Image Classification
Transformers
TensorBoard
Safetensors
swin
Generated from Trainer
Eval Results (legacy)
# Load model directly
from transformers import AutoImageProcessor, AutoModelForImageClassification
processor = AutoImageProcessor.from_pretrained("Angy309/swin-tiny-patch4-window7-224-LEGO")
model = AutoModelForImageClassification.from_pretrained("Angy309/swin-tiny-patch4-window7-224-LEGO")Quick Links
swin-tiny-patch4-window7-224-LEGO
This model is a fine-tuned version of microsoft/swin-tiny-patch4-window7-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.1108
- Accuracy: 0.9718
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: 5e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.2301 | 1.0 | 45 | 0.7921 | 0.7132 |
| 0.5433 | 2.0 | 90 | 0.3047 | 0.8918 |
| 0.4067 | 3.0 | 135 | 0.2028 | 0.9279 |
| 0.3297 | 4.0 | 180 | 0.1282 | 0.9577 |
| 0.3334 | 5.0 | 225 | 0.1108 | 0.9718 |
Captura entrenamiento
Evaluación
Framework versions
- Transformers 4.40.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
- Downloads last month
- 6
Model tree for Angy309/swin-tiny-patch4-window7-224-LEGO
Base model
microsoft/swin-tiny-patch4-window7-224Evaluation results
- Accuracy on imagefolderself-reported0.972
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-classification", model="Angy309/swin-tiny-patch4-window7-224-LEGO") pipe("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/hub/parrots.png")