Image Classification
Transformers
PyTorch
TensorBoard
vit
vision
Generated from Trainer
Eval Results (legacy)
# Load model directly
from transformers import AutoImageProcessor, AutoModelForImageClassification
processor = AutoImageProcessor.from_pretrained("stlxx/vit-base-beans")
model = AutoModelForImageClassification.from_pretrained("stlxx/vit-base-beans", device_map="auto")Quick Links
vit-base-beans
This model is a fine-tuned version of google/vit-huge-patch14-224-in21k on the beans dataset. It achieves the following results on the evaluation set:
- Loss: 0.9760
- Accuracy: 0.8195
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: 4
- eval_batch_size: 4
- 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 |
|---|---|---|---|---|
| 1.0596 | 1.0 | 259 | 1.0507 | 0.7143 |
| 1.0165 | 2.0 | 518 | 1.0165 | 0.7895 |
| 1.0113 | 3.0 | 777 | 0.9941 | 0.8045 |
| 1.0067 | 4.0 | 1036 | 0.9804 | 0.8195 |
| 0.9746 | 5.0 | 1295 | 0.9760 | 0.8195 |
Framework versions
- Transformers 4.31.0.dev0
- Pytorch 1.13.1+cu117-with-pypi-cudnn
- Datasets 2.12.0
- Tokenizers 0.13.3
- Downloads last month
- 2
Dataset used to train stlxx/vit-base-beans
Evaluation results
- Accuracy on beansvalidation set self-reported0.820
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-classification", model="stlxx/vit-base-beans") pipe("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/hub/parrots.png")