---
library_name: peft
license: gemma
base_model: google/paligemma-3b-pt-224
tags:
- generated_from_trainer
model-index:
- name: paligemma_ocr_final
  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. -->

# paligemma_ocr_final

This model is a fine-tuned version of [google/paligemma-3b-pt-224](https://huggingface.co/google/paligemma-3b-pt-224) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.2630

## 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: 2e-05
- 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 |
|:-------------:|:------:|:----:|:---------------:|
| 13.3076       | 0.2888 | 20   | 2.7240          |
| 9.9169        | 0.5776 | 40   | 2.3545          |
| 9.1021        | 0.8664 | 60   | 2.2845          |
| 9.0472        | 1.1552 | 80   | 2.2641          |
| 8.893         | 1.4440 | 100  | 2.2649          |
| 8.8682        | 1.7329 | 120  | 2.2630          |


### Framework versions

- PEFT 0.13.2
- Transformers 4.46.1
- Pytorch 2.4.1+cu124
- Datasets 2.19.1
- Tokenizers 0.20.1