<!-- 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. -->
exper_batch_32_e4
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the sudo-s/herbier_mesuem1 dataset. It achieves the following results on the evaluation set:
- Loss: 0.3909
- Accuracy: 0.9067
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: 0.0002
- train_batch_size: 32
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
- mixed_precision_training: Apex, opt level O1
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 3.4295 | 0.31 | 100 | 3.4027 | 0.2837 |
| 2.5035 | 0.62 | 200 | 2.4339 | 0.5247 |
| 1.6542 | 0.94 | 300 | 1.7690 | 0.6388 |
| 1.1589 | 1.25 | 400 | 1.3106 | 0.7460 |
| 0.9363 | 1.56 | 500 | 0.9977 | 0.7803 |
| 0.6946 | 1.88 | 600 | 0.8138 | 0.8207 |
| 0.3488 | 2.19 | 700 | 0.6593 | 0.8489 |
| 0.2935 | 2.5 | 800 | 0.5725 | 0.8662 |
| 0.2557 | 2.81 | 900 | 0.5088 | 0.8855 |
| 0.1509 | 3.12 | 1000 | 0.4572 | 0.8971 |
| 0.1367 | 3.44 | 1100 | 0.4129 | 0.9090 |
| 0.1078 | 3.75 | 1200 | 0.3909 | 0.9067 |
Framework versions
- Transformers 4.19.4
- Pytorch 1.5.1
- Datasets 2.3.2
- Tokenizers 0.12.1
