I wrote a scraper for comments from pornhub, I want to feed them into a machine-learning AI to generate newer and sadder comments.
- I'm stupid so I'm going to skate over details by using Keras.
- Input will be a small dataset (1-50MB) of text catted together.
- I'm going to steal as much code as possible to get a works on my machine implementation.
- This is not gonna be super clean or pythonic or useful at all.
- json_dump.py is just a script to extract comments from the porny app json output
- regresssion.py generates and runs the model
http://karpathy.github.io/2015/05/21/rnn-effectiveness/ https://towardsdatascience.com/recurrent-neural-networks-by-example-in-python-ffd204f99470 https://deepsense.ai/keras-or-pytorch/ https://www.analyticsvidhya.com/blog/2018/03/text-generation-using-python-nlp/ https://minimaxir.com/2018/05/text-neural-networks/ https://machinelearningmastery.com/text-generation-lstm-recurrent-neural-networks-python-keras/ https://adventuresinmachinelearning.com/word2vec-keras-tutorial