chan20/model

Simple Feed-Forward Neural Network

This is a simple PyTorch feed-forward neural network trained on synthetic data.

Model Details

• Architecture: Feed-forward Neural Network
• Input Size: 10 features
• Hidden Layer: 32 neurons with ReLU activation
• Output Layer: 2 classes (Binary Classification)
• Framework: PyTorch

Training Data

The model was trained on 1000 samples of synthetic data generated using torch.randn.

• Features: 10 random float values per sample.
• Labels: Binary (0 or 1), randomly assigned.
• Split: 80% Training, 20% Testing.

Training Procedure

• Optimizer: Adam
• Loss Function: CrossEntropyLoss
• Batch Size: 32
• Epochs: 20

Usage

Installation

pip install torch

Inference Code

import torch
import torch.nn as nn
import json

# Define Model Architecture
class SimpleNN(nn.Module):
    def __init__(self, input_size, hidden_size, output_size):
        super(SimpleNN, self).__init__()
        self.fc1 = nn.Linear(input_size, hidden_size)
        self.relu = nn.ReLU()
        self.fc2 = nn.Linear(hidden_size, output_size)

    def forward(self, x):
        out = self.fc1(x)
        out = self.relu(out)
        out = self.fc2(out)
        return out

# Load Configuration
with open("config.json", "r") as f:
    config = json.load(f)

# Load Model
model = SimpleNN(config["input_size"], config["hidden_size"], config["output_size"])
model.load_state_dict(torch.load("model.pth"))
model.eval()

# Predict
dummy_input = torch.randn(1, 10)
output = model(dummy_input)
_, prediction = torch.max(output, 1)
print(f"Prediction: {prediction.item()}")