https://github.com/takuti/criteo-ffm
Running field-aware factorization machines on the Criteo data
https://github.com/takuti/criteo-ffm
ctr-prediction factorization-machines field-aware-factorization-machines hive kaggle
Last synced: about 1 year ago
JSON representation
Running field-aware factorization machines on the Criteo data
- Host: GitHub
- URL: https://github.com/takuti/criteo-ffm
- Owner: takuti
- Created: 2018-05-10T04:43:10.000Z (about 8 years ago)
- Default Branch: master
- Last Pushed: 2018-05-30T20:20:11.000Z (about 8 years ago)
- Last Synced: 2025-03-28T15:48:23.374Z (about 1 year ago)
- Topics: ctr-prediction, factorization-machines, field-aware-factorization-machines, hive, kaggle
- Language: C++
- Homepage: https://www.kaggle.com/c/criteo-display-ad-challenge
- Size: 424 KB
- Stars: 7
- Watchers: 2
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
FFM on the Criteo data
===
Try to replicate the result of [Kaggle Display Advertising Challenge](https://www.kaggle.com/c/criteo-display-ad-challenge) by using the following implementations of field-aware factorization machines (FFMs):
- [LIBFFM](https://github.com/guestwalk/libffm)
- [Hivemall](https://github.com/apache/incubator-hivemall)
## Data
Download and convert the full dataset into CSV format:
```sh
./data.sh
ln -s train.csv tr.csv
ln -s test.csv te.csv
```
Or, use tiny data:
```sh
ln -s train.tiny.csv tr.csv
ln -s test.tiny.csv te.csv
```
## Usage
Build LIBFFM and preprocess data as the [winning solution did](https://www.csie.ntu.edu.tw/~r01922136/kaggle-2014-criteo.pdf):
```sh
make
```
Move to [libffm/](./libffm/) or [hivemall/](./hivemall/) and follow the instructions.
## References
- [guestwalk/kaggle-2014-criteo](https://github.com/guestwalk/kaggle-2014-criteo)
- [chenhuang-learn/ffm](https://github.com/chenhuang-learn/ffm)
- https://gist.github.com/myui/aaeef548a17eb90c4e88f824c3ca1bcd