News Comments Upvotes
Installation
pip install git+https://github.com/jfilter/text-classification-keras#egg=keras_text
Setup
Create a config.py
with the following values
path_data = '/Users/filter/data/pol_comments_selection.csv'
path_embedding = '/Users/filter/data/guardian-twokenized-lower-50.vec'
path_for_proc_data = 'imdb_proc_data.bin'
Usage
-
Build (preprocess) the dataset
python run build
-
Train
python run train