๐จ๐ปโ๐ป Ever tried deploying a model? and ended up entangled in scripts?
Hereโs how MLEM lets us do โsingle command deploymentsโ โก๏ธ:
- Deployment Options
- Setting up the Environment
- Run the command
- Getting Predictions
@Iterativeai @DVCorg
#mlem #aws #deployment
๐งต[1/6]
Hereโs how MLEM lets us do โsingle command deploymentsโ โก๏ธ:
- Deployment Options
- Setting up the Environment
- Run the command
- Getting Predictions
@Iterativeai @DVCorg
#mlem #aws #deployment
๐งต[1/6]
๐ MLEM gives us a simple and powerful API to deploy to platforms such as:
โ @awscloud Sagemaker
โ Docker Container
โ Heroku
โ Kubernetes
๐งต[2/6]
โ @awscloud Sagemaker
โ Docker Container
โ Heroku
โ Kubernetes
๐งต[2/6]
๐ฑ Setting up the Environment
Setting up the environment varies for each case, but here let's take the example of Heroku.
We can either set โHEROKU_API_KEYโ environment variable or use Heroku CLI to run โheroku loginโ.
๐งต[3/6]
Setting up the environment varies for each case, but here let's take the example of Heroku.
We can either set โHEROKU_API_KEYโ environment variable or use Heroku CLI to run โheroku loginโ.
๐งต[3/6]
๐โโ๏ธ Run the command
Run โmlem deployment run heroku app.mlem โmodel <model> โapp_name <app name>โ
Voila! All Done โ
๐งต[4/6]
Run โmlem deployment run heroku app.mlem โmodel <model> โapp_name <app name>โ
Voila! All Done โ
๐งต[4/6]
๐ Getting Predictions
Our model above is reachable by HTTP request, we can open the URL and see OpenAPI spec there, or send requests to get predictions.
We can also use built-in MLEM functionality for the same using โmlem deployment apply app.mlem data.csv --jsonโ
๐งต[5/6]
Our model above is reachable by HTTP request, we can open the URL and see OpenAPI spec there, or send requests to get predictions.
We can also use built-in MLEM functionality for the same using โmlem deployment apply app.mlem data.csv --jsonโ
๐งต[5/6]
โข โข โข
Missing some Tweet in this thread? You can try to
force a refresh
ใ
