metadata
license: apache-2.0
base_model: facebook/dinov2-small-imagenet1k-1-layer
tags:
- image-classification
- vision
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: dinov2-small-imagenet1k-1-layer-finetuned-galaxy10-decals
results: []
dinov2-small-imagenet1k-1-layer-finetuned-galaxy10-decals
This model is a fine-tuned version of facebook/dinov2-small-imagenet1k-1-layer on the matthieulel/galaxy10_decals dataset. It achieves the following results on the evaluation set:
- Loss: 1.0579
- Accuracy: 0.6387
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: 1e-07
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 3.2794 | 0.9940 | 124 | 3.2713 | 0.0924 |
| 2.2032 | 1.9960 | 249 | 2.1443 | 0.2401 |
| 1.9114 | 2.9980 | 374 | 1.8624 | 0.3777 |
| 1.6905 | 4.0 | 499 | 1.6865 | 0.4453 |
| 1.5548 | 4.9940 | 623 | 1.5515 | 0.4853 |
| 1.4678 | 5.9960 | 748 | 1.4546 | 0.5068 |
| 1.3977 | 6.9980 | 873 | 1.3747 | 0.5372 |
| 1.3531 | 8.0 | 998 | 1.3144 | 0.5536 |
| 1.3098 | 8.9940 | 1122 | 1.2655 | 0.5660 |
| 1.2152 | 9.9960 | 1247 | 1.2197 | 0.5817 |
| 1.2257 | 10.9980 | 1372 | 1.1856 | 0.5924 |
| 1.1706 | 12.0 | 1497 | 1.1540 | 0.6082 |
| 1.1836 | 12.9940 | 1621 | 1.1286 | 0.6122 |
| 1.1769 | 13.9960 | 1746 | 1.1120 | 0.6161 |
| 1.1197 | 14.9980 | 1871 | 1.0943 | 0.6234 |
| 1.1373 | 16.0 | 1996 | 1.0772 | 0.6308 |
| 1.1111 | 16.9940 | 2120 | 1.0714 | 0.6330 |
| 1.1274 | 17.9960 | 2245 | 1.0624 | 0.6336 |
| 1.0801 | 18.9980 | 2370 | 1.0584 | 0.6381 |
| 1.0979 | 19.8798 | 2480 | 1.0579 | 0.6387 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.1.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1