Matcha: A open source tool for provisioning MLOps environments to the cloud.
Project description
matcha
🍵
Welcome to matcha
, the open source tool for provisioning MLOps environments to the cloud.
Data Scientists and practitioners build ML pipelines and productionize them with the following capabilities:
- A way to run model training pipelines
- A way to track experiments
- A way to deploy and serve models
The current version of matcha
provisions the above with a single command, using sensible defaults for the infrastructure.
More components could be added here, for example, managing and versioning datasets, managing models (registration and governance), and monitoring models. Adding all of these is on our roadmap.
matcha
uses open source to solve this problem.
By default, matcha
will provision
an MLOps environment to Azure with sensible defaults - enough for you to get going with cloud-based training and deployment. From there, you can use matcha
to get
information about your provisioned environment.
Please see the following for more information, including install instructions, documentation, and how you can contribute to matcha
:
Licence
This library is released under the Apache License. See LICENSE.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for matcha_ml-0.2.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 223ae73c1c932286ca89432581be15565ea46939adef2ae1f7d0eb58b50fa408 |
|
MD5 | d61b6ae0c96c11de7969404d8a9345f1 |
|
BLAKE2b-256 | 776dfee0b047de28775cf7e7826676d7972d193d3d654cb59b95880ce2efe79c |