vit-hybrid-base-bit-384_rice-leaf-disease-augmented-v4_v5_fft
This model is a fine-tuned version of google/vit-hybrid-base-bit-384 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2360
- Accuracy: 0.9664
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: 5e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine_with_restarts
- lr_scheduler_warmup_steps: 512
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Accuracy | Validation Loss |
|---|---|---|---|---|
| 1.3979 | 0.5 | 128 | 0.8289 | 0.5143 |
| 0.225 | 1.0 | 256 | 0.9228 | 0.2283 |
| 0.0585 | 1.5 | 384 | 0.9094 | 0.2963 |
| 0.062 | 2.0 | 512 | 0.9530 | 0.1543 |
| 0.0166 | 2.5 | 640 | 0.9295 | 0.3726 |
| 0.0189 | 3.0 | 768 | 0.9430 | 0.2889 |
| 0.0126 | 3.5 | 896 | 0.9497 | 0.2025 |
| 0.0031 | 4.0 | 1024 | 0.9597 | 0.2456 |
| 0.0001 | 4.5 | 1152 | 0.9631 | 0.2122 |
| 0.0 | 5.0 | 1280 | 0.9631 | 0.2111 |
| 0.0 | 5.5 | 1408 | 0.9631 | 0.2112 |
| 0.0 | 6.0 | 1536 | 0.9631 | 0.2110 |
| 0.0 | 6.5 | 1664 | 0.9631 | 0.2110 |
| 0.0 | 7.0 | 1792 | 0.9631 | 0.2117 |
| 0.0 | 7.5 | 1920 | 0.9631 | 0.2121 |
| 0.0 | 8.0 | 2048 | 0.9631 | 0.2133 |
| 0.0 | 8.5 | 2176 | 0.9631 | 0.2144 |
| 0.0 | 9.0 | 2304 | 0.9631 | 0.2148 |
| 0.0 | 9.5 | 2432 | 0.9631 | 0.2155 |
| 0.0 | 10.0 | 2560 | 0.9631 | 0.2154 |
| 0.0 | 10.5 | 2688 | 0.9631 | 0.2154 |
| 0.0 | 11.0 | 2816 | 0.9631 | 0.2154 |
| 0.0 | 11.5 | 2944 | 0.9631 | 0.2152 |
| 0.0 | 12.0 | 3072 | 0.9631 | 0.2162 |
| 0.0 | 12.5 | 3200 | 0.9631 | 0.2184 |
| 0.0 | 13.0 | 3328 | 0.9631 | 0.2174 |
| 0.0 | 13.5 | 3456 | 0.2183 | 0.9664 |
| 0.0 | 14.0 | 3584 | 0.2189 | 0.9664 |
| 0.0 | 14.5 | 3712 | 0.2194 | 0.9664 |
| 0.0 | 15.0 | 3840 | 0.2194 | 0.9664 |
| 0.0 | 15.5 | 3968 | 0.2194 | 0.9664 |
| 0.0 | 16.0 | 4096 | 0.2194 | 0.9664 |
| 0.0 | 16.5 | 4224 | 0.2212 | 0.9664 |
| 0.0 | 17.0 | 4352 | 0.2221 | 0.9664 |
| 0.0 | 17.5 | 4480 | 0.2231 | 0.9664 |
| 0.0 | 18.0 | 4608 | 0.2227 | 0.9664 |
| 0.0 | 18.5 | 4736 | 0.2237 | 0.9664 |
| 0.0 | 19.0 | 4864 | 0.2237 | 0.9664 |
| 0.0 | 19.5 | 4992 | 0.2240 | 0.9664 |
| 0.0 | 20.0 | 5120 | 0.2240 | 0.9664 |
| 0.0 | 20.5 | 5248 | 0.2241 | 0.9664 |
| 0.0 | 21.0 | 5376 | 0.2239 | 0.9664 |
| 0.0 | 21.5 | 5504 | 0.2245 | 0.9664 |
| 0.0 | 22.0 | 5632 | 0.2267 | 0.9664 |
| 0.0 | 22.5 | 5760 | 0.2271 | 0.9664 |
| 0.0 | 23.0 | 5888 | 0.2285 | 0.9664 |
| 0.0 | 23.5 | 6016 | 0.2285 | 0.9664 |
| 0.0 | 24.0 | 6144 | 0.2289 | 0.9664 |
| 0.0 | 24.5 | 6272 | 0.2292 | 0.9664 |
| 0.0 | 25.0 | 6400 | 0.2293 | 0.9664 |
| 0.0 | 25.5 | 6528 | 0.2298 | 0.9664 |
| 0.0 | 26.0 | 6656 | 0.2309 | 0.9664 |
| 0.0 | 26.5 | 6784 | 0.2316 | 0.9664 |
| 0.0 | 27.0 | 6912 | 0.2330 | 0.9664 |
| 0.0 | 27.5 | 7040 | 0.2348 | 0.9664 |
| 0.0 | 28.0 | 7168 | 0.2350 | 0.9664 |
| 0.0 | 28.5 | 7296 | 0.2357 | 0.9664 |
| 0.0 | 29.0 | 7424 | 0.2358 | 0.9664 |
| 0.0 | 29.5 | 7552 | 0.2360 | 0.9664 |
| 0.0 | 30.0 | 7680 | 0.2360 | 0.9664 |
Framework versions
- Transformers 4.48.3
- Pytorch 2.5.1+cu124
- Datasets 3.3.2
- Tokenizers 0.21.1