Model Deployment
Deploy your model
Leave it to us.
It’s the easiest part to deploy your function or application to Each servers. You can use Magicpoint CLI tool to deploy your application!
If you have not configured your framework you can visit Magicpoint Configuration page.
You can follow these steps to deploy your application:
- Prepare magicpoint.yaml
- Execute command
Set your environment
magicpoint.yaml
application:
name: Magicpoint Test Application
env:
- EXAMPLE_ENV_VARIABLE: YOUR_VALUE
build:
gpu: A100
gpu_memory: 40Gi
cpu: 1
scaling:
type: QueueDepth
max_task_per_replica: 3
min_replicas:
max_replicas: 2
python_version: 3.8
python_packages:
- pillow==8.2.0
cuda: "11.1"
Execute just one command
magicpoint deploy app.py
[✅] Validating magicpoint.yaml
[✅] Validation app.py for Magicpoint Test Application
[✅] Deploying Magicpoint Test Application
[✅] Deployed on: https://magicpoint-test-application-x2h2gvhax.eachlabs.a
🎉 You are all set! You can track your apps on console!