Question:

[jzm0144]

How can one use your model to train on data that is already in the text form and has not come from reddit?

For example, I want to train the model on Cornell Movie Lines dataset.

Any suggestions?

[saran-gangster]

keep your data in reddit phrase directory and run "reddit_phrase.py"