|
|
--- |
|
|
library_name: peft |
|
|
license: gemma |
|
|
base_model: google/paligemma-3b-pt-224 |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
model-index: |
|
|
- name: pali_191805 |
|
|
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. --> |
|
|
|
|
|
# pali_191805 |
|
|
|
|
|
This model is a fine-tuned version of [google/paligemma-3b-pt-224](https://huggingface.co/google/paligemma-3b-pt-224) on the None dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.8563 |
|
|
|
|
|
## 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: 2 |
|
|
- eval_batch_size: 2 |
|
|
- seed: 42 |
|
|
- gradient_accumulation_steps: 8 |
|
|
- 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: 2 |
|
|
- num_epochs: 5 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:------:|:----:|:---------------:| |
|
|
| 18.2133 | 0.0444 | 50 | 1.7766 | |
|
|
| 11.9694 | 0.0889 | 100 | 1.3043 | |
|
|
| 9.7625 | 0.1333 | 150 | 1.1940 | |
|
|
| 9.0576 | 0.1778 | 200 | 1.1325 | |
|
|
| 9.3286 | 0.2222 | 250 | 1.0906 | |
|
|
| 8.5435 | 0.2667 | 300 | 1.0586 | |
|
|
| 8.2508 | 0.3111 | 350 | 1.0357 | |
|
|
| 8.3642 | 0.3556 | 400 | 1.0151 | |
|
|
| 8.0343 | 0.4 | 450 | 0.9982 | |
|
|
| 8.1537 | 0.4444 | 500 | 0.9818 | |
|
|
| 7.6705 | 0.4889 | 550 | 0.9672 | |
|
|
| 7.6794 | 0.5333 | 600 | 0.9557 | |
|
|
| 7.3842 | 0.5778 | 650 | 0.9470 | |
|
|
| 7.5392 | 0.6222 | 700 | 0.9343 | |
|
|
| 7.3926 | 0.6667 | 750 | 0.9233 | |
|
|
| 7.5391 | 0.7111 | 800 | 0.9141 | |
|
|
| 7.3299 | 0.7556 | 850 | 0.9053 | |
|
|
| 7.3423 | 0.8 | 900 | 0.8974 | |
|
|
| 7.4747 | 0.8444 | 950 | 0.8911 | |
|
|
| 7.252 | 0.8889 | 1000 | 0.8832 | |
|
|
| 7.1392 | 0.9333 | 1050 | 0.8783 | |
|
|
| 6.9769 | 0.9778 | 1100 | 0.8719 | |
|
|
| 7.0285 | 1.0222 | 1150 | 0.8665 | |
|
|
| 6.8336 | 1.0667 | 1200 | 0.8613 | |
|
|
| 6.748 | 1.1111 | 1250 | 0.8563 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- PEFT 0.14.0 |
|
|
- Transformers 4.51.3 |
|
|
- Pytorch 2.6.0+cu124 |
|
|
- Datasets 3.6.0 |
|
|
- Tokenizers 0.21.1 |
