Apply keras generated LSTM to 'run.py'

[jeff1evesque]

We need to verify, then apply our keras lstm model to the --local case.

[jeff1evesque]

We should also relocate the ./model/chatbot.h5 into chatbot/model/chatbot.h5 for better organization.

[jeff1evesque]

The following is our limited trained lstm traceback during prediction:

root@development:/vagrant# python3 run.py --local
Using TensorFlow backend.

> Do you know the weather for next week?
2018-12-30 03:32:10.289126: I tensorflow/core/platform/cpu_feature_guard.cc:137] Your CPU supports instructions that this TensorFlow binary was not compiled to use: SSE4.1 SSE4.2 AVX AVX2
['missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'incident incident incident incident incident incident incident incident incident incident incident incident incident incident incident incident incident incident incident incident', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'has has has has has has has has has has has has has has has has has has has has', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'has has has has has has has has has has has has has has has has has has has has', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'often often often often often often often often often often often often often often often often often often often often', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'However However However However However However However However However However However However However However However However However However However However', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'has has has has has has has has has has has has has has has has has has has has', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed', 'missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed missed']

Note: we only used half of the reddit-2005-12 during train.