<!-- 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. -->
PE_mobile_vit_v2
This model is a fine-tuned version of jonglet/mobile_vit on an unknown dataset. It achieves the following results on the evaluation set:
-
Accuracy: 0.9643
-
F1: 0.9470
-
Precision: 0.9792
-
Recall: 0.9375
-
Loss: 0.2799
-
Classification Report: precision recall f1-score support
0 1.00 1.00 1.00 4 1 1.00 1.00 1.00 3 2 1.00 1.00 1.00 3 3 1.00 1.00 1.00 4 4 1.00 1.00 1.00 3 5 1.00 1.00 1.00 4 6 0.83 1.00 0.91 5 7 1.00 0.50 0.67 2accuracy 0.96 28 macro avg 0.98 0.94 0.95 28 weighted avg 0.97 0.96 0.96 28
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: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- 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
Training results
| Training Loss | Epoch | Step | Accuracy | F1 | Precision | Recall | Validation Loss | Classification Report |
|---|---|---|---|---|---|---|---|---|
| 1.1217 | 1.0 | 172 | 0.7857 | 0.7429 | 0.8438 | 0.7875 | 0.7660 | precision recall f1-score support |
0 0.00 0.00 0.00 4
1 1.00 1.00 1.00 3
2 0.75 1.00 0.86 3
3 1.00 1.00 1.00 4
4 1.00 1.00 1.00 3
5 1.00 1.00 1.00 4
6 0.80 0.80 0.80 5
7 0.20 0.50 0.29 2
accuracy 0.79 28
macro avg 0.72 0.79 0.74 28 weighted avg 0.74 0.79 0.76 28 | | 0.6866 | 2.0 | 344 | 0.9286 | 0.9196 | 0.9643 | 0.9062 | 0.3667 | precision recall f1-score support
0 1.00 0.75 0.86 4
1 1.00 1.00 1.00 3
2 1.00 1.00 1.00 3
3 1.00 1.00 1.00 4
4 1.00 1.00 1.00 3
5 1.00 1.00 1.00 4
6 0.71 1.00 0.83 5
7 1.00 0.50 0.67 2
accuracy 0.93 28
macro avg 0.96 0.91 0.92 28 weighted avg 0.95 0.93 0.93 28 | | 0.6053 | 3.0 | 516 | 0.9643 | 0.9470 | 0.9792 | 0.9375 | 0.2799 | precision recall f1-score support
0 1.00 1.00 1.00 4
1 1.00 1.00 1.00 3
2 1.00 1.00 1.00 3
3 1.00 1.00 1.00 4
4 1.00 1.00 1.00 3
5 1.00 1.00 1.00 4
6 0.83 1.00 0.91 5
7 1.00 0.50 0.67 2
accuracy 0.96 28
macro avg 0.98 0.94 0.95 28 weighted avg 0.97 0.96 0.96 28 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1
