File size: 2,257 Bytes
00adc38 |
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 73 74 75 |
---
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
- image-classification
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: vit-base-melSpecImagesCREMA
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-base-melSpecImagesCREMA
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the Supreeta03/CREMA-melSpecImages dataset.
It achieves the following results on the evaluation set:
- Loss: 1.1416
- Accuracy: 0.5808
## 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: 10
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 1.5606 | 0.76 | 100 | 1.4424 | 0.4079 |
| 1.2841 | 1.53 | 200 | 1.4981 | 0.3695 |
| 1.0159 | 2.29 | 300 | 1.1693 | 0.5518 |
| 0.9868 | 3.05 | 400 | 1.0969 | 0.5931 |
| 0.8477 | 3.82 | 500 | 1.1719 | 0.5797 |
| 0.5495 | 4.58 | 600 | 1.2348 | 0.5806 |
| 0.2671 | 5.34 | 700 | 1.3457 | 0.5854 |
| 0.1388 | 6.11 | 800 | 1.3891 | 0.5787 |
| 0.1548 | 6.87 | 900 | 1.4216 | 0.5979 |
| 0.0906 | 7.63 | 1000 | 1.6401 | 0.5643 |
| 0.1047 | 8.4 | 1100 | 1.6780 | 0.5873 |
| 0.0583 | 9.16 | 1200 | 1.6795 | 0.5768 |
| 0.0228 | 9.92 | 1300 | 1.6926 | 0.5883 |
### Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
|