metadata
license: apache-2.0
tags:
- image-classification
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: exper4_mesum5
results: []
exper4_mesum5
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the sudo-s/herbier_mesuem5 dataset. It achieves the following results on the evaluation set:
- Loss: 3.4389
- Accuracy: 0.1331
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: 2e-05
- 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 |
|---|---|---|---|---|
| 3.3793 | 0.23 | 100 | 3.4527 | 0.1308 |
| 3.2492 | 0.47 | 200 | 3.4501 | 0.1331 |
| 3.3847 | 0.7 | 300 | 3.4500 | 0.1272 |
| 3.3739 | 0.93 | 400 | 3.4504 | 0.1320 |
| 3.4181 | 1.16 | 500 | 3.4452 | 0.1320 |
| 3.214 | 1.4 | 600 | 3.4503 | 0.1320 |
| 3.282 | 1.63 | 700 | 3.4444 | 0.1325 |
| 3.5308 | 1.86 | 800 | 3.4473 | 0.1337 |
| 3.2251 | 2.09 | 900 | 3.4415 | 0.1361 |
| 3.4385 | 2.33 | 1000 | 3.4408 | 0.1343 |
| 3.3702 | 2.56 | 1100 | 3.4406 | 0.1325 |
| 3.366 | 2.79 | 1200 | 3.4411 | 0.1355 |
| 3.2022 | 3.02 | 1300 | 3.4403 | 0.1308 |
| 3.2768 | 3.26 | 1400 | 3.4394 | 0.1320 |
| 3.3444 | 3.49 | 1500 | 3.4394 | 0.1314 |
| 3.2981 | 3.72 | 1600 | 3.4391 | 0.1331 |
| 3.3349 | 3.95 | 1700 | 3.4389 | 0.1331 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.12.0+cu113
- Datasets 2.3.2
- Tokenizers 0.12.1