- make chat UI into a flask object
- add intent to log output to make it easier to track
- load images, webpages, etc. from intent
- bind chat bot to telegram or similar
- check for cyrillic support with NLTK\
- center the UI in the screen
- I've received some index out of range errors. Need to set up error catching.
- clean up the UI
- done in the code: pop responses so that users don't necessarily see same response twice in the same session
- call function from intent
- set up a requirements document (poetry? requirements.txt?)
- figure out a json UI
- It's really not convenient to work with json though editing the json file here in repl.it is not hard.
- convert UI from tk to pysimple GUI: Thanks Horst!
- response intent is included in the chat UI: Thanks Horst!
- get logging to work: Thanks Horst!
- Generates a log file with time stamp
- I need to see how the chat brain is parsing input.
- I don't see any evidence in the code for recording what the users type
- bind return key in UI to send user input: Thanks Horst!
Creating a simple Python chatbot using natural language processing and deep learning.
-
run the libraries.py first Check your dependencies. A requirements file or poetry config would be useful here.
-
update the intents.json file
-
run train_chatbot.py
-
run chatgui.py
- How to call a Python function from a json file?
- how to process logs more effectively?
- the intents file is to support chatbot for a Toastmasters club.
- convert_json_to_csv.py I found this on the web here https://datatofish.com/json-string-to-csv-python/ and was hoping a csv file would be easier to edit than json. No such luck.