metadata
library_name: peft
license: gemma
base_model: vidore/colpaligemma-3b-pt-448-base
tags:
- colpali-finetuned
- generated_from_trainer
model-index:
- name: version1_2-8epochs
results: []
version1_2-8epochs
This model is a fine-tuned version of vidore/colpaligemma-3b-pt-448-base on the 3sara/validated_colpali_italian_documents_with_images dataset. It achieves the following results on the evaluation set:
- Loss: 0.2972
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: 8
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 0.0103 | 1 | 0.3841 |
| 0.135 | 1.0205 | 100 | 0.2809 |
| 0.0817 | 2.0410 | 200 | 0.3178 |
| 0.0183 | 3.0923 | 300 | 0.3017 |
| 0.0746 | 4.1128 | 400 | 0.2900 |
| 0.1553 | 5.1538 | 500 | 0.3342 |
| 0.022 | 6.1744 | 600 | 0.2990 |
| 0.0446 | 7.2154 | 700 | 0.2972 |
Framework versions
- PEFT 0.15.2
- Transformers 4.51.3
- Pytorch 2.7.1+cu126
- Datasets 3.6.0
- Tokenizers 0.21.1