<!-- 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. -->
9b_idefics_lora-finetune_nycaptions
This model is a fine-tuned version of HuggingFaceM4/idefics-9b on the newyorker_caption_contest dataset. It achieves the following results on the evaluation set:
- Loss: 1.4041
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant_with_warmup
- lr_scheduler_warmup_steps: 100
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3.0629 | 0.34 | 50 | 2.6949 |
| 1.2509 | 0.68 | 100 | 1.1454 |
| 1.0642 | 1.02 | 150 | 1.0957 |
| 1.0314 | 1.36 | 200 | 1.0888 |
| 0.9441 | 1.7 | 250 | 1.0925 |
| 0.9214 | 2.04 | 300 | 1.1144 |
| 0.8548 | 2.38 | 350 | 1.1244 |
| 0.8407 | 2.72 | 400 | 1.1548 |
| 0.724 | 3.06 | 450 | 1.1988 |
| 0.6888 | 3.4 | 500 | 1.2206 |
| 0.65 | 3.74 | 550 | 1.2622 |
| 0.5636 | 4.08 | 600 | 1.3041 |
| 0.4913 | 4.42 | 650 | 1.3550 |
| 0.5128 | 4.76 | 700 | 1.4041 |
Framework versions
- Transformers 4.33.2
- Pytorch 2.0.1+cu117
- Datasets 2.14.5
- Tokenizers 0.13.3
