checkpoints
This model is a fine-tuned version of vidore/colpaligemma-3b-pt-448-base on the ferferefer/colpali_prueba dataset. It achieves the following results on the evaluation set:
- Loss: 0.0147
- Model Preparation Time: 0.0064
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 OptimizerNames.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: 1.5
Training results
| Training Loss | Epoch | Step | Validation Loss | Model Preparation Time |
|---|---|---|---|---|
| No log | 0.0034 | 1 | 0.0574 | 0.0064 |
| 0.0116 | 0.3404 | 100 | 0.0284 | 0.0064 |
| 0.0061 | 0.6809 | 200 | 0.0278 | 0.0064 |
| 0.0043 | 1.0204 | 300 | 0.0170 | 0.0064 |
| 0.0135 | 1.3609 | 400 | 0.0167 | 0.0064 |
Framework versions
- Transformers 4.47.1
- Pytorch 2.4.1+cu121
- Datasets 3.3.2
- Tokenizers 0.21.0
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API:
The model has no pipeline_tag.
Model tree for ferferefer/colpali_prueba_lora
Base model
google/paligemma-3b-pt-448
Finetuned
vidore/colpaligemma-3b-pt-448-base