Light

tealgreen0503 / probspace-spam-mail-competition Goto Github PK

View Code? Open in Web Editor NEW

0.0 1.0 0.0 2.38 MB

ProbSpace: スパムメール判別 6th Place Solution

Jupyter Notebook 100.00%

probspace-spam-mail-competition's Introduction

ProbSpace スパムメール判別 6th Place Solution

https://prob.space/competitions/spam_mail

pre-process

メール内の一番最初の改行コードを[SEP]に置換
- メールのタイトルと本文をBERTに区別して入力
- 文頭の"Subject: "は除去しなかった

modeling

Stratified 5-Fold
models
- BERT
- RoBERTa
- Electra
- 上記3つの出力の平均をとってアンサンブル
各層の[CLS]のhidden stateの重み付き平均を出力
- 重みが合計1になるように制約をかけて，訓練可能なパラメータとした
- Google QUEST Q&A Labeling 1st place より
  - https://www.kaggle.com/c/google-quest-challenge/discussion/129840
各層ごとにlearning rateを設定
- 出力から遠い層ほど小さくなるように設定
- How to Fine-Tune BERT for Text Classification? より
  - https://arxiv.org/abs/1905.05583
multi-sample dropout
- Multi-Sample Dropout for Accelerated Training and Better Generalization より
  - https://arxiv.org/abs/1905.09788
cosine_schedul_with_warmup

post-process

出力結果が7838:17000になるように最適化
- SIGNATE Sutudent Cup 2020 より
  - https://signate.jp/competitions/281/discussions/20200816040343-8180
タイトル，本文ともにnullであるメール（'Subject: \r\n'）を0と予測

what did not work

本文中に多く登場した'enron', 'ect', 'hou'をtokenizerの辞書に追加
XLNet
- もっとしっかりチューニングをすれば効果的だったかもしれない

result

Public Score

5位

Private score

6位

Recommend Projects

React

A declarative, efficient, and flexible JavaScript library for building user interfaces.
Vue.js

🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript

TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
TensorFlow

An Open Source Machine Learning Framework for Everyone
Django

The Web framework for perfectionists with deadlines.
Laravel

A PHP framework for web artisans
D3

Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

javascript

JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
web

Some thing interesting about web. New door for the world.
server

A server is a program made to process requests and deliver data to clients.
Machine learning

Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Visualization

Some thing interesting about visualization, use data art
Game

Some thing interesting about game, make everyone happy.

Recommend Org

Facebook

We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft

Open source projects and samples from Microsoft.
Google

Google ❤️ Open Source for everyone.
Alibaba

Alibaba Open Source for everyone
D3

Data-Driven Documents codes.
Tencent

China tencent open source team.