ml-tooling / best-of-ml-python Goto Github PK

Project details:

• Project Name: small-text
• Github URL: https://github.com/webis-de/small-text
• Category: nlp
• License: MIT License
• Package Managers: pypi:small-text conda:conda-forge/small-text

Additional context:

Small-Text provides state-of-the-art Active Learning for Text Classification. Several pre-implemented Query Strategies, Initialization Strategies, and Stopping Critera are provided, which can be easily mixed and matched to build active learning experiments or applications.

Project details:

• Project Name: MLup
• Github URL: https://github.com/nxexox/pymlup
• Category: Model Serialization & Deployment
• License: MIT
• Package Managers: pypi:pymlup

Additional context:

MLup is framework for easy and fast run ML in production.

Update details:

https://github.com/ml-tooling/best-of-ml-python#database-clients
POINTS/LINKS TO
https://github.com/ml-tooling/best-of-web-python

Is that intentional?

Project details:

• Project Name: julearn
• Github URL: https://github.com/juaml/julearn
• Category: sklearn-utils
• License: AGPL-3.0 license
• Package Managers: pypi:julearn
  Additional context:

From julearn docs - What is Julearn.

In the same way as seaborn provides an abstraction of matplotlib’s functionality aiming for powerful data visualization with minor coding, we built julearn on top of scikit-learn.

Julearn is a library that provides users with the possibility of easy testing ML models directly from pandas dataframes, while keeping the flexibiliy of using scikit-learn’s models.

Project details:

• Project Name: Population Shift Monitoring
• Github URL: https://github.com/ing-bank/popmon
• Category: data-viz or others
• License: MIT Licence
• Package Managers: pypi:popmon

Additional context:

popmon is a package that allows one to check the stability of a dataset. popmon works with both pandas and spark datasets.

Project details:

• Project Name: underthesea
• Github URL: https://github.com/undertheseanlp/underthesea
• Category: Text Data & NLP
• License: GPL-3.0 license
• Package Managers: pypi:underthesea

Additional context:

Underthesea is a suite of open source Python modules data sets and tutorials supporting research and development in Vietnamese Natural Language Processing.

Project details:

• Project Name: H2O-3
• Github URL: https://github.com/h2oai/h2o-3
• Category: distributed-ml
• License: Apache-2.0
• Package Managers: pypi:h2o conda:h2oai/h2o cran:h2o

Project details:
An open-source framework to evaluate, test and monitor ML models in production.

• Project Name: evidently
• Github URL: https://github.com/evidentlyai/evidently
• Category: Model Interpretability or Others
• License: Apache 2.0
• Package Managers: pypi:evidently conda:conda-forge/evidently

Additional context:

Project details:

• Project Name: GFPGAN Image restoration
• Github URL: GFPGAN
• Category: Computer Vision
• License: under multiple licenses
• Package Managers: pypi

Project details:

• Project Name: Geomstats
• Github URL: https://github.com/geomstats/geomstats
• Category: Machine Learning frameworks
• License: MIT
• Package Managers: pypi:geomstats conda:conda-forge/geomstats

Hey 👋 Creator of Ciphey here. Wanted to say thank you for including us in the ML hall of fame! :)

Project details:

• Project Name: tqdm
• Github URL: https://github.com/tqdm/tqdm
• Category: Other
• License: MPL-2.0, MIT
• Package Managers: pypi:tqdm, conda:defaults/tqdm, conda:conda-forge/tqdm, snap:tqdm, dockerhub:tqdm/tqdm, apt:python-tqdm, apt:python3-tqdm

Additional context:

In general useful (e.g. data pre-processing with tqdm.pandas) as well as ML-specific (e.g. tqdm.keras)

If you think it's appropriate I can open a PR.

• Project Name: Upgini
• Github URL: https://github.com/upgini/upgini/blob/main/README.md#upgini--dataset-search-library-for-your-machine-learning-pipelines
• Category: Data Loading & Extraction
• Package Managers: pypi:upgini

Additional context:
Upgini is a python library for for an automated useful dataset search to boost supervised ML tasks.
upgini.com

Update details:

• Project Name: ivy

The pypi link for ivy has been updated from ivy-core to just ivy
(ivy-core not (found)[https://pypi.org/project/ivy-core/] in the pypi website)
correct pypi link for ivy is https://pypi.org/project/ivy/

The wrong pypi link reflects incorrect download statistics(also it is inactive), the correct is supposed to be around 1500+ monthly downloads, with the old ivy-core here the figure is about 36 which is misrepresentative and might give a skewed score to the project.
Also the latest figure is from 01.06.2022 because of the same which is very outdated.

Project details:

• Project Name: nnAudio
• Github URL: https://github.com/KinWaiCheuk/nnAudio
• Category: audio
• License: MIT
• Package Managers: pypi:nnAudio

Additional context:

Project details:

• Project Name: sentence-transformers
• Github URL: https://github.com/UKPLab/sentence-transformers
• Category: Text Data & NLP
• License: Apache-2.0 License
• Package Managers: pypi:sentence-transformers

Additional context:

Project details:

• Project Name: Lea
• Github URL: https://bitbucket.org/piedenis/lea
• Category: Probabilty and Statistics
• License:GNU gpl v3
• Package Managers: pip

Additional context:

This project has no Github, only bitbucket

Update details:

• Project Name: Jina

• Category: Machine Learning Framework

• Description: A deep learning-powered search framework for building neural search in the cloud.

Additional context:

I'm from Jina team.
Jina is currently listed under NLP category here. But Jina is a Search Framework that can do more than just NLP search.
We are listed in LF AI Landscape under Deep Learning Platform category (Link Here), so Machine Learning Framework/Platform is most likely the right category for Jina.
Thank you!

Project details:

• Project Name: Lightning-Hydra-Template
• Github URL: https://github.com/ashleve/lightning-hydra-template
• Category: ml-experiments
• License: MIT License
• Package Managers: None except Github. It is just a project template.

Additional context:

Found. via atomistic deep learning package SchNetPack.

Currently many projects with high quality-score are buried behind the "Show 17 hidden projects ..." buttons, whereas some projects with much lower quality-score are on the front page.

Project details:

• Project Name:
• Github URL:
• Category:
• License:
• Package Managers:

Additional context:

Project details:

• Project Name: CML - Continuous Machine Learning
• Github URL: https://github.com/iterative/cml
• Category: data-pipelines
• License: Apache-2.0
• Package Managers:

Project details:

• Project Name: Orbit (Object-oRiented BayesIan Time-series)
• Github URL: https://github.com/uber/orbit
• Category: Time-series
• License: APACHE 2.0
• Package Managers:
  numpy>=1.18.2
  pandas>=1.0.3
  pystan==2.19.1.1
  matplotlib>=3.3.2
  scipy>=1.4.1
  torch
  tqdm
  seaborn>=0.10.0
  pyro-ppl>=1.4.0
  statsmodels>=0.11.1
  arviz
  Additional context:

Project details:

• Project Name: pydeck
• Github URL: https://github.com/visgl/deck.gl/tree/master/bindings/pydeck
• Category: geospatial-data
• License: Apache-2.0 License
• Package Managers: pypi:pydeck conda:conda-forge/pydeck

Additional context:

• Large-scale interactive data visualization in Python
• The pydeck library is a set of Python bindings for making spatial visualizations with deck.gl, optimized for a Jupyter environment.

Project details:

• Project Name: Auto_ViML
• Github URL: https://github.com/AutoViML/Auto_ViML
• Category: Hyperparameter Optimizers and AutoML
• License: Apache 2
• Package Managers:
  • npm:
  • [ X] PyPI:
  • Conda:
  • Dockerhub:
  • Gem:
  • CRAN:
  • Helmhub:
  • Maven:
  • Go:
  • Packagist:
  • NuGet:
  • Brew:
  • Apt:
  • ...

Additional context:
Auto_ViML: Automatically Build Multiple ML Models with a Single Line of Code.

Project details:

• Project Name:
• Github URL: https://github.com/bodywork-ml/bodywork-core
• Category: Data Pipelines & Streaming
• License: AGPL-3.0
• Package Managers: pypi:bodywork-core

Additional context:

DevOps tool for machine learning projects developed in Python. It helps you to serve models as microservices, schedule batch jobs and run reproducible pipelines, in containers on Kubernetes.

Project details:

• Project Name: H2O Wave
• Github URL: https://github.com/h2oai/wave
• Category:
• License: Apache-2.0
• Package Managers: pypi:h2o-wave conda:h2oai/h2o-wave
  Additional context:

H2O Wave: Realtime Web Apps and Dashboards for Python and R

The link for the best-of-generator repository (https://github.com/ml-tooling/best-of-generator) isn't working. Has it been moved/made private?

Project details:

• Project Name: Panel
• Github URL: https://github.com/holoviz/panel
• Category: Data Visualization
• License: BSD 3-Clause License.
• Package Managers: Conda
  Additional context:

Project details:
Missing data imputation is a widely used method for dealing with missing data in machine learning and statistical workflows. miceforest aims to provide extremely accurate imputations using lightgbm, while being as lightweight and fast as possible. This package can:

1. Impute multiple datasets, so the user can perform Multiple Imputation by Chained Equations (MICE)
2. Plot the imputed correlations, distributions, feature importance, and more
3. Train models on 1 dataset, and impute a different dataset (useful for production environments)
4. Can be GPU accelerated through the lightgbm api.
5. Can impute data in place, which means the dataset never has to be copied. Useful for huge datasets.

• Project Name:
• Github URL: https://github.com/AnotherSamWilson/miceforest
• Category: Other? New Category? See Additional Context.
• License: MIT
• Package Managers: pypi, conda

Additional context:
I am wondering if Missing Data Imputation should be it's own category - it is very often used in machine learning, especially predictive modeling. There is another missing data imputation project already on here, fancyimpute. What do you think?

Project details:

• Project Name: TSFEL - Time Series Feature Extraction Library
• Github URL: https://github.com/fraunhoferportugal/tsfel
• Category: time-series-data
• License: BSD3
• Package Managers: pypi:tsfel

Additional context:

• TSFEL is optimized for time series and automatically extracts over 60 different features on the statistical, temporal and spectral domains.
• Disclaimer: I'm one of the maintainers.
• Thank you for your effort on compiling this curated list!

Project details:

• Project Name: numpy-ml
• Github URL: https://github.com/ddbourgin/numpy-ml
• Category: machine-learning-frameworks
• License:GPL-3.0 license
• Package Managers: pypi:numpy-ml

Additional context:

Project details:

• Project Name: skops
• Github URL: https://github.com/skops-dev/skops
• Category: sklearn-utils
• License: MIT license
• Package Managers: pypi:skops conda-forge/skops

Additional context:

Recommended for model persistence in scikit-learn 1.2.0 user guide, chapter model persistence.

Conflict of interest: No affiliation with the project.

Hi,
I would like to add Neptune.ai to the list.

Project details:

name: "Neptune.ai"
github_id: "neptune-ai/neptune-client"
pypi_id: "neptune-client"
conda_id: "conda-forge/neptune-client"
category: "ml-experiments"

Additional context:

Flexible metadata store for MLOps, built for research and production teams that run a lot of experiments.

Thanks in advance!

Project details:

• Project Name: Tiny Genetic Programming in Python
• Github URL: https://github.com/moshesipper/tiny_gp
• Category: others
• License: GNU General Public License v3.0
• Package Managers:

Additional context:

Hi,

I would like to add WEFE to the list.

Project details:

• Project Name: WEFE: The Word Embeddings Fairness Evaluation Framework
• Github URL: https://github.com/dccuchile/wefe
• Category: Model Interpretability
• License: MIT
• Package Managers: pip:wefe

Additional context:

• WEFE is a framework that standardizes the bias measurement and mitigation in Word Embeddings models

Thanks!

Project details:

• Project Name: MLEM
• Github URL: https://github.com/iterative/mlem
• Category: model-serialisation
• License: Apache-2.0
• Package Managers: pypi:mlem

Update details:

xlrd no longer supports any formats other than .xls due to security issues and recommends using openpyxl instead.

• Project Name:

xlrd

Additional context:

• xlrd documentation showing a warning that it no longer supports anything other than xls: https://xlrd.readthedocs.io/en/latest/
• A stackoverflow answer by the xlrd maintainer where he recommends using openpyxl instead: https://stackoverflow.com/a/65266497/4800086
• openpyxl documentation: https://openpyxl.readthedocs.io/en/stable/
• openpyxl source: https://foss.heptapod.net/openpyxl/openpyxl

ChatGPT独享账号
http://bwww.vip/igit.html

Project details:

• Project Name: featurewiz
• Github URL: https://github.com/AutoViML/featurewiz
• Category: Hyperparameter Optimization & AutoML
• License: Apache 2
• Package Managers:
  • npm:
  • [ X] PyPI:
  • Conda:
  • Dockerhub:
  • Gem:
  • CRAN:
  • Helmhub:
  • Maven:
  • Go:
  • Packagist:
  • NuGet:
  • Brew:
  • Apt:
  • ...

Additional context:

featurewiz: Select the best features from your data set fast with a single line of code.

Project details:

• Project Name: vulkan-kompute
• Github URL: https://github.com/EthicalML/vulkan-kompute
• Category: GPU Utilities
• License: Apache-2.0 License
• Package Managers:
  • pypi:kp

Additional context:

Video from conference

Video: https://www.youtube.com/watch?v=AJRyZ09IUdg

I'm one of the developers of InterpretML. InterpretML is currently listed in the interpretability category, which makes sense and is also our primary focus. Recently though, we have added the ability to create differentially private EBM models (DP-EBMs) which we and at least one other outside paper have found perform significantly better than other known differentially private models. InterpretML would therefore seem to fit into the privacy category as well.

Papers:
http://proceedings.mlr.press/v139/nori21a/nori21a.pdf

https://arxiv.org/pdf/2210.02910.pdf

Project details:

• Project Name: Netron
• Github URL: https://github.com/lutzroeder/netron
• Category: interpretability
• License: MIT Licence
• Package Managers: pypi:netron

Additional context:
Netron is a viewer for neural network, deep learning and machine learning models.

Netron supports ONNX, TensorFlow Lite, Caffe, Keras, Darknet, PaddlePaddle, ncnn, MNN, Core ML, RKNN, MXNet, MindSpore Lite, TNN, Barracuda, Tengine, CNTK, TensorFlow.js, Caffe2 and UFF.

Netron has experimental support for PyTorch, TensorFlow, TorchScript, OpenVINO, Torch, Vitis AI, kmodel, Arm NN, BigDL, Chainer, Deeplearning4j, MediaPipe, ML.NET and scikit-learn.

Project details:

• Project Name: Auto_TS
• Github URL: https://github.com/AutoViML/Auto_TS
• Category: Time Series Data
• License: Apache 2
• Package Managers:
  • npm:
  • [X ] PyPI:
  • Conda:
  • Dockerhub:
  • Gem:
  • CRAN:
  • Helmhub:
  • Maven:
  • Go:
  • Packagist:
  • NuGet:
  • Brew:
  • Apt:
  • ...

Additional context:
Auto_TS: Automatically build ARIMA, SARIMAX, VAR, FB Prophet and ML Models on Time Series data sets with a Single Line of Code.

Project details:

• Project Name: scikit-activeml
• Github URL: https://github.com/scikit-activeml/scikit-activeml
• Category: Probabilistics & Statistics
• License: BSD-3-Clause license
• Package Managers: pypi:scikit-activeml

Additional information:

Preprint: Kottke et al., 2021.

Conflicts of interest: I am not affiliated with this package.

Project details:

• Project Name: Streamsync
• Github URL: streamsync-cloud/streamsync
• Category: Others
• License: Apache-2
• Package Managers: pypi:streamsync

Additional context:

No-code in the front, Python in the back. An open source framework for creating data apps.

Update details:

Hi,
Thanks for listing SMAC3 here. However, I'm surprised to see SMAC3 listed as BSD 1-clause, although it is BSD 3-clause. Could you please look into this again?

• Project Name: SMAC3

Thanks,
Marius

Project details:

• Project Name: jupyterlab-executor
• Github URL: https://github.com/gavincyi/jupyterlab-executor
• Category: JupyterLab Extensions
• License: BSD-3
• Package Managers: pypi and npm

Additional context:

Project details:

• Project Name: Maze
• Github URL: https://github.com/enlite-ai/maze
• Category: reinforcement-learning
• License: Custom
• Package Managers: pypi:maze-rl dockerhub:enliteai/maze

Additional context:

New application-oriented deep reinforcement learning framework. Disclaimer: I'm one of the maintainers.