ivanpastukhov / finch Goto Github PK

simplifies code by using high-level TF interfaces: tf.layers, tf.data, tf.estimator;

produces lots of clean Notebooks to make life easy;

can be tested with CPU;

Python 3 and Jupyter Notebook are required

(CPU User) $ pip3 install tensorflow sklearn scipy bunch tqdm wget

(GPU User) $ pip3 install tensorflow-gpu sklearn scipy bunch tqdm wget

Penn Treebank（PTB 语料库）

• Skip-Gram     Notebook

• CBOW     Notebook

IMDB Movie Reviews Sentiment（电影评论情感分类）

• TF-IDF + LR     Notebook

• Text-CNN     Notebook

• Gated Linear Unit     Notebook

• Bi-RNN     Notebook

• ConvLSTM     Notebook

• Attention-Pooling     Notebook

Language Model（语言模型）

• RNN LM     Notebook

• Character-Aware LM     Notebook

• Dilated-CNN LM     Notebook

• Self-Attention LM     Notebook

Speech Recognition（语音识别）

• CTC     Notebook

User-Item Matching     Folder

Question Matching     Folder

POS Tagging（词性识别）

• Bi-RNN + CRF     Notebook

• CNN + CRF     Notebook

Chinese Segmentation（中文分词）

• Bi-RNN + CRF     Notebook

• CNN + CRF     Notebook

Learning to Sort（机器排序）

• Seq2Seq + Attention + Beam-Search     Notebook

• Pointer Network     Notebook

• Transformer     Notebook

Learning to Dialog（机器对话）

• Transformer     Notebook

Learning to Reconstruct（重建句子）

• VAE (RNN)     Notebook

• VAE (Dilated-CNN)     Notebook

• VAE (Self-Attention)     Notebook

Learning to Control（定向重建句子）

• VAE + Discriminator     Folder

Memory Network（记忆网络）

• End-to-End Memory Network     Notebook

• Dynamic Memory Network     Notebook

Notes

Link Prediction（关系预测）

• DistMult     Notebook (WN18)

Loading Data (导入数据)

• TFRecord: Fixed Length

• TFRecord: Padded Batch

Project Template（项目模版）

• TF Estimator Template

• TF Dataset Template