s# Deep Learning - course-Udacity This is notes based on Udacity Deep Learning Nanodegree Foundation to learn the 'Deep Learning', also ref to:

• Udacity DL by Vincent

Course Notes

1. Intro Neural Networks

   1.1 Sentiment networks:

   Use sentiment as example intro NN basic, like BP/SGD/ and some interesting idea of process the data and skill imporved accuracy and efficiency.

2. Convolutional Neural Networks
3. Recurrent Neural Networks
4. Generative Adversarial Networks

DLND Projects:

• project 1: First Neural Networks
• project 2: Image Classification
• project 3: TV Script Generation
• project 4: Language Translation
• project 5: Face generation

DLND Exercise:

1. Youtube Live Ex:

• 1.4 : word_vector_game_of_thrones

  Understand word2vector, try to use w2v to read

2. CNN

• Image Classify by tf

  use tf show weight/bias in CNN difference layer results, interesting. ref to Siray live

  3. RNN

Deep-learning Advanced Study.

Other