Python 3 and Jupyter Notebook are required

(CPU User) $ pip3 install tensorflow==1.6.0 sklearn scipy tqdm

(GPU User) $ pip3 install tensorflow-gpu==1.6.0 sklearn scipy tqdm

feeding is the old approach, not efficient for industry use

Starting to move existing code into new efficient interfaces tf.data.Dataset and tf.estimator.Estimator

feeding

• Skip-Gram     Model     PTB Test

• CBOW     Model     PTB Test

tf.estimator.Estimator API

• IMDB Movie Reviews Sentiment

  • CNN

  • Bi-RNN

  • Attention (no RNN / CNN)

  • TF-IDF + LR

tf.data.Dataset API

• Writing《Anna Karenina》

  • RNN Language Model

  • Attention Language Model (no RNN / CNN)

feeding

• Char-Aware CNN + RNN     Model     PTB Test    

• IMDB Movie Reviews

  • VAE

  • Controllable VAE

tf.data.Dataset API

• POS Tagging

  • Bi-RNN + CRF

  • CNN + CRF

• Chinese Segmentation

  • Bi-RNN + CRF

  • CNN + CRF

tf.estimator.Estimator API

• Learning to Sort

  • Attention Is All You Need

feeding

• Learning to Sort

  • Seq2Seq     Model     Test

    • Seq2Seq + Attention     Model     Test

    • Seq2Seq + BiLSTM Encoder     Model     Test

    • Seq2Seq + Beam-Search     Model     Test

    • Seq2Seq + BiLSTM Encoder + Attention + Beam-Search     Model     Test

  • Pointer Network     Model     Test

• Speech Recognition

  • CTC

Facebook bAbI

• End-To-End Memory Network

• Dynamic Memory Network

Movielens

• Hybrid Neural Network