This is usefull code for developing text classification algorithms in python that I've developed over the years.
I make heavy use of sklearn, spacy, gensim, keras and tensorflow.
The repo contains notebooks addressing:
- text processing
- word embedding training and pretrained word embeddings (GLOVE, etc)
- tf_idf preprocessing
- classifiers: logit, lstm and bidirectional lstm (bilstm) with and without attention, feed forward networks (ffn), convolutional networks (cnn) and ensemble of these