|
--- |
|
license: apache-2.0 |
|
tags: |
|
- image-classification |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: exper1_mesum5 |
|
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. --> |
|
|
|
# exper1_mesum5 |
|
|
|
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 sudo-s/herbier_mesuem5 dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.6401 |
|
- Accuracy: 0.8278 |
|
|
|
## 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: 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: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:| |
|
| 3.9352 | 0.23 | 100 | 3.8550 | 0.1959 | |
|
| 3.1536 | 0.47 | 200 | 3.1755 | 0.2888 | |
|
| 2.6937 | 0.7 | 300 | 2.6332 | 0.4272 | |
|
| 2.3748 | 0.93 | 400 | 2.2833 | 0.4970 | |
|
| 1.5575 | 1.16 | 500 | 1.8712 | 0.5888 | |
|
| 1.4063 | 1.4 | 600 | 1.6048 | 0.6314 | |
|
| 1.1841 | 1.63 | 700 | 1.4109 | 0.6621 | |
|
| 1.0857 | 1.86 | 800 | 1.1832 | 0.7112 | |
|
| 0.582 | 2.09 | 900 | 1.0371 | 0.7479 | |
|
| 0.5971 | 2.33 | 1000 | 0.9839 | 0.7462 | |
|
| 0.4617 | 2.56 | 1100 | 0.9233 | 0.7657 | |
|
| 0.4621 | 2.79 | 1200 | 0.8417 | 0.7828 | |
|
| 0.2128 | 3.02 | 1300 | 0.7644 | 0.7970 | |
|
| 0.1883 | 3.26 | 1400 | 0.7001 | 0.8183 | |
|
| 0.1501 | 3.49 | 1500 | 0.6826 | 0.8201 | |
|
| 0.1626 | 3.72 | 1600 | 0.6568 | 0.8254 | |
|
| 0.1053 | 3.95 | 1700 | 0.6401 | 0.8278 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.20.1 |
|
- Pytorch 1.12.0+cu113 |
|
- Datasets 2.3.2 |
|
- Tokenizers 0.12.1 |
|
|
