End of training

README.md

@@ -15,68 +15,47 @@ should probably proofread and complete it, then remove this comment. -->
 
 # finetune_colqwen2-v1.0
 
-This model is a fine-tuned version of [vidore/colqwen2-base](https://huggingface.co/vidore/colqwen2-base) on the custom dataset.
+This model is a fine-tuned version of [vidore/colqwen2-base](https://huggingface.co/vidore/colqwen2-base) on the None dataset.
+It achieves the following results on the evaluation set:
+- eval_loss: 0.4260
+- eval_model_preparation_time: 0.0093
+- eval_runtime: 188.0038
+- eval_samples_per_second: 0.532
+- eval_steps_per_second: 0.266
+- epoch: 0.4796
+- step: 100
 
 ## Model description
 
-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-VL-2B](https://huggingface.co/Qwen/Qwen2-VL-2B-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)
+More information needed
 
-This version is the untrained base version to guarantee deterministic projection layer initialization.
-<p align="center"><img width=800 src="https://github.com/illuin-tech/colpali/blob/main/assets/colpali_architecture.webp?raw=true"/></p>
+## Intended uses & limitations
 
+More information needed
 
-## Usage
+## Training and evaluation data
 
-```
-model = ColQwen2.from_pretrained(
-    'toxic-pandas/finetune_colqwen2-v1.0',
-    torch_dtype=torch.bfloat16,
-    device_map=device,
-)
-```
+More information needed
 
-## Limitations
+## Training procedure
 
-- 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.
-
-## Dataset
-
-With the help of the GT4-o mini model, a dataset was formed for the completion of the colqwen2-v1.0 model, containing the following fields:
-
-- document_filename: Filename of the document.
-- document_url: Original URL of the document.
-- search_query: The query used to fetch the document.
-- search_topic: Topic related to the document.
-- search_subtopic: Subtopic related to the document.
-- search_language: Language specified for the search.
-- search_filetype: Filetype filter applied during the search.
-- page_number: The page's number within the document.
-- page_description: A natural language description of the page.
-- page_language: Language used on the page.
-- page_contains_table: Boolean indicating the presence of tables.
-- page_contains_figure: Boolean indicating the presence of figures.
-- page_contains_paragraph: Boolean indicating the presence of paragraphs.
-- page_image: The image of the current page.
-- query_type: Type of query (e.g., Extractive, Open-ended, Boolean, Compare-contrast, Enumerative, Numerical).
-- query_answerability: Answerability level of the query (Fully answerable, Partially answerable, Unanswerable).
-- query_modality: Modality used for query generation.
-- query_language: Language of the query.
-- query_reasoning: Reasoning traces used in query generation.
-- query: The actual query text.
-- query_is_self_contained: Boolean indicating if the query is self-contained.
-- query_is_self_contained_reasoning: Reasoning traces for determining self-contained nature.
-- answer: Expected answer to the question from the "query" field.
-
-## Training hyperparameters
+### Training hyperparameters
 
 The following hyperparameters were used during training:
 - learning_rate: 3e-05
+- train_batch_size: 2
+- eval_batch_size: 2
+- seed: 42
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 16
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 100
 - num_epochs: 1
 
 ### Framework versions
 
-- Transformers 4.46.3
+- Transformers 4.46.3
+- Pytorch 2.5.1+cu124
+- Datasets 3.1.0
+- Tokenizers 0.20.3

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3bb816833141e593bd989adbf1a0446c738b2feaf5ad8750219420a3341ff6e2
-size 74018232
+oid sha256:737694a530fa391d7c203a966a1bb565333b58b2e44b34dbb569873213c40be9
+size 147877240

runs/Dec07_23-42-45_l-0037930f-7a7f-43a1-be61-b29649ff25d0/events.out.tfevents.1733619250.l-0037930f-7a7f-43a1-be61-b29649ff25d0.8134.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98bfe3d085b76891ff92ca67d1b1fe913cb121cd8cd9b1efe645b4da4e83971c
+size 419

runs/Dec08_01-50-09_l-0037930f-7a7f-43a1-be61-b29649ff25d0/events.out.tfevents.1733622806.l-0037930f-7a7f-43a1-be61-b29649ff25d0.14785.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e96a77d9965b71f6a40d6eab1ef4760136a7f168040cc9f6a1d78a6a4e9725f
+size 13450

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fec017bd2b6a69984ac01cda5d0f949d6d333672167fbaf8cc184975fbebeaec
+oid sha256:9998ff654976e43ce1429edcdbc9ebe6798f7e1651880877429a9fedf934186f
 size 5304