resnet-50-mnist-87
This model is a fine-tuned version of microsoft/resnet-50 on the mnist dataset. It achieves the following results on the evaluation set:
- Loss: 0.0872
- Accuracy: 0.9757
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: 32
- eval_batch_size: 256
- seed: 87
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
2.1328 | 1.0 | 1594 | 0.9452 | 0.7733 |
0.8675 | 2.0 | 3188 | 0.2649 | 0.9407 |
0.6966 | 3.0 | 4782 | 0.2234 | 0.9443 |
0.579 | 4.0 | 6376 | 0.1318 | 0.9632 |
0.5492 | 5.0 | 7970 | 0.1083 | 0.9688 |
0.52 | 6.0 | 9564 | 0.0995 | 0.9722 |
0.5 | 7.0 | 11158 | 0.0950 | 0.9741 |
0.4909 | 8.0 | 12752 | 0.0924 | 0.9739 |
0.4774 | 9.0 | 14346 | 0.0887 | 0.9744 |
0.4802 | 10.0 | 15940 | 0.0872 | 0.9757 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.1.0
- Datasets 2.15.0
- Tokenizers 0.13.3
- Downloads last month
- 2