Machine learning lib.
Project description
modelkit
Python framework for production ML systems.
modelkit is a Python framework meant to make your ML models robust, reusable and performant in all situations you need to use them.
It is meant to bridge the gap between the different uses of your algorithms. With modelkit you can ensure that the same exact code will run in production, on your machine, or on data processing pipelines.
Features
modelkit's key features are:
- simple
modelkitis just a Python library, use pip to install it and you are done. - custom
modelkitis useful whenever you need to go beyond off-the-shelf models: custom processing, heuristics, business logic, different frameworks, etc. - framework agnostic you bring your own framework to the table, and you can use whatever code or library you want. Similarly,
modelkitis not opinionated about how you build or train your models. - organized
modelkitencourages you to version and share you ML library and artifacts with others, as a Python package or as a service. Let others use and evaluate your models! - fast
modelkitadd minimal overhead to prediction calls. Model predictions can be batched for speed (you define the batching logic). - fast to code Models only need to define their prediction logic and that's it. No cumbersome pre or postprocessing logic, branching options, etc... The boilerplate code is minimal and sensible.
- fast to deploy Models can be served in a single CLI call using fastapi
And more:
- composable Models can depend on other models, and evaluate them however you need to
- extensible Models can rely on arbitrary supporting configurations files called assets hosted on local or cloud object stores
- type-safe Models' inputs and outputs can be validated by pydantic, you get type annotations for your predictions and can catch errors with static type analysis tools during development.
- async Models support async and sync prediction functions.
modelkitsupports calling async code from sync code so you don't have to suffer from partially async code. - testable Models carry their own unit test cases, and unit testing fixtures are available for pytest
- robust
modelkithelps you follow software development best practices: all configurations and artifacts are explicitly versioned and tested.
Installation
Install with pip:
pip install modelkit
Documentation
Refer to the documentation for more information.
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 Distributions
No source distribution files available for this release.See tutorial on generating distribution archives.
Built Distribution
modelkit-0.0.9-py3-none-any.whl
(61.9 kB
view details)
File details
Details for the file modelkit-0.0.9-py3-none-any.whl.
File metadata
- Download URL: modelkit-0.0.9-py3-none-any.whl
- Upload date:
- Size: 61.9 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/3.4.1 importlib_metadata/4.0.1 pkginfo/1.7.0 requests/2.25.0 requests-toolbelt/0.9.1 tqdm/4.54.1 CPython/3.7.10
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | 11e550297b6c99f512ec1746d4376ab0413cd03e17bd24f3346983e8810a34ca |
|
| MD5 | 1d856a602a0526dbc6a3426b9fe10106 |
|
| BLAKE2b-256 | e4740b1e344317ae2a2f2112f40280e3e12f9801b4906173bd79cdb03efde3d5 |
