ResNet-18 trained on STL-10 Subset
This model is a ResNet-18 architecture trained on the STL-10 Subset dataset.
Model Details
- Architecture: ResNet-18
- Dataset: STL-10 Subset (10 classes)
- Classes: airplane, bird, car, cat, deer, dog, horse, monkey, ship, truck
- Best Validation Accuracy: 90.40%
- Framework: PyTorch
Usage
import torch
from torchvision import models
from huggingface_hub import hf_hub_download
# Load model
model = models.resnet18(pretrained=False)
model.fc = torch.nn.Linear(model.fc.in_features, 10)
# Download weights
model_path = hf_hub_download(repo_id="DuckyDuck123/resnet18-stl10", filename="pytorch_model.pth")
checkpoint = torch.load(model_path, map_location='cpu')
model.load_state_dict(checkpoint['model_state_dict'])
model.eval()
Training Details
- Optimizer: Adam
- Learning Rate: 0.001
- Batch Size: 32
- Epochs: 10