Coursera-DeepLearning-Part2-week3 take away

一. Hyperparameter tuning

1. Tuning process

1-最重要：red - learning_rate
2-次要重要：yellow
3-再次重要：purple

---

2.Using an appropriate scale to pick hyperparameters

---

3. Hyperparameters tuning in practice: Pandas vs. Caviar

---

二. Batch Normalization

4. Normalizing activations in a network

---

5. Fitting Batch Norm into a neural network

• 此方法能让神经网络变得更有鲁棒性
• 让神经网络对于超参数的选择上不再那么敏感
• 可以让你更容易地训练非常深的网络

---

6. Why does Batch Norm work?

---

7. Batch Norm at test time

---

三. Multi-class classification

8. Softmax Regression

---

9. Training a softmax classifier