This project use TensorFlow framework to do many interesting applications. Many popular deep leaning architecture will be implemented is this project, including Neural Networks, RNN, LSTM, Auto-encoder, CNN, etc.
-
Basic
Basic models -
Sentiment Classification
Neural Network (NN) and Convolutional NN (CNN) for classifying tweets into 3 classes, negative (0), neutral (2), positive (4). -
Poetry Generate
Chinese poetry generation using LSTM -
Char RNN
Use Char_RNN to generate Enlighs sentence trained with tiny shakespeare corpus. -
Chitchat Bot
Use movie dialog corpus to train a simple chit chat bot. -
LSTM for Text Classification
Use LSTM model to classify texts, including char level and word level. -
Chinese Speech Recognition
Use TensorFlow to train a Chinese Speech Recognition (ASR). -
Gender classification according to Chinese Name
Gender classification using Chinese names. -
CAPTCHA identification
CAPTCHA identification using cnn-rnn
- TensorFlow 0.12.1
- Python 3.5.1
- OS: Windows 10 or Ubuntu 16.04
Yunchao He (yunchaohe@gmail.com)
Weibo: @Yunchao_He