I’d love to do a project involving sentence generation, but not sure how possible this is with Processing. May have to learn Python to do this, but first, I wanted to see if anybody knew if Processing could work for generating sentences (similar to NLP engines)?
For the record, I have little knowledge of this area - I’m just looking to pair sentences (random memories with images) using Processing.
Thanks,
Dan
Hi Dockhands,
I have recently completed a similar project in Python. I switched to Python to work with ML. Since NLPs are a pretty advanced use of advanced ML techniques, I would highly recommend switching to Python. Processing has none of the libraries or assets available to make LSTMs, or really any kind of RNN. Speaking as someone with experience, USE PYTHON. Don’t bother with Processing unless you are making simple deep nets or similar networks. Recurrent neural nets, like Long Short-Term Memory (LSTM) nets are not simple. Learn Python basics, then learn Tensorflow basics, then take a quick look at Keras, and you will be all set. This will take magnitudes less time to learn than if you were to write an LSTM in Processing. Believe me- I’ve tried, and failed. I leave you, friend, with a good luck and a few resources.
http://colah.github.io/posts/2015-08-Understanding-LSTMs/
https://machinelearningmastery.com/text-generation-lstm-recurrent-neural-networks-python-keras/
Best of luck, and have fun! Once you learn Python, a whole new world of ML opens!