Med-Amine-saighi / kaggle-google-quest-challenge

60th Place Solution for Google QUEST Q&A Labeling

Home Page:https://www.kaggle.com/c/google-quest-challenge

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

kaggle-google-quest-challenge

This respository contains my code for competition in kaggle.

60th Place Solution for Google QUEST Q&A Labeling

Public score: 0.43938 (64th)
Private score: 0.40094 (60th)

Prerequisite

Pull PyTorch image from NVIDIA GPU CLOUD (NGC)

docker login nvcr.io
docker image pull nvcr.io/nvidia/pytorch:20.01-py3
docker run --gpus all -it --ipc=host --name=bert nvcr.io/nvidia/pytorch:20.01-py3

Usage

pip install iterative-stratification
pip install category_encoders

# train BERT model
python train_bert.py

About

60th Place Solution for Google QUEST Q&A Labeling

https://www.kaggle.com/c/google-quest-challenge


Languages

Language:Jupyter Notebook 92.9%Language:Python 7.1%