Clone repository and run Docker

Clone the starter

We've created a starter for you to use as a base for this tutorial. Clone the starter from GitHub:

Copy
git clone https://github.com/hasura/ai-workshop-hasuracon23.git

Then, head into the directory:

Copy
cd ai-workshop-hasuracon23

Next, we'll need to use git-lfs to pull down the large files in the repository. If you don't have git-lfs, install it globally and then run the following command to install it on the repository:

Copy
git lfs install

Next, pull down the large files:

Copy
git lfs pull

Set the right variable for your system

Our docker-compose file references a variable called ARCH which can be set to either arm64 or amd64. This variable is used to pull the right images for your system. If you're running on an M1 or M2 Mac, you'll want to set this variable to arm64. If you're running on an Intel Mac, you'll want to set this variable to amd64:

Copy
# For M1/M2 Macs
export ARCH="arm64"
# For Intel Macs
export ARCH="amd64"

Start the containers

Next, we'll need to start the containers for the application. We'll use Docker Compose to start the containers. From the root of the repository, run the following commands in sequence:

Copy
docker load < hasura-image-arm64.tar.gz
docker load < weaviate-gdc-image-arm64.tar.gz
docker compose up -d

The first two commands will load the images for Hasura and VectorDB into Docker. The last command will start the containers for the application. You should see output similar to the following after all of the containers have started:

Copy
 ⠿ Container ai-workshop-hasuracon23-handlers-1        Started
 ⠿ Container ai-workshop-hasuracon23-postgres-1        Started
 ⠿ Container ai-workshop-hasuracon23-weaviate-gdc-1    Started
 ⠿ Container ai-workshop-hasuracon23-graphql-engine-1  Started

At this point, you're ready to shape your data and configure your Postgres instance!