qianrenjian / RecAlg

End-to-end Recommendation Algorithm Package

The currently supported algorithms:

1. MemoNet: MemoNet: MemoNet: Memorizing All Cross Features' Representations Efficiently via Multi-Hash Codebook Network for CTR Prediction
2. FiBiNet++: FiBiNet++: Reducing Model Size by Low Rank Feature Interaction Layer for CTR Prediction
3. FiBiNet: FiBiNET: Combining Feature Importance and Bilinear feature Interaction for Click-Through Rate Prediction .

The following algorithms are planned to be supported:

1. GateNet: GateNet: Gating-Enhanced Deep Network for Click-Through Rate Prediction
2. ContextNet: ContextNet: A Click-Through Rate Prediction Framework Using Contextual information to Refine Feature Embedding
3. MaskNet: MaskNet: Introducing Feature-Wise Multiplication to CTR Ranking Models by Instance-Guided Mask

Prerequisites

• Python >= 3.6.8
• TensorFlow-GPU == 1.14

Getting Started

Installation

• Install TensorFlow-GPU 1.14

• Clone this repo

Dataset

• Links of datasets are:

  • https://www.kaggle.com/c/criteo-display-ad-challenge
  • https://www.kaggle.com/c/avazu-ctr-prediction
  • https://www.kaggle.com/c/kddcup2012-track2

• You can download the original datasets and preprocess them by yourself. Run python -u -m rec_alg.preprocessing.{dataset_name}.{dataset_name}_process to preprocess the datasets. dataset_name can be criteo, avazu or kdd12.

• This repo also contains a demo dataset of criteo, which contains 100,000 samples and has been preprocessed. It is used to help demonstrate models here.

Training

MemoNet

You can use python -u -m rec_alg.model.memonet.run_memonet to train a specific model on a dataset. Parameters could be found in the code.

FiBiNet & FiBiNet ++

You can use python -u -m rec_alg.model.fibinet.run_fibinet --version {version} --config {config_path} to train a specific model on a dataset.

Some important parameters are list below, and other hyper-parameters can be found in the code.

• version: model version, supports v1, ++, and custom, default to ++. For custom, you can adjust all parameter values flexibly.
• config_path: specifies the paths of the input/output files and the fields of the dataset. It is generated during dataset preprocessing. Support values: ./config/criteo/config_dense.json, ./config/avazu/config_sparse.json.
• mode: running mode, supports train, retrain, test.

Acknowledgement

Part of the code comes from DeepCTR.