File size: 2,015 Bytes
169c57f f2d32b7 169c57f ebfade5 169c57f f2d32b7 169c57f |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
- precision
- recall
model-index:
- name: vit-focal-skin
results: []
---
<!-- 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-focal-skin
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5783
- Accuracy: 0.8653
- F1: 0.8726
- Precision: 0.8851
- Recall: 0.8653
## 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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
| 0.1024 | 1.0 | 626 | 0.4875 | 0.8187 | 0.8247 | 0.8725 | 0.8187 |
| 0.1279 | 2.0 | 1252 | 0.4645 | 0.8187 | 0.8223 | 0.8295 | 0.8187 |
| 0.0912 | 3.0 | 1878 | 0.4883 | 0.8497 | 0.8454 | 0.8462 | 0.8497 |
| 0.0397 | 4.0 | 2504 | 0.5439 | 0.8238 | 0.8274 | 0.8342 | 0.8238 |
| 0.0004 | 5.0 | 3130 | 0.5795 | 0.8601 | 0.8668 | 0.8787 | 0.8601 |
| 0.0002 | 6.0 | 3756 | 0.5783 | 0.8653 | 0.8726 | 0.8851 | 0.8653 |
### Framework versions
- Transformers 4.29.2
- Pytorch 1.13.1
- Datasets 2.14.5
- Tokenizers 0.13.3
|