The MLflow Inference (models serve) Apps. Support for x86_64 | arm64 | ppc64le.
1.9K
GitHub Actions in the laineil/containers/app/pysvc project builds and pushes this image to Docker Hub.
Please visit the project documentation site for help to use and contribute to this image and others.
The container can be easily setup with the following environment variables:
MLF_WORKERS: Number of gunicorn worker processes to handle requests. Default: 1MLF_EX_MODEL: Whether to enable mlflow example model (see mlflow-example) or not. Default: enable (disable)MLF_MODEL_URI: Set up the model URI (see Storage Format). Default: exmodel
$ docker run -it \
--name mlflow_infrence \
-p [expose_port]:5000 \
laineil/mlflow-inference:[tag]
$ docker run -it \
--name mlflow_infrence \
-v [model_uri_path]:/volume/mlflow/[model_uri] \
-p [expose_port]:5000 \
-e MLF_EX_MODEL=disable \
-e MLF_MODEL_URI=[model_uri] \
laineil/mlflow-inference:[tag]
$ curl -X POST http://[server_host]:[expose_port]/invocations -H "Content-Type:application/json" -d '{"dataframe_split": {"columns":["fixed acidity", "volatile acidity", "citric acid", "residual sugar", "chlorides", "free sulfur dioxide", "total sulfur dioxide", "density", "pH", "sulphates", "alcohol"],"data":[[6.2, 0.66, 0.48, 1.2, 0.029, 29, 75, 0.98, 3.33, 0.39, 12.8]]}}'
{"predictions": [x.xxxxxxxxxxxxxxx]}
$ docker logs mlflow_infrence
$ docker container stop mlflow_infrence
$ docker container rm mlflow_infrence
$ docker volume rm mlflow_infrence
BSD 3-Clause License
Copyright (c) 2023, Laineil
All rights reserved.
Content type
Image
Digest
sha256:7d358f318…
Size
354.2 MB
Last updated
3 months ago
docker pull laineil/mlflow-inference