MNIST CNN Model (Seed 1)
This is a Convolutional Neural Network (CNN) trained on the MNIST dataset with random seed 1.
Run Details
- Repository: abir-hr196/cnn_transfer
- Seed-Specific Run: seed_1
- WandB Run: run-seed-1 if current_wandb_run.url else "N/A"
Model Performance
- Best Validation Accuracy: 99.23%
Configuration
- Architecture: CNN
- Dataset: MNIST
- Epochs: 10
- Batch Size: 128
- Learning Rate: 0.0003
- Random Seed Used: 1
How to Use
import torch
# from mnist_transformer import MNISTCNN # Assuming MNISTCNN class definition
# model = MNISTCNN()
# model.load_state_dict(torch.hub.load_state_dict_from_url("https://huggingface.co/abir-hr196/cnn_transfer/resolve/main/seed_1/pytorch_model.bin"))
# model.eval()