Menulis loop pelatihan pertama kita

Pengantar Deep Learning dengan PyTorch

Jasmin Ludolf

Senior Data Science Content Developer, DataCamp

Melatih jaringan saraf

Buat model
Pilih fungsi loss
Definisikan dataset
Atur optimizer
Jalankan training loop:
- Hitung loss (forward pass)
- Hitung gradien (backpropagation)
- Perbarui parameter model

Memperkenalkan dataset Gaji Data Science

 experience_level  employment_type  remote_ratio  company_size  salary_in_usd  
        0                0               0.5             1            0.036 
        1                0               1.0             2            0.133     
        2                0               0.0             1            0.234  
        1                0               1.0             0            0.076  
        2                0               1.0             1            0.170

Fitur: kategorikal, target: gaji (USD)
Keluaran akhir: linear layer
Loss: khusus regresi

Loss Mean Squared Error

Loss MSE adalah rata-rata selisih kuadrat antara prediksi dan ground truth

def mean_squared_loss(prediction, target):
  return np.mean((prediction - target)**2)

di PyTorch:

criterion = nn.MSELoss()
# Prediction and target are float tensors
loss = criterion(prediction, target)

Sebelum training loop

# Create the dataset and the dataloader
dataset = TensorDataset(torch.tensor(features).float(),
                        torch.tensor(target).float())


dataloader = DataLoader(dataset, batch_size=4, shuffle=True)


# Create the model
model = nn.Sequential(nn.Linear(4, 2), 
                      nn.Linear(2, 1))


# Create the loss and optimizer
criterion = nn.MSELoss()
optimizer = optim.SGD(model.parameters(), lr=0.001)

Training loop

for epoch in range(num_epochs):

    for data in dataloader:

        # Set the gradients to zero
        optimizer.zero_grad()

        # Get feature and target from the data loader
        feature, target = data

        # Run a forward pass
        pred = model(feature)
        # Compute loss and gradients
        loss = criterion(pred, target)
        loss.backward()

        # Update the parameters
        optimizer.step()

Ayo berlatih!

Pengantar Deep Learning dengan PyTorch