A chat bot microservice app that can be trained on different datasets.
- python 3.x
- Rasa NLU
- MITIE
- scikit-learn
- Flask
- Flask-cors
- virtualenv
- gunicorn
- nginx
- Rasa NLU Trainer
- Install python 3.6 or above
- Create a virtual environment to install dependencies
$ virtualenv venv
- Enter the virtual environment
$ source venv/bin/activate
- Install dependencies
$ pip install -r requirements.txt
- Use Rasa NLU Trainer to generate
training.json
. - Place it at
./data/training.json
- run
./chatai.sh train
- Create
./data/answers.json
. - Use intents in the training data as key and value as answers.
- Run
./chatBot start