resnet-50-mnist-21
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: 0.0801
- Accuracy: 0.9769
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: 21
- 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 |
---|---|---|---|---|
2.124 | 1.0 | 1594 | 0.8829 | 0.8139 |
0.8392 | 2.0 | 3188 | 0.2524 | 0.9339 |
0.6709 | 3.0 | 4782 | 0.1497 | 0.9591 |
0.5665 | 4.0 | 6376 | 0.1200 | 0.9668 |
0.5399 | 5.0 | 7970 | 0.0994 | 0.9712 |
0.5078 | 6.0 | 9564 | 0.0935 | 0.9719 |
0.4973 | 7.0 | 11158 | 0.0901 | 0.9748 |
0.4846 | 8.0 | 12752 | 0.0838 | 0.975 |
0.4833 | 9.0 | 14346 | 0.0801 | 0.9769 |
0.4761 | 10.0 | 15940 | 0.0793 | 0.9763 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.1.0
- Datasets 2.15.0
- Tokenizers 0.13.3
- Downloads last month
- 2