rocketheat / Chat

WSGI Server to chat with chatbot

Chat

WSGI Server to chat with chatbot

To implement:

1. Run the code using the chatbot_tf10 enviroment with: conda env create -f chatbot_cpu_tf10.yml
2. source activate chatbot_cpu_tf10
3. Add your saved model (the ckpt files) to the chatbot folder.
4. In chatbot_load.py file (in part 5 in the bottom) change the ip_address variable to your local ip address (ex. '192.168.1.1').
5. Make sure port 5000 is open on your computer. In linux using a code like sudo ufw allow 5000.
6. Run chatbot_load.py file from terminal with python chatbot_load.py
7. you can access your chatbot with the ip address noted above :5000 (ex. 192.168.1.1:5000) either on your local computer or a remote computer connected to your local network. If you didn't update the ip_address variable, you can access the server locally on localhost:5000.