SOTAVerified

Bag of Tricks for Efficient Text Classification

2016-07-06EACL 2017Code Available1· sign in to hype

Armand Joulin, Edouard Grave, Piotr Bojanowski, Tomas Mikolov

Code Available — Be the first to reproduce this paper.

Reproduce

Code

Abstract

This paper explores a simple and efficient baseline for text classification. Our experiments show that our fast text classifier fastText is often on par with deep learning classifiers in terms of accuracy, and many orders of magnitude faster for training and evaluation. We can train fastText on more than one billion words in less than ten minutes using a standard multicore~CPU, and classify half a million sentences among~312K classes in less than a minute.

Tasks

Benchmark Results

DatasetModelMetricClaimedVerifiedStatus
Amazon Review FullFastTextAccuracy60.2Unverified
Amazon Review PolarityFastTextAccuracy94.6Unverified
Sogou NewsfastText, h=10, bigramAccuracy96.8Unverified
Yelp Binary classificationfastText, h=10, bigramError4.3Unverified
Yelp Fine-grained classificationFastTextError36.1Unverified

Reproductions