metadata
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: CGIAR
results: []
CGIAR
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6887
- Accuracy: 0.7211
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: 0.0001
- 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
- lr_scheduler_warmup_steps: 50
- num_epochs: 40
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.8524 | 1.0 | 652 | 0.7425 | 0.6910 |
| 0.7133 | 2.0 | 1304 | 0.7050 | 0.7106 |
| 0.6627 | 3.0 | 1956 | 0.6887 | 0.7211 |
| 0.5699 | 4.0 | 2608 | 0.6967 | 0.7244 |
| 0.4901 | 5.0 | 3260 | 0.6904 | 0.7292 |
| 0.4374 | 6.0 | 3912 | 0.7500 | 0.7255 |
| 0.3277 | 7.0 | 4564 | 0.8507 | 0.7083 |
| 0.2602 | 8.0 | 5216 | 0.8675 | 0.7259 |
| 0.2324 | 9.0 | 5868 | 0.9844 | 0.7315 |
| 0.1703 | 10.0 | 6520 | 1.0635 | 0.7112 |
| 0.1349 | 11.0 | 7172 | 1.0992 | 0.7229 |
| 0.1203 | 12.0 | 7824 | 1.2086 | 0.7215 |
| 0.1148 | 13.0 | 8476 | 1.2128 | 0.7250 |
| 0.087 | 14.0 | 9128 | 1.2238 | 0.7300 |
| 0.0842 | 15.0 | 9780 | 1.2935 | 0.7294 |
| 0.0765 | 16.0 | 10432 | 1.3468 | 0.7282 |
| 0.0609 | 17.0 | 11084 | 1.3041 | 0.7271 |
| 0.0543 | 18.0 | 11736 | 1.4513 | 0.7259 |
| 0.0507 | 19.0 | 12388 | 1.4640 | 0.7382 |
| 0.0457 | 20.0 | 13040 | 1.5730 | 0.7298 |
| 0.0376 | 21.0 | 13692 | 1.5676 | 0.7315 |
| 0.0434 | 22.0 | 14344 | 1.6493 | 0.7315 |
| 0.0324 | 23.0 | 14996 | 1.6090 | 0.7294 |
| 0.0318 | 24.0 | 15648 | 1.7164 | 0.7348 |
| 0.0256 | 25.0 | 16300 | 1.6981 | 0.7413 |
| 0.0252 | 26.0 | 16952 | 1.6465 | 0.7317 |
| 0.0189 | 27.0 | 17604 | 1.7949 | 0.7265 |
| 0.0211 | 28.0 | 18256 | 1.7796 | 0.7284 |
| 0.0184 | 29.0 | 18908 | 1.8446 | 0.7319 |
| 0.0128 | 30.0 | 19560 | 1.8685 | 0.7300 |
| 0.0098 | 31.0 | 20212 | 1.9648 | 0.7278 |
| 0.008 | 32.0 | 20864 | 1.9593 | 0.7303 |
| 0.006 | 33.0 | 21516 | 1.9797 | 0.7325 |
| 0.0064 | 34.0 | 22168 | 1.9700 | 0.7401 |
| 0.0077 | 35.0 | 22820 | 2.0103 | 0.7307 |
| 0.0036 | 36.0 | 23472 | 2.0696 | 0.7334 |
| 0.0019 | 37.0 | 24124 | 2.0095 | 0.7374 |
| 0.0024 | 38.0 | 24776 | 2.0186 | 0.7394 |
| 0.001 | 39.0 | 25428 | 2.0157 | 0.7432 |
| 0.0013 | 40.0 | 26080 | 2.0033 | 0.7438 |
Framework versions
- Transformers 4.36.1
- Pytorch 2.0.0
- Datasets 2.15.0
- Tokenizers 0.15.0