Links

Installation in AWS

Easily install Giskard in AWS in 4 steps

1. Initialize EC2 instance

  • In the AWS console, go to the service EC2 and select one of the following zones: N. Virginia (us-east-1), Paris (eu-west-3), or Singapore (ap-southeast-1)
  • Launch an EC2 instance

2. Configure your EC2 instance

  • Application and OS image: Select the default Ubuntu server 22.04 LTS 64-bit (x86)
  • Instance type: We recommend you to choose at least a t2.large instance type (2vCPU, 8GB memory)
  • Key pair: Choose your usual key pair. If you don't have one, go to the Amazon document to create the right one
  • Network settings: You need to open the ports 19000 and 40051 (TCP connection) to access the Giskard frontend (port 19000) and upload your model (port 40051). To do so, click on Edit and add the following security groups:
  • Storage: Choose a minimum of 30 Gigs of SSD (this will mainly depend on the size of your datasets)

3. Launch the instance and install Giskard

  • Click on Launch instance to create the instance
  • Connect in SSH to your instance. You can for example use the EC2 Instance connect to open a terminal directly in your AWS platform
  • Installation of the Giskard requirements (git and docker)
sudo apt install git
curl -fsSL https://get.docker.com -o get-docker.sh
sudo sh get-docker.sh
  • Installation of Giskard
git clone https://github.com/Giskard-AI/giskard.git
cd giskard
sudo docker compose up -d --force-recreate --no-build

4. Connect to your instance and start uploading an ML model

  • Get your IP address: Click on the ID of the instance you just created and copy its Public IPv4 address (or Public IPv4 DNS)
  • Go to http://<your IP address>:19000 in your web browser
  • The user id is admin and the password is admin
That's it, you are now ready to use Giskard in AWS! Now you can start uploading a model!
You can stop the instance and restart it when you need to save your AWS compute costs. However, note that the IP address will not necessarily be the same. So make sure you copy it again when it's launched