std": -1.7399626517273863, "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": {"ndcg_at_1": 0.86, "ndcg_at_3": 0.93047, "ndcg_at_5": 0.93478, "ndcg_at_10": 0.93834, "ndcg_at_20": 0.93834, "ndcg_at_100": 0.93834, "ndcg_at_1000": 0.93834, "map_at_1": 0.86, "map_at_3": 0.91333, "map_at_5": 0.91583, "map_at_10": 0.9175, "map_at_20": 0.9175, "map_at_100": 0.9175, "map_at_1000": 0.9175, "recall_at_1": 0.86, "recall_at_3": 0.98, "recall_at_5": 0.99, "recall_at_10": 1.0, "recall_at_20": 1.0, "recall_at_100": 1.0, "recall_at_1000": 1.0, "precision_at_1": 0.86, "precision_at_3": 0.32667, "precision_at_5": 0.198, "precision_at_10": 0.1, "precision_at_20": 0.05, "precision_at_100": 0.01, "precision_at_1000": 0.001, "mrr_at_1": 0.86, "mrr_at_3": 0.9166666666666665, "mrr_at_5": 0.9166666666666665, "mrr_at_10": 0.9183333333333333, "mrr_at_20": 0.9183333333333333, "mrr_at_100": 0.9183333333333333, "mrr_at_1000": 0.9183333333333333, "naucs_at_1_max": 0.5858043508382976, "naucs_at_1_std": 0.34131217957599924, "naucs_at_1_diff1": 0.8498337259249, "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}, "infovqa_subsampled": {"ndcg_at_1": 0.752, "ndcg_at_3": 0.80538, "ndcg_at_5": 0.81718, "ndcg_at_10": 0.82621, "ndcg_at_20": 0.83154, "ndcg_at_100": 0.84329, "ndcg_at_1000": 0.84596, "map_at_1": 0.752, "map_at_3": 0.79267, "map_at_5": 0.79937, "map_at_10": 0.80308, "map_at_20": 0.80469, "map_at_100": 0.80643, "map_at_1000": 0.80655, "recall_at_1": 0.752, "recall_at_3": 0.842, "recall_at_5": 0.87, "recall_at_10": 0.898, "recall_at_20": 0.918, "recall_at_100": 0.98, "recall_at_1000": 1.0, "precision_at_1": 0.752, "precision_at_3": 0.28067, "precision_at_5": 0.174, "precision_at_10": 0.0898, "precision_at_20": 0.0459, "precision_at_100": 0.0098, "precision_at_1000": 0.001, "mrr_at_1": 0.752, "mrr_at_3": 0.7933333333333331, "mrr_at_5": 0.7995333333333332, "mrr_at_10": 0.803242857142857, "mrr_at_20": 0.8049362776112772, "mrr_at_100": 0.8065472483651968, "mrr_at_1000": 0.8066659177387367, "naucs_at_1_max": 0.5521093170633048, "naucs_at_1_std": -0.07785006404638299, "naucs_at_1_diff1": 0.8997463426144405, "naucs_at_3_max": 0.6002557449502772, "naucs_at_3_std": -0.012867943736110006, "naucs_at_3_diff1": 0.85642900594669, "naucs_at_5_max": 0.6126712582389099, "naucs_at_5_std": 0.006642968229280956, "naucs_at_5_diff1": 0.8472635710582841, "naucs_at_10_max": 0.5868009030671214, "naucs_at_10_std": -0.006387547952498705, "naucs_at_10_diff1": 0.8159541858629611, "naucs_at_20_max": 0.6058049235954548, "naucs_at_20_std": 0.0355150190157345, "naucs_at_20_diff1": 0.8257953587939248, "naucs_at_100_max": 0.7703081232492869, "naucs_at_100_std": -0.2799253034547204, "naucs_at_100_diff1": 0.8830532212885076, "naucs_at_1000_max": 1.0, "naucs_at_1000_std": 1.0, "naucs_at_1000_diff1": 1.0}, "docvqa_subsampled": {"ndcg_at_1": 0.438, "ndcg_at_3": 0.50464, "ndcg_at_5": 0.52539, "ndcg_at_10": 0.54515, "ndcg_at_20": 0.55891, "ndcg_at_100": 0.58614, "ndcg_at_1000": 0.60242, "map_at_1": 0.438, "map_at_3": 0.48767, "map_at_5": 0.49927, "map_at_10": 0.50725, "map_at_20": 0.51082, "map_at_100": 0.51421, "map_at_1000": 0.51485, "recall_at_1": 0.438, "recall_at_3": 0.554, "recall_at_5": 0.604, "recall_at_10": 0.666, "recall_at_20": 0.722, "recall_at_100": 0.874, "recall_at_1000": 1.0, "precision_at_1": 0.438, "precision_at_3": 0.18467, "precision_at_5": 0.1208, "precision_at_10": 0.0666, "precision_at_20": 0.0361, "precision_at_100": 0.00874, "precision_at_1000": 0.001, "mrr_at_1": 0.438, "mrr_at_3": 0.4856666666666668, "mrr_at_5": 0.4982666666666667, "mrr_at_10": 0.5064968253968254, "mrr_at_20": 0.5101578425616352, "mrr_at_100": 0.5137831471648001, "mrr_at_1000": 0.5143743988006569, "naucs_at_1_max": 0.3411814722367848, "naucs_at_1_std": 0.19351178785768525, "naucs_at_1_diff1": 0.8626562650787815, "naucs_at_3_max": 0.25447268440039383, "naucs_at_3_std": 0.23527321240702792, "naucs_at_3_diff1": 0.7822508539209857, "naucs_at_5_max": 0.21922361789517097, "naucs_at_5_std": 0.24044767820737398, "naucs_at_5_diff1": 0.7541297238480713, "naucs_at_10_max": 0.1511643138558891, "naucs_at_10_std": 0.2974449919221844, "naucs_at_10_diff1": 0.7404250173815784, "naucs_at_20_max": 0.07575239988792164, "naucs_at_20_std": 0.40653199402762796, "naucs_at_20_diff1": 0.7356189043230615, "naucs_at_100_max": -0.06645545519625845, "naucs_at_100_std": 0.6961981935326662, "naucs_at_100_diff1": 0.6942713965850758, "naucs_at_1000_max": 1.0, "naucs_at_1000_std": 1.0, "naucs_at_1000_diff1": 1.0}, "arxivqa_subsampled": {"ndcg_at_1": 0.716, "ndcg_at_3": 0.76712, "ndcg_at_5": 0.78742, "ndcg_at_10": 0.80395, "ndcg_at_20": 0.81394, "ndcg_at_100": 0.82137, "ndcg_at_1000": 0.82402, "map_at_1": 0.716, "map_at_3": 0.755, "map_at_5": 0.7661, "map_at_10": 0.77314, "map_at_20": 0.77581, "map_at_100": 0.77698, "map_at_1000": 0.77709, "recall_at_1": 0.716, "recall_at_3": 0.802, "recall_at_5": 0.852, "recall_at_10": 0.902, "recall_at_20": 0.942, "recall_at_100": 0.98, "recall_at_1000": 1.0, "precision_at_1": 0.716, "precision_at_3": 0.26733, "precision_at_5": 0.1704, "precision_at_10": 0.0902, "precision_at_20": 0.0471, "precision_at_100": 0.0098, "precision_at_1000": 0.001, "mrr_at_1": 0.716, "mrr_at_3": 0.7533333333333331, "mrr_at_5": 0.764733333333333, "mrr_at_10": 0.7717634920634919, "mrr_at_20": 0.7747616656992816, "mrr_at_100": 0.7758456524880377, "mrr_at_1000": 0.7759607523201341, "naucs_at_1_max": 0.7289289296839049, "naucs_at_1_std": -0.07936298430464478, "naucs_at_1_diff1": 0.9118126211468693, "naucs_at_3_max": 0.7440127459185849, "naucs_at_3_std": -0.039296554366786994, "naucs_at_3_diff1": 0.8523732214704197, "naucs_at_5_max": 0.7376670092497422, "naucs_at_5_std": -0.1464476770232166, "naucs_at_5_diff1": 0.8692584394023237, "naucs_at_10_max": 0.7347605708950243, "naucs_at_10_std": -0.1508793993787991, "naucs_at_10_diff1": 0.8310752872577601, "naucs_at_20_max": 0.710067935220066, "naucs_at_20_std": -0.24474387456132013, "naucs_at_20_diff1": 0.8414308252036463, "naucs_at_100_max": 0.852007469654521, "naucs_at_100_std": 0.2826797385620789, "naucs_at_100_diff1": 0.8095238095237989, "naucs_at_1000_max": 1.0, "naucs_at_1000_std": 1.0, "naucs_at_1000_diff1": 1.0}, "tabfquad_subsampled": {"ndcg_at_1": 0.82857, "ndcg_at_3": 0.87759, "ndcg_at_5": 0.89097, "ndcg_at_10": 0.90108, "ndcg_at_20": 0.90465, "ndcg_at_100": 0.90682, "ndcg_at_1000": 0.90682, "map_at_1": 0.82857, "map_at_3": 0.86607, "map_at_5": 0.87357, "map_at_10": 0.87758, "map_at_20": 0.87853, "map_at_100": 0.87889, "map_at_1000": 0.87889, "recall_at_1": 0.82857, "recall_at_3": 0.91071, "recall_at_5": 0.94286, "recall_at_10": 0.975, "recall_at_20": 0.98929, "recall_at_100": 1.0, "recall_at_1000": 1.0, "precision_at_1": 0.82857, "precision_at_3": 0.30357, "precision_at_5": 0.18857, "precision_at_10": 0.0975, "precision_at_20": 0.04946, "precision_at_100": 0.01, "precision_at_1000": 0.001, "mrr_at_1": 0.8285714285714286, "mrr_at_3": 0.8648809523809524, "mrr_at_5": 0.873095238095238, "mrr_at_10": 0.8768296485260769, "mrr_at_20": 0.8780735435199719, "mrr_at_100": 0.8784422839749192, "mrr_at_1000": 0.8784422839749192, "naucs_at_1_max": 0.5117778966425275, "naucs_at_1_std": 0.18279912771560283, "naucs_at_1_diff1": 0.8180752139565501, "naucs_at_3_max": 0.5747899159663853, "naucs_at_3_std": 0.3599813258636801, "naucs_at_3_diff1": 0.799458450046687, "naucs_at_5_max": 0.7672152194211035, "naucs_at_5_std": 0.5452264239028954, "naucs_at_5_diff1": 0.8789098972922513, "naucs_at_10_max": 0.9176337201547285, "naucs_at_10_std": 0.8259303721488543, "naucs_at_10_diff1": 0.8522742430305419, "naucs_at_20_max": 0.8513849984438296, "naucs_at_20_std": 0.7424525365701908, "naucs_at_20_diff1": 0.9564270152505505, "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": {"ndcg_at_1": 0.546, "ndcg_at_3": 0.65066, "ndcg_at_5": 0.67483, "ndcg_at_10": 0.69932, "ndcg_at_20": 0.71343, "ndcg_at_100": 0.72476, "ndcg_at_1000": 0.72826, "map_at_1": 0.546, "map_at_3": 0.62528, "map_at_5": 0.63862, "map_at_10": 0.64883, "map_at_20": 0.65269, "map_at_100": 0.65429, "map_at_1000": 0.65446, "recall_at_1": 0.546, "recall_at_3": 0.72399, "recall_at_5": 0.78292, "recall_at_10": 0.85809, "recall_at_20": 0.91401, "recall_at_100": 0.97474, "recall_at_1000": 1.0, "precision_at_1": 0.546, "precision_at_3": 0.24133, "precision_at_5": 0.15658, "precision_at_10": 0.08581, "precision_at_20": 0.0457, "precision_at_100": 0.00975, "precision_at_1000": 0.001, "mrr_at_1": 0.5460012026458209, "mrr_at_3": 0.6248747243936668, "mrr_at_5": 0.6380737622770104, "mrr_at_10": 0.6487178936517485, "mrr_at_20": 0.6526362286471448, "mrr_at_100": 0.6541821288626329, "mrr_at_1000": 0.6543520127179482, "naucs_at_1_max": 0.24133986493040024, "naucs_at_1_std": -0.12843954118121406, "naucs_at_1_diff1": 0.7071602689895572, "naucs_at_3_max": 0.26030616268930545, "naucs_at_3_std": -0.11795132646338957, "naucs_at_3_diff1": 0.5997176411168043, "naucs_at_5_max": 0.2730219894711174, "naucs_at_5_std": -0.10596692743471213, "naucs_at_5_diff1": 0.5663532060729257, "naucs_at_10_max": 0.36225362853714277, "naucs_at_10_std": 0.02304171698956151, "naucs_at_10_diff1": 0.5517101887320957, "naucs_at_20_max": 0.4125290416811556, "naucs_at_20_std": 0.167993127442173, "naucs_at_20_diff1": 0.5211985072803259, "naucs_at_100_max": 0.3575181920149435, "naucs_at_100_std": 0.2654763254976142, "naucs_at_100_diff1": 0.5593828030956866, "naucs_at_1000_max": NaN, "naucs_at_1000_std": NaN, "naucs_at_1000_diff1": NaN}, "shift_project": {"ndcg_at_1": 0.64, "ndcg_at_3": 0.79488, "ndcg_at_5": 0.8078, "ndcg_at_10": 0.81451, "ndcg_at_20": 0.8173, "ndcg_at_100": 0.82289, "ndcg_at_1000": 0.82437, "map_at_1": 0.64, "map_at_3": 0.75833, "map_at_5": 0.76583, "map_at_10": 0.76875, "map_at_20": 0.76966, "map_at_100": 0.77041, "map_at_1000": 0.7705, "recall_at_1": 0.64, "recall_at_3": 0.9, "recall_at_5": 0.93, "recall_at_10": 0.95, "recall_at_20": 0.96, "recall_at_100": 0.99, "recall_at_1000": 1.0, "precision_at_1": 0.64, "precision_at_3": 0.3, "precision_at_5": 0.186, "precision_at_10": 0.095, "precision_at_20": 0.048, "precision_at_100": 0.0099, "precision_at_1000": 0.001, "mrr_at_1": 0.64, "mrr_at_3": 0.7616666666666667, "mrr_at_5": 0.7666666666666666, "mrr_at_10": 0.7711111111111112, "mrr_at_20": 0.7711111111111112, "mrr_at_100": 0.7718950246765373, "mrr_at_1000": 0.7719893642991787, "naucs_at_1_max": 0.11254630220973243, "naucs_at_1_std": -0.10817154170392139, "naucs_at_1_diff1": 0.747333631370545, "naucs_at_3_max": 0.18342670401493752, "naucs_at_3_std": -0.27189542483660173, "naucs_at_3_diff1": 0.6793183940242762, "naucs_at_5_max": -0.14785914365746142, "naucs_at_5_std": -0.6669334400426802, "naucs_at_5_diff1": 0.7159530478858213, "naucs_at_10_max": -0.2681605975723533, "naucs_at_10_std": -0.47301587301586673, "naucs_at_10_diff1": 0.6578898225957086, "naucs_at_20_max": -0.5852007469654444, "naucs_at_20_std": -0.8085901027077429, "naucs_at_20_diff1": 0.6050420168067251, "naucs_at_100_max": 0.7222222222222041, "naucs_at_100_std": -0.17133520074697067, "naucs_at_100_diff1": 1.0, "naucs_at_1000_max": NaN, "naucs_at_1000_std": NaN, "naucs_at_1000_diff1": NaN}}

special_tokens_map.json

@@ -0,0 +1,39 @@
+{
+  "additional_special_tokens": [
+    {
+      "content": "<image>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    }
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "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:ffd310e50986db7a039948ab83441d612689e7f989198e31b5c8984ca458adf6
+size 17763459

training_config.yml

@@ -0,0 +1,61 @@
+config:
+  (): colpali_engine.utils.train_colpali_engine_models.ColModelTrainingConfig
+  output_dir: !path ../../../models/right_pad/train_colpali_hardneg_long
+  processor:
+    () : colpali_engine.utils.wrapper.AutoProcessorWrapper
+    pretrained_model_name_or_path:  "./models/colpaligemma-3b-pt-448-base" # "./models/paligemma-3b-mix-448"
+    max_length: 50
+  model:
+    (): colpali_engine.utils.wrapper.AllPurposeWrapper
+    class_to_instanciate: !ext colpali_engine.models.paligemma_colbert_architecture.ColPali
+    pretrained_model_name_or_path: "./models/colpaligemma-3b-pt-448-base"
+    torch_dtype:  !ext torch.bfloat16
+#    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_train_set_ir_negs
+  eval_dataset_loader: !import ../data/test_data.yaml
+
+  max_length: 50
+  run_eval: true
+  add_suffix: true
+  loss_func:
+    (): colpali_engine.loss.colbert_loss.ColbertPairwiseNegativeCELoss
+    in_batch_term: true
+  tr_args:
+    (): transformers.training_args.TrainingArguments
+    output_dir: null
+    overwrite_output_dir: true
+    num_train_epochs: 5
+    per_device_train_batch_size: 4
+    # 6 x 8 gpus = 48 batch size
+    # gradient_accumulation_steps: 4
+    per_device_eval_batch_size: 4
+    eval_strategy: "steps"
+    # dataloader_num_workers: 8
+    # bf16: true
+    save_steps: 500
+    logging_steps: 10
+    eval_steps: 50
+    warmup_steps: 1000
+    learning_rate: 5e-5
+    save_total_limit: 1
+    resume_from_checkpoint: true
+    # optim: "paged_adamw_8bit"
+
+  peft_config:
+    (): peft.LoraConfig
+    r: 32
+    lora_alpha: 32
+    lora_dropout: 0.1
+    init_lora_weights: "gaussian"
+    bias: "none"
+    task_type: "FEATURE_EXTRACTION"
+    target_modules: '(.*(language_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).*$|.*(custom_text_proj).*$)'
+