neural_network_pytorch.py

neural_network_pytorch.py

import torch
import torch.nn as nn
import pandas as pd

# Ler o arquivo CSV
df = pd.read_csv('diabetes.csv')

# Extrair os dados para feature_set e labels
feature_set = torch.tensor(df.drop('diabetico', axis=1).values, dtype=torch.float32)
labels = torch.tensor(df['diabetico'].values, dtype=torch.float32)

class NeuralNetwork(nn.Module):
    def __init__(self):
        super(NeuralNetwork, self).__init__()
        self.hidden = nn.Linear(3, 4)
        self.output = nn.Linear(4, 1)
        self.sigmoid = nn.Sigmoid()

    def forward(self, x):
        x = self.sigmoid(self.hidden(x))
        x = self.sigmoid(self.output(x))
        return x

model = NeuralNetwork()
criterion = nn.BCELoss()  # Binary Cross Entropy Loss
optimizer = torch.optim.SGD(model.parameters(), lr=0.05)  # Stochastic Gradient Descent

for epoch in range(20000):
    optimizer.zero_grad()  # Zera os gradientes
    output = model(feature_set)
    loss = criterion(output, labels.view(-1, 1))  # Calcula a perda
    loss.backward()  # Propagação para trás
    optimizer.step()  # Atualização dos pesos

# Predição
single_point = torch.tensor([[0, 1, 0]], dtype=torch.float32)  # ou [[1, 0, 0]]
result = model(single_point)
threshold = 0.5
prediction = 1 if result.item() > threshold else 0
print(prediction)