idefics-9b-instruct-ft-instruct-compact
This model is a fine-tuned version of HuggingFaceM4/idefics-9b-instruct on an unknown dataset.
It achieves the following results on the evaluation set:
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: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss |
Epoch |
Step |
Validation Loss |
| 3.9471 |
0.18 |
25 |
3.9005 |
| 3.513 |
0.36 |
50 |
3.4236 |
| 2.8952 |
0.54 |
75 |
2.5822 |
| 1.9127 |
0.73 |
100 |
1.7289 |
| 1.2521 |
0.91 |
125 |
1.0073 |
| 0.7576 |
1.09 |
150 |
0.7103 |
| 0.6537 |
1.27 |
175 |
0.6171 |
| 0.5969 |
1.45 |
200 |
0.5845 |
| 0.5778 |
1.63 |
225 |
0.5681 |
| 0.5663 |
1.82 |
250 |
0.5559 |
| 0.558 |
2.0 |
275 |
0.5471 |
| 0.5595 |
2.18 |
300 |
0.5409 |
| 0.5417 |
2.36 |
325 |
0.5367 |
| 0.5422 |
2.54 |
350 |
0.5341 |
| 0.54 |
2.72 |
375 |
0.5321 |
| 0.5474 |
2.91 |
400 |
0.5312 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.0.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0
