---
library_name: transformers
license: gemma
base_model: vidore/colpaligemma-3b-pt-448-base
tags:
- colpali
- generated_from_trainer
pipeline_tag: visual-document-retrieval
model-index:
- name: finetune_colpali_v1_2-german_ver3-4bit
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# finetune_colpali_v1_2-german_ver3-4bit

This model is a fine-tuned version of [vidore/colpaligemma-3b-pt-448-base](https://huggingface.co/vidore/colpaligemma-3b-pt-448-base) on the German_docx dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0815
- Model Preparation Time: 0.0061

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 4
- 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: 3

### Training results

| Training Loss | Epoch  | Step | Validation Loss | Model Preparation Time |
|:-------------:|:------:|:----:|:---------------:|:----------------------:|
| No log        | 0.0146 | 1    | 0.3622          | 0.0061                 |
| 1.6392        | 0.1460 | 10   | 0.3430          | 0.0061                 |
| 1.1999        | 0.2920 | 20   | 0.3049          | 0.0061                 |
| 1.2826        | 0.4380 | 30   | 0.2768          | 0.0061                 |
| 0.7583        | 0.5839 | 40   | 0.2492          | 0.0061                 |
| 0.603         | 0.7299 | 50   | 0.2258          | 0.0061                 |
| 1.014         | 0.8759 | 60   | 0.1958          | 0.0061                 |
| 0.8131        | 1.0219 | 70   | 0.1688          | 0.0061                 |
| 0.6346        | 1.1679 | 80   | 0.1591          | 0.0061                 |
| 0.5089        | 1.3139 | 90   | 0.1502          | 0.0061                 |
| 0.4616        | 1.4599 | 100  | 0.1341          | 0.0061                 |
| 0.4498        | 1.6058 | 110  | 0.1136          | 0.0061                 |
| 0.4422        | 1.7518 | 120  | 0.1062          | 0.0061                 |
| 0.3519        | 1.8978 | 130  | 0.0989          | 0.0061                 |
| 0.2382        | 2.0438 | 140  | 0.0925          | 0.0061                 |
| 0.242         | 2.1898 | 150  | 0.0894          | 0.0061                 |
| 0.3462        | 2.3358 | 160  | 0.0907          | 0.0061                 |
| 0.1371        | 2.4818 | 170  | 0.0862          | 0.0061                 |
| 0.2691        | 2.6277 | 180  | 0.0838          | 0.0061                 |
| 0.0869        | 2.7737 | 190  | 0.0833          | 0.0061                 |
| 0.3401        | 2.9197 | 200  | 0.0815          | 0.0061                 |


### Framework versions

- Transformers 4.46.1
- Pytorch 2.3.1
- Datasets 3.1.0
- Tokenizers 0.20.1