Skip to content

Latest commit

 

History

History
26 lines (20 loc) · 751 Bytes

README.md

File metadata and controls

26 lines (20 loc) · 751 Bytes

item_embed_rec

Recommendation using item embedding based on user history

GradientBoosting(machine learning using only CPU) version

gbc - using GradientBoosting for classification
tf_wals - WALS(Weighted Alternating Least Sqaure by Tensorflow)
wp_predict - method hub for many models

din_model - base_model.py, din_model.py

Deep Interest Network by Alibaba, rewritten in tensorflow

WALS

tf_wals.py - main for Weighted Alternating squares MF model tf_wals_lib.py - wals model tf_wals_low.py - low level version of wals model

w2v_lr

using built-in linear classifier

w2vgbc.py

using scikit-learn gradientboostingclassifier

wp_rnn_37.py

bidirectional GRU

preprocessing modules

many files, make*, preprocess*, *preprocess