plant-seedlings-resnet-50
This model is a fine-tuned version of microsoft/resnet-50 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.4375
- Accuracy: 0.8492
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: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.3644 | 0.53 | 100 | 2.3284 | 0.2294 |
| 1.8235 | 1.06 | 200 | 1.8020 | 0.4396 |
| 1.4216 | 1.6 | 300 | 1.2533 | 0.6341 |
| 1.1626 | 2.13 | 400 | 0.8825 | 0.7186 |
| 0.9175 | 2.66 | 500 | 0.7782 | 0.7598 |
| 0.819 | 3.19 | 600 | 0.6772 | 0.7814 |
| 0.7846 | 3.72 | 700 | 0.6027 | 0.8006 |
| 0.6403 | 4.26 | 800 | 0.5450 | 0.8148 |
| 0.6702 | 4.79 | 900 | 0.5621 | 0.8119 |
| 0.6038 | 5.32 | 1000 | 0.4854 | 0.8374 |
| 0.7048 | 5.85 | 1100 | 0.4793 | 0.8325 |
| 0.4157 | 6.38 | 1200 | 0.4724 | 0.8345 |
| 0.4833 | 6.91 | 1300 | 0.4614 | 0.8404 |
| 0.5064 | 7.45 | 1400 | 0.4901 | 0.8340 |
| 0.4966 | 7.98 | 1500 | 0.4365 | 0.8522 |
| 0.5718 | 8.51 | 1600 | 0.4375 | 0.8423 |
| 0.41 | 9.04 | 1700 | 0.4322 | 0.8536 |
| 0.4956 | 9.57 | 1800 | 0.4375 | 0.8492 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3