As of March 2023, the official Rasa Docker image is not compatible with ARM-based Macs. The solution described here should only be used as a workaround until official support arrives. It has been tested on a Macbook Pro with M1 Processor, but might also work on other ARM-based machines like the Raspberry Pi.
At the time of writing, the only dependency that could neither be satisfied from Pip nor Anaconda is tensorflow-text.
Therefore, projects that rely on Rasa features utilizing this package will not work.
The easiest way to get started is by using one of the available images on Dockerhub:
khalosa/rasa-aarch64:3.0.9khalosa/rasa-aarch64:3.1.0khalosa/rasa-aarch64:3.2.1khalosa/rasa-aarch64:3.2.5khalosa/rasa-aarch64:3.3.1khalosa/rasa-aarch64:3.5.2
docker run -it --rm khalosa/rasa-aarch64:3.5.2Alternatively, if you want to build the image yourself:
RASA_VERSION="3.5.2" ./build-docker.shNote that the build script will probably only work for the latest supported version.