Tudorx95/mnist-cnn-model

MNIST CNN Model

Model Information

• Framework: PyTorch
• Dataset: MNIST (28x28 grayscale images)
• Architecture: MNISTNet
• Total Parameters: 390,858
• Trainable Parameters: 390,858

Training Configuration

• Epochs: 5
• Batch Size: 128
• Learning Rate: 0.001
• Optimizer: Adam
• Device: cuda

Performance Metrics

• Accuracy: 0.9943
• Precision: 0.9943
• Recall: 0.9942
• F1 Score: 0.9942

Usage

import torch
from huggingface_hub import hf_hub_download

# Download model
model_path = hf_hub_download(
    repo_id="Tudorx95/mnist-cnn-model",
    filename="MNIST_CNN.pth"
)

# Load model
checkpoint = torch.load(model_path)
# Create your model instance and load state dict
# model.load_state_dict(checkpoint['model_state_dict'])

Training Date

2026-02-11 19:12:53