resnet-50-mnist-100
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: 0.0778
- Accuracy: 0.9777
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: 100
- 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 |
---|---|---|---|---|
2.1226 | 1.0 | 1594 | 0.6676 | 0.854 |
0.7717 | 2.0 | 3188 | 0.2226 | 0.9502 |
0.6491 | 3.0 | 4782 | 0.1477 | 0.9619 |
0.5622 | 4.0 | 6376 | 0.1083 | 0.9702 |
0.5311 | 5.0 | 7970 | 0.0977 | 0.9717 |
0.5052 | 6.0 | 9564 | 0.0851 | 0.9744 |
0.4913 | 7.0 | 11158 | 0.0846 | 0.9752 |
0.4825 | 8.0 | 12752 | 0.0778 | 0.9777 |
0.453 | 9.0 | 14346 | 0.0803 | 0.9774 |
0.4677 | 10.0 | 15940 | 0.0775 | 0.9771 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.1.0
- Datasets 2.15.0
- Tokenizers 0.13.3
- Downloads last month
- 2