notice
This is documentation for Rasa Documentation v2.x, which is no longer actively maintained.
For up-to-date documentation, see the latest version (3.x).
Installation
Quick Installation
Isolate your python project using a virtual environment.
- Ubuntu
- macOS
- Windows
Create a new virtual environment by choosing a Python interpreter and making a ./venv
directory to hold it:
Activate the virtual environment:
Install Rasa Open Source using pip (requires Python 3.7, or 3.8).
You are now ready to go! So what's next? You can create a new project by running:
You can learn about the most important Rasa commands in the Command Line Interface.
note
Due to lack of official TensorFlow support for the Apple M1, Rasa Open Source is currently unable to train a model using M1.
Step-by-step Installation Guide
Prefer following video instructions? Watch our installation series on Youtube, it explains the installation in walkthroughs for all major platforms.
1. Python Environment Setup
Check if your Python environment is already configured:
If these packages are already installed, these commands should display version numbers for each step, and you can skip to the next step.
Otherwise, proceed with the instructions below to install them.
- Ubuntu
- macOS
- Windows
Fetch the relevant packages using apt
, and install virtualenv using pip
.
2. Virtual Environment Setup
This step is optional, but we strongly recommend isolating python projects using virtual environments. Tools like virtualenv and virtualenvwrapper provide isolated Python environments, which are cleaner than installing packages system-wide (as they prevent dependency conflicts). They also let you install packages without root privileges.
- Ubuntu
- macOS
- Windows
Create a new virtual environment by choosing a Python interpreter and making a ./venv
directory to hold it:
Activate the virtual environment:
3. Install Rasa Open Source
- Ubuntu / macOS / Windows
First make sure your pip
version is up to date:
To install Rasa Open Source:
Telemetry reporting
When you run Rasa Open Source for the first time, you’ll see a message notifying you about anonymous usage data that is being collected. You can read more about how that data is pulled out and what it is used for in the telemetry documentation.
Congratulations! You have successfully installed Rasa Open Source!
Building from Source
If you want to use the development version of Rasa Open Source, you can get it from GitHub:
Additional Dependencies
For some machine learning algorithms you need to install additional python packages. They aren't installed by default to keep the footprint small.
The page on Tuning Your Model will help you pick the right configuration for your assistant and alert you to additional dependencies.
Just give me everything!
If you don't mind the additional dependencies lying around, you can use
to install all needed dependencies for every configuration.
Dependencies for spaCy
For more information on spaCy models, check out the spaCy docs.
You can install it with the following commands:
This will install Rasa Open Source as well as spaCy and its language model
for the English language, but many other languages are availabe too.
We recommend using at least the "medium" sized models (_md
) instead of the spaCy's
default small en_core_web_sm
model. Small models require less
memory to run, but will likely reduce intent classification performance.
Changed in 2.5
Support for Spacy 3 was added. In prior versions of Rasa Open Source, to install spaCy with
its language model for the English language, you need to additionally run
python3 -m spacy link en_core_web_md en
.
Dependencies for MITIE
First, run
and then download the
MITIE models.
The file you need is total_word_feature_extractor.dat
. Save this
anywhere. If you want to use MITIE, you need to
tell it where to find this file (in this example it was saved in the
data
folder of the project directory).
Upgrading Versions
To upgrade your installed version of Rasa Open Source to the latest version from PyPI:
To download a specific version, specify the version number: