Usr2Vec

Code for learning user representations as described in the paper Modelling Context with User Embeddings for Sarcasm Detection in Social Media paper

requirements:

• python >= 2.7
• sma_toolkit
• numpy
• gensim
• joblib
• theano

Running the code

0. pretrain word embeddings using [gensim] (https://radimrehurek.com/gensim/models/word2vec.html) with the hierarchical softmax option (see the [documention] (https://radimrehurek.com/gensim/models/word2vec.html) on how to do this--tl;dr set the flag hs=1). Save the embeddings in binary format.
1. clone or download the [my_utils] (https://github.com/samiroid/utils) module
2. edit file setup.sh to change the paths to my_utils and the word embeddings; run setup.sh
3. edit file build_data.sh to change the paths to the word embeddings and the file containing the user's tweets; run build_data.sh
4. edit file run.sh to change the paths to the word embeddings (binary format) and the ouput user embeddings; run run.sh
5. kick-back and relax :)