CorNet

Correlation Networks for Extreme Multi-label Text Classification

Prerequisites

• python==3.6.3
• pytorch==1.2.0
• torchgpipe==0.0.5
• click==7.0
• ruamel.yaml==0.16.5
• numpy==1.16.2
• scipy==1.2.1
• scikit-learn==0.20.3
• gensim==3.7.2
• nltk==3.2.4
• tqdm==4.31.1
• joblib==0.13.2
• logzero==1.5.0

Datasets

• EUR-Lex
• AmazonCat-13K
• Wiki-500K

Pretrained Word Embeddings in gensim format

• GloVe embeddings (840B,300d)

Run

Preprocess (the EUR-Lex dataset is already tokenized in advance)

./scripts/preprocess_eurlex.sh

or (the other datasets need to be tokenized using NLTK)

./scripts/preprocess_others.sh

Train and evaluate

./scripts/run_models.sh

Baselines

The codes for the baseline models are adapted from the following repositories: XML-CNN, BERT, MeSHProbeNet, and AttentionXML.