Image Classification
Transformers
Safetensors
vit
vision
Generated from Trainer
Eval Results (legacy)
# Load model directly
from transformers import AutoImageProcessor, AutoModelForImageClassification
processor = AutoImageProcessor.from_pretrained("super-j/vit-base-mnist")
model = AutoModelForImageClassification.from_pretrained("super-j/vit-base-mnist", device_map="auto")Quick Links
vit-base-mnist
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the mnist dataset. It achieves the following results on the evaluation set:
- Loss: 0.0247
- Accuracy: 0.9949
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: 8
- eval_batch_size: 8
- seed: 1337
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.3215 | 1.0 | 6375 | 0.0630 | 0.9856 |
| 0.4689 | 2.0 | 12750 | 0.0377 | 0.9906 |
| 0.3258 | 3.0 | 19125 | 0.0364 | 0.9908 |
| 0.3094 | 4.0 | 25500 | 0.0269 | 0.9936 |
| 0.2981 | 5.0 | 31875 | 0.0247 | 0.9949 |
Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.1.0
- Datasets 2.14.6
- Tokenizers 0.14.1
- Downloads last month
- 3
Model tree for super-j/vit-base-mnist
Base model
google/vit-base-patch16-224-in21kDataset used to train super-j/vit-base-mnist
Evaluation results
- Accuracy on mnistself-reported0.995
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-classification", model="super-j/vit-base-mnist") pipe("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/hub/parrots.png")