Training a Neural Network

• Randomly initialize the weights
• Implement forward propagation to get h_Θ?(x⁽ⁱ⁾) for any x⁽ⁱ⁾
• Implement the cost function
• Implement backpropagation to compute partial derivatives
• Use gradient checking to confirm that your backpropagation works. Then disable gradient checking.
• Use gradient descent or a built-in optimization function to minimize the cost function with the weights in theta.

訓練一個神經網絡

• 隨機初始化權重
• 運用前向傳播得到所有樣本x⁽ⁱ⁾的h_Θ?(x⁽ⁱ⁾)
• 計算損失函數
• 運用反向傳播計算偏導
• 運用梯度檢查確保梯度下降算法的正確運行，然後關閉梯度檢查
• 運用梯度下降算法或者別的優化算法優化權重以最小化損失函數

訓練神經網絡的一般步驟