More information on csv files used about cs291k HOT 3 CLOSED

If you want it to get to work right away all you really need is two files of good and bad tweets respectively. These are essentially just a massive list of tweets (a single tweet per line).

When you run train.py these will get fed in by line 56 & 57, and then everything will work nicely.

Feel free to refer to the research paper too: https://www.academia.edu/35947062/Twitter_Sentiment_Analysis_using_combined_LSTM-CNN_Models

from cs291k.

Thank you for the information. The paper is also good.
My another doubt is, does it take a long time to load the whole glove file every time we run?

from cs291k.

Yea, I don't remember the specific timing, but it probably did take a minute or two.

from cs291k.