<!-- 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. -->
git-large-r-coco-childrensbooks
This model is a fine-tuned version of microsoft/git-large-r-coco on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.1193
- Wer Score: 2.9897
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
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer Score |
|---|---|---|---|---|
| 6.8915 | 4.35 | 50 | 3.7130 | 10.8615 |
| 1.541 | 8.7 | 100 | 0.2006 | 0.7385 |
| 0.0556 | 13.04 | 150 | 0.1085 | 2.4906 |
| 0.0133 | 17.39 | 200 | 0.1258 | 5.4701 |
| 0.0052 | 21.74 | 250 | 0.1167 | 4.3573 |
| 0.0016 | 26.09 | 300 | 0.1184 | 3.7897 |
| 0.0013 | 30.43 | 350 | 0.1163 | 3.3932 |
| 0.0009 | 34.78 | 400 | 0.1177 | 3.1009 |
| 0.0008 | 39.13 | 450 | 0.1191 | 2.9778 |
| 0.0007 | 43.48 | 500 | 0.1192 | 2.9932 |
| 0.0007 | 47.83 | 550 | 0.1193 | 2.9897 |
Framework versions
- Transformers 4.33.1
- Pytorch 1.11.0a0+17540c5
- Datasets 2.14.5
- Tokenizers 0.13.3
