CS 20 : Tensorflow for Deep Learning Research

Refactoring code examples of CS 20 : Tensorflow for Deep Learning Research following tensorflow 2.0 (current tf 1.12)

• notice

  • {filename}_kd.ipynb is implemented by using tf.keras and tf.data
  • {filename}_de.ipynb is implemented by using tf.data and eager execution
  • {filename}_kde.ipynb is implemented by using tf.keras, tf.data and eager execution

• syllabus : http://web.stanford.edu/class/cs20si/syllabus.html

• github : https://github.com/chiphuyen/stanford-tensorflow-tutorials

01. Overview of Tensorflow

• Lec01 Overview of Tensorflow example code
  • Lec01_Overview of Tensorflow.ipynb

02. Operations

• Lec02 Operations example code
  • Lec02_Operations.ipynb

03. Linear and Logistic Regression

• Simple usage of tf.data

  • How to simply use tf.data.ipynb

• Lec03 Linear and Logistic Regression example code

  • Lec03_Linear Regression with mse loss.ipynb
  • Lec03_Linear Regression with huber loss by low-level.ipynb
  • Lec03_Linear Regression with huber loss by high-level.ipynb
  • Lec03_Linear Regression with tf.data.ipynb
  • Lec03_Linear Regression with tf.data_de.ipynb
  • Lec03_Logistic Regression with ce loss.ipynb
  • Lec03_Logistic Regression with tf.data.ipynb
  • Lec03_Logistic Regression with tf.data_de.ipynb

04. Eager Execution

• Lec04 Eager execution example code
  • Lec04_Eager execution.ipynb
  • Lec04_Automatic differentiation and gradient tape.ipynb
  • Lec04_Custom training basics.ipynb
  • Lec04_Custom training walkthrough.ipynb
  • Lec04_Custom training subclassing.ipynb

05. Variable sharing and managing experiments

• Simple usage of tf.keras

  • How to use keras.ipynb

• Lec05 Variable sharing and managing experiments example code

  • Lec05_Variable sharing.ipynb
  • Lec05_Randomization.ipynb
  • Lec05_Applied example with tf.placeholder.ipynb
  • Lec05_Applied example with tf.data.ipynb
  • Lec05_Applied example with tf.data_kde.ipynb
  • Lec05_Word2vec_simple.ipynb

06. Introduction to ConvNet

07. ConvNet in TensorFlow

• Lec07 ConvNet in TensorFlow example code
  • Lec07_ConvNet mnist by low-level.ipynb
  • Lec07_ConvNet mnist by high-level.ipynb
  • Lec07_ConvNet mnist by high-level_kd.ipynb
  • Lec07_ConvNet mnist by high-level_kde.ipynb
  • Lec07_ConvNet mnist with Weight initialization and Drop out.ipynb
  • Lec07_ConvNet mnist with Weight initialization and Drop out_kde.ipynb
  • Lec07_ConvNet mnist with Weight initialization and Batch norm.ipynb
  • Lec07_ConvNet mnist with Weight initialization and Batch norm_kde.ipynb

08. Style Transfer

09. Variational Auto-Encoders

10. Generative Adversarial Networks

11. Recurrent Neural Networks

• Presentation

  • To quickly implementing RNN.ipynb

• Lec11 Recurrent Neural Networks example code

• many to one, word sentiment classification example

  • Lec11_Many to One Classification by RNN.ipynb
  • Lec11_Many to One Classification by RNN_kde.ipynb
  • Lec11_Many to One Classification by LSTM.ipynb
  • Lec11_Many to One Classification by GRU.ipynb
  • Lec11_Many to One Classification by Bi-directional RNN.ipynb
  • Lec11_Many to One Classification by Bi-directional LSTM.ipynb
  • Lec11_Many to One Classification by Bi-directional GRU.ipynb
  • Lec11_Many to One Classification by Stacked RNN with Drop out.ipynb
  • Lec11_Many to One Classification by Stacked LSTM with Drop out.ipynb
  • Lec11_Many to One Classification by Stacked GRU with Drop out.ipynb
  • Lec11_Many to One Classification by Stacked Bi-directional RNN with Drop out.ipynb
  • Lec11_Many to One Classification by Stacked Bi-directional LSTM with Drop out.ipynb
  • Lec11_Many to One Classification by Stacked Bi-directional GRU with Drop out.ipynb

• many to many, simple pos-tagger example

  • Lec11_Many to Many Classification by RNN.ipynb
  • Lec11_Many to Many Classification by LSTM.ipynb
  • Lec11_Many to Many Classification by GRU.ipynb
  • Lec11_Many to Many Classification by Bi-directional RNN.ipynb
  • Lec11_Many to Many Classification by Bi-directional LSTM.ipynb
  • Lec11_Many to Many Classification by Bi-directional GRU.ipynb
  • Lec11_Many to Many Classification by Stacked RNN with Drop out.ipynb
  • Lec11_Many to Many Classification by Stacked LSTM with Drop out.ipynb
  • Lec11_Many to Many Classification by Stacked GRU with Drop out.ipynb
  • Lec11_Many to Many Classification by Stacked Bi-directional RNN with Drop out.ipynb
  • Lec11_Many to Many Classification by Stacked Bi-directional LSTM with Drop out.ipynb
  • Lec11_Many to Many Classification by Stacked Bi-directional GRU with Drop out.ipynb

12. Seq2Seq with Attention

• Lec12 Seq2Seq with Attention example code

• encoder decoder (many to many), simple neural machine translation example

  • Lec12_Seq2Seq by Encoder RNN and Decoder RNN.ipynb
  • Lec12_Seq2Seq by Encoder Bi-directional RNN and Decoder RNN.ipynb
  • Lec12_Seq2Seq with Attention by Encoder RNN and Decoder RNN.ipynb
  • Lec12_Seq2Seq with Attention by Encoder Bi-directional RNN and Decoder RNN.ipynb