Downloads · 30 days
7
2% of all-time downloads
vit-base_rvl_cdip_symce
jordyvl/vit-base_rvl_cdip_symce
vit-base_rvl_cdip_symce is a image classification model from jordyvl. Use it when you need a label for an image. It is set up for transformers. The card lists the license as apache-2.0.
should probably proofread and complete it, then remove this comment. --
All-time downloads
361
Public
Repo size
687 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin343 MB · 100%
At a glance
- Task
- Image Classification
- Library
- transformers
- License
- apache-2.0
- Model type
- vit
- Access
- Public
- Created
- Aug 1, 2023
- Updated
- Aug 3, 2023
- SHA
- 0d981455
Model card
From the Hugging Face model README
<!-- 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. -->
vit-base_rvl_cdip_symce
This model is a fine-tuned version of jordyvl/vit-base_rvl-cdip on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6253
- Accuracy: 0.8982
- Brier Loss: 0.1796
- Nll: 1.1468
- F1 Micro: 0.8982
- F1 Macro: 0.8984
- Ece: 0.0846
- Aurc: 0.0197
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: 64
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Brier Loss | Nll | F1 Micro | F1 Macro | Ece | Aurc |
|---|---|---|---|---|---|---|---|---|---|---|
| 0.1665 | 1.0 | 2500 | 0.3898 | 0.8939 | 0.1621 | 1.1704 | 0.8939 | 0.8938 | 0.0463 | 0.0167 |
| 0.1439 | 2.0 | 5000 | 0.3927 | 0.8949 | 0.1602 | 1.1860 | 0.8949 | 0.8954 | 0.0506 | 0.0165 |
| 0.0889 | 3.0 | 7500 | 0.4389 | 0.8941 | 0.1684 | 1.1449 | 0.8941 | 0.8946 | 0.0637 | 0.0172 |
| 0.0574 | 4.0 | 10000 | 0.4870 | 0.8953 | 0.1741 | 1.1605 | 0.8953 | 0.8952 | 0.0719 | 0.0179 |
| 0.0372 | 5.0 | 12500 | 0.5259 | 0.8929 | 0.1792 | 1.1860 | 0.8929 | 0.8935 | 0.0775 | 0.0185 |
| 0.0225 | 6.0 | 15000 | 0.5579 | 0.8959 | 0.1784 | 1.1504 | 0.8959 | 0.8963 | 0.0799 | 0.0196 |
| 0.0126 | 7.0 | 17500 | 0.5905 | 0.8949 | 0.1811 | 1.1714 | 0.8949 | 0.8950 | 0.0836 | 0.0197 |
| 0.0081 | 8.0 | 20000 | 0.6011 | 0.8973 | 0.1791 | 1.1720 | 0.8973 | 0.8975 | 0.0828 | 0.0198 |
| 0.0048 | 9.0 | 22500 | 0.6198 | 0.8975 | 0.1800 | 1.1518 | 0.8975 | 0.8977 | 0.0847 | 0.0198 |
| 0.0038 | 10.0 | 25000 | 0.6253 | 0.8982 | 0.1796 | 1.1468 | 0.8982 | 0.8984 | 0.0846 | 0.0197 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1.post200
- Datasets 2.9.0
- Tokenizers 0.13.2
- Task
- Image Classification
- Library
- transformers
- Type
- vit
- License
- apache-2.0
- Created
- Aug 1, 2023
- Updated
- Aug 3, 2023