metadata
library_name: peft
license: gemma
base_model: google/paligemma2-28b-pt-448
tags:
- generated_from_trainer
datasets:
- imagefolder
metrics:
- accuracy
model-index:
- name: PG2-28b-pt-448-COND_GEN
results: []
PG2-28b-pt-448-COND_GEN
This model is a fine-tuned version of google/paligemma2-28b-pt-448 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.0000
- Accuracy: 0.0
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
- 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: 2
- num_epochs: 2
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.4437 | 0.1001 | 200 | 0.0036 | 1.0 |
| 0.3054 | 0.2001 | 400 | 0.0002 | 1.0 |
| 0.3504 | 0.3002 | 600 | 0.0001 | 1.0 |
| 0.2364 | 0.4002 | 800 | 0.0001 | 0.0 |
| 0.257 | 0.5003 | 1000 | 0.0001 | 0.0 |
| 0.3007 | 0.6003 | 1200 | 0.0002 | 0.0 |
| 0.2468 | 0.7004 | 1400 | 0.0001 | 0.0 |
| 0.278 | 0.8004 | 1600 | 0.0001 | 0.0 |
| 0.2576 | 0.9005 | 1800 | 0.0001 | 0.0 |
| 0.3052 | 1.0005 | 2000 | 0.0005 | 0.0 |
| 0.0982 | 1.1006 | 2200 | 0.0001 | 0.0 |
| 0.103 | 1.2006 | 2400 | 0.0001 | 0.0 |
| 0.1191 | 1.3007 | 2600 | 0.0001 | 0.0 |
| 0.1137 | 1.4007 | 2800 | 0.0001 | 1.0 |
| 0.1567 | 1.5008 | 3000 | 0.0001 | 0.0 |
| 0.0758 | 1.6008 | 3200 | 0.0000 | 0.0 |
| 0.0969 | 1.7009 | 3400 | 0.0001 | 0.0 |
| 0.1034 | 1.8009 | 3600 | 0.0000 | 0.0 |
| 0.0513 | 1.9010 | 3800 | 0.0000 | 0.0 |
Framework versions
- PEFT 0.14.0
- Transformers 4.48.0
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0