This is a sample shopping bot to help customers buy products from e-commerce sites. Built using Rasa 2.8.5 and Rasa x 0.39.3
Run:
First create an virtual environment in anaconda prompt using the command :
conda create -n rasavirtualenv python = 3.6
Activate your environment using the command :
conda activate rasavirtualenv
Install latest rasa version:
pip3 install rasa
or install speicific version 2.8(if any error occurs with latest one)
pip3 install rasa==2.8.5
Install Rasa X:
pip3 install rasa-x==0.39.3 --extra-index-url https://pypi.rasa.com/simple
Use rasa train
to train a model.
Use rasa shell
to talk with the bot(console).
Use rasa x
to talk with the bot(browser).
For custom actions, first set up your action server in one terminal window:
rasa run actions
For entity extraction, run the duckling server (In another window):
docker run -p 8000:8000 rasa/duckling
TO talk to your bot and debug at the same time:
rasa shell --debug
Note that , If you get errors follow these:
- For c++ 14 related error : Install [MS Visual Studio Build Tools] (https://www.scivision.dev/python-windows-visual-c-14-required)
- If Rasa doesn't start: First:
pip3 install SQLAlchemy==1.3.22
If it still doesn't launch: First:
pip install sanic-jwt==1.6.0
Delete rasa.db and events.db files (only first time)
data/stories.yml
- contains stories
data/nlu.yml
- contains NLU training data
data/rules.yml
- contains the rules upon which the bot responds to queries
actions/actions.py
- contains custom action/api code
domain.yml
- the domain file, including bot response templates
config.yml
- training configurations for the NLU pipeline and policy ensemble
tests/test_stories.yml
- end-to-end test stories
- Ask if a product is available or not.
- Ask how many product in stock,price of product,view products by shop ,category or price range.
- Add Product in Cart
- Subscribe to product updates
Note: This project is still in development all features aren't implemented yet. Be patient.