metadata
library_name: peft
license: gemma
base_model: google/paligemma-3b-pt-224
tags:
- generated_from_trainer
model-index:
- name: paligemma_ocr_final
results: []
paligemma_ocr_final
This model is a fine-tuned version of google/paligemma-3b-pt-224 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.1325
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: 0.0001
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Use OptimizerNames.ADAMW_HF 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: 2
- num_epochs: 2
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 10.1617 | 0.2888 | 20 | 2.2262 |
| 8.9231 | 0.5776 | 40 | 2.2125 |
| 8.7636 | 0.8664 | 60 | 2.1961 |
| 8.4317 | 1.1552 | 80 | 2.1833 |
| 8.3217 | 1.4440 | 100 | 2.1497 |
| 8.165 | 1.7329 | 120 | 2.1325 |
Framework versions
- PEFT 0.13.2
- Transformers 4.46.1
- Pytorch 2.4.1+cu124
- Datasets 2.19.1
- Tokenizers 0.20.1