RASA Core+NLU English and Oriya

Chat Bot Natural Language Understanding and Processing for Oriya and English Language

System Requirement

1. Ubuntu 16+
2. Mac OS Latest
3. Windows 10 (Visual C++ Build tool)

Install Mini Conda (Part of Anaconda)

https://conda.io/miniconda.html

Install Python

Install Python using Conda. Rasa best works on Python 3.6

conda install python=3.6

Create Environment for Rasa Project under Python 3.6

We except you have install python 3.6 using Conda command

conda create -n rasa python=3.6

Install Visual studio Code and Open a Empty Folder

https://code.visualstudio.com/download

Activate python 3.6 using Conda (important step)

source activate rasa

Install Required Python Packages

Install Rasa NLU, Tensorflow, Spacy, Spacy ENG lang and Rasa Core

pip install  --no-cache-dir rasa_nlu
pip install  --no-cache-dir rasa_nlu[spacy]
python -m spacy download en_core_web_md
python -m spacy link en_core_web_md en
pip install  --no-cache-dir rasa_nlu[tensorflow]
pip install -U rasa_core==0.11.11

Clone this repo

git@github.com:bikashkumars/rasa_core_nlu_oriya.git

Run Rasa command to create Model

python -m rasa_nlu.train -c nlu_config.yml --data nlu.md -o models --fixed_model_name nlu --project current --verbose

Test

python index.py

Run as Server

python -m rasa_nlu.server --path projects

Information

So far, you have suceefully completed RASA NLU. Now we will move to Rasa Core

Train Rasa Core using Story and Domain

python -m rasa_core.train -d domain.yml -s stories.md -o models/dialogue

Run rasa Code as Server

python -m rasa_core.run -d models/dialogue -u models/current/nlu

Versions

pip show rasa_core Rasa Core - Version: 0.12.2

pip show rasa_nlu 0.13.8

pip show tensorflow Version: 1.10.0

Known Error rasa_core.exceptions.UnsupportedDialogueModelError: The model version is to old to be loaded by this Rasa Core instance. Either retrain the model, or run withan older version. Model version: 0.11.3 Instance version: 0.12.2 Minimal compatible version: 0.12.0

pip install -U rasa_core==0.11.11

Process to Run this

Step-1 Train Rasa NLU (Intent, Entity)

python -m rasa_nlu.train -c nlu_config.yml --data nlu.md -o models --fixed_model_name nlu --project current --verbose

Step-2 Generate Dialogs using Rasa Core (Flow)

python -m rasa_core.train -d domain.yml -s stories.md -o models/dialogue

Step-3 Run this Rasa Core Server

python -m rasa_core.run -d models/dialogue -u models/current/nlu

Step-4 Start Bot and start chatting

python bot.py