The whole project

files/presentation.key/Metadata/BuildVersionHistory.plist

@@ -0,0 +1,9 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
+<plist version="1.0">
+<array>
+	<string>Template: Showroom (2015-09-16 11:52)</string>
+	<string>M6.6.1-2560-1</string>
+	<string>M7.3.1-5249-2</string>
+</array>
+</plist>

files/presentation.key/Metadata/DocumentIdentifier

@@ -0,0 +1 @@
+9333E508-7DCC-48E4-AF96-C66012FD5A86

reference_code/Adversarial-Learning-for-Neural-Dialogue-Generation-in-Tensorflow-revert-10-master/.gitattributes

@@ -0,0 +1,17 @@
+# Auto detect text files and perform LF normalization
+* text=auto
+
+# Custom for Visual Studio
+*.cs     diff=csharp
+
+# Standard to msysgit
+*.doc	 diff=astextplain
+*.DOC	 diff=astextplain
+*.docx diff=astextplain
+*.DOCX diff=astextplain
+*.dot  diff=astextplain
+*.DOT  diff=astextplain
+*.pdf  diff=astextplain
+*.PDF	 diff=astextplain
+*.rtf	 diff=astextplain
+*.RTF	 diff=astextplain

reference_code/Adversarial-Learning-for-Neural-Dialogue-Generation-in-Tensorflow-revert-10-master/.gitignore

@@ -0,0 +1,47 @@
+# Windows image file caches
+Thumbs.db
+ehthumbs.db
+
+# Folder config file
+Desktop.ini
+
+# Recycle Bin used on file shares
+$RECYCLE.BIN/
+
+# Windows Installer files
+*.cab
+*.msi
+*.msm
+*.msp
+
+# Windows shortcuts
+*.lnk
+
+# =========================
+# Operating System Files
+# =========================
+
+# OSX
+# =========================
+
+.DS_Store
+.AppleDouble
+.LSOverride
+
+# Thumbnails
+._*
+
+# Files that might appear in the root of a volume
+.DocumentRevisions-V100
+.fseventsd
+.Spotlight-V100
+.TemporaryItems
+.Trashes
+.VolumeIcon.icns
+
+# Directories potentially created on remote AFP share
+.AppleDB
+.AppleDesktop
+Network Trash Folder
+Temporary Items
+.apdisk

reference_code/Adversarial-Learning-for-Neural-Dialogue-Generation-in-Tensorflow-revert-10-master/README.md

@@ -0,0 +1,78 @@
+# Adversarial-Learning-for-Neural-Dialogue-Generation-in-Tensorflow
+
+the paper: Adversarial Learning for Neural Dialogue Generation    https://arxiv.org/pdf/1701.06547.pdf
+
+
+the paper translation in Chinese :http://blog.csdn.net/liuyuemaicha/article/details/60581187
+## Config:
+
+TensorFlow 0.12.0  Python 2.7
+
+## Introduction to Project: al_neural_dialogue 
+
+### 1.
+
+
+ gen_data：    training data for gen model
+
+ disc_data:    training data for disc model
+
+  disc：       code about disc model
+
+  gen:         code about gen model
+
+ utils：       code about data operation and model config
+
+**notice:**
+
+**gen_data** include  chitchat.train.answer, chitchat.train.query, chitchat.dev.answer, chitchat.dev.query (total four files)
+
+**disc_data**  include disc.dev.answer,disc.dev.query, disc.dev.gen 和 disc.train.answer, disc.train.query，disc.tran.gen   (total six files)
+
+**formula of training data**   one sentence one row and splited with space, eg:  i don ' t want to !
+
+
+### 2.run
+
+
+python al_neural_dialogue_train.py
+
+
+**introduction**
+
+
+def main(_):
+
+'''
+
+    # step_1 training gen model
+
+
+    # gen_pre_train()
+
+    # model test
+    # gen_test()
+
+    # step_2 gen training data for disc
+    # gen_disc()
+
+    # step_3 training disc model
+    # disc_pre_train()
+
+    # step_4 training al model
+    # al_train()
+
+    # model test
+    # gen_test() 
+'''
+
+**model introduction**
+
+1、disc model : hierarchical rnn (paper——Building end-to-end dialogue systems using generative hierarchical neural network models)
+
+2、gen model : seq2seq model with attention (GRU cell)
+
+3、method of reward : Monte Carlo Search
+
+4、optimal：Policy Gradient
+