Awesome Python Data Science Overview

Probably the best curated list of data science software in Python.

🏠 Home · 🔥 Feed · 📮 Subscribe · ❤️ Sponsor · 😺 krzjoa/awesome-python-data-science · ⭐ 1.8K · 🏷️ Programming Languages

[ Daily / Weekly / Overview ]

Awesome Python Data Science

Probably the best-curated list of data science software in Python

Contents

• Contents
• Machine Learning
  • General Purpose Machine Learning
  • Automated Machine Learning
  • Ensemble Methods
  • Imbalanced Datasets
  • Random Forests
  • Extreme Learning Machine
  • Kernel Methods
  • Gradient Boosting
• Deep Learning
  • PyTorch
  • TensorFlow
  • MXNet
  • Others
• Web Scraping
• Data Manipulation
  • Data Frames
  • Pipelines
  • Data-centric AI
  • Synthetic Data
• Feature Engineering
  • General
  • Feature Selection
• Visualization
  • General Purposes
  • Interactive plots
  • Map
  • Automatic Plotting
  • NLP
• Deployment
• Model Explanation
• Reinforcement Learning
• Probabilistic Methods
• Genetic Programming
• Optimization
• Time Series
• Natural Language Processing
• Computer Audition
• Computer Vision
• Statistics
• Distributed Computing
• Experimentation
• Data Validation
• Evaluation
• Computations
• Spatial Analysis
• Quantum Computing
• Conversion
• Related Resources
• Contributing
• License

Machine Learning

General Purpose Machine Learning

• scikit-learn - Machine learning in Python.
• Shogun - Machine learning toolbox.
• xLearn (⭐3k) - High Performance, Easy-to-use, and Scalable Machine Learning Package.
• cuML (⭐3.2k) - RAPIDS Machine Learning Library.
• modAL (⭐1.9k) - Modular active learning framework for Python3.
• Sparkit-learn (⭐1.1k) - PySpark + scikit-learn = Sparkit-learn.
• mlpack (⭐4.2k) - A scalable C++ machine learning library (Python bindings).
• dlib (⭐12k) - Toolkit for making real-world machine learning and data analysis applications in C++ (Python bindings).
• MLxtend (⭐4.3k) - Extension and helper modules for Python's data analysis and machine learning libraries.
• hyperlearn (⭐1.4k) - 50%+ Faster, 50%+ less RAM usage, GPU support re-written Sklearn, Statsmodels.
• Reproducible Experiment Platform (REP) (⭐666) - Machine Learning toolbox for Humans.
• scikit-multilearn (⭐812) - Multi-label classification for python.
• seqlearn (⭐655) - Sequence classification toolkit for Python.
• pystruct (⭐667) - Simple structured learning framework for Python.
• sklearn-expertsys (⭐483) - Highly interpretable classifiers for scikit learn.
• RuleFit (⭐342) - Implementation of the rulefit.
• metric-learn (⭐1.3k) - Metric learning algorithms in Python.
• pyGAM (⭐755) - Generalized Additive Models in Python.
• Karate Club (⭐1.8k) - An unsupervised machine learning library for graph-structured data.
• Little Ball of Fur (⭐624) - A library for sampling graph structured data.
• causalml (⭐3.8k) - Uplift modeling and causal inference with machine learning algorithms.

Automated Machine Learning

• TPOT (⭐8.9k) - Automated Machine Learning tool that optimizes machine learning pipelines using genetic programming.
• auto-sklearn (⭐6.7k) - An automated machine learning toolkit and a drop-in replacement for a scikit-learn estimator.
• MLBox (⭐1.4k) - A powerful Automated Machine Learning python library.
• AutoGluon (⭐5.3k) - AutoML for Image, Text, Tabular, Time-Series, and MultiModal Data.

Ensemble Methods

• ML-Ensemble - High performance ensemble learning.
• Stacking (⭐196) - Simple and useful stacking library written in Python.
• stacked_generalization (⭐114) - Library for machine learning stacking generalization.
• vecstack (⭐668) - Python package for stacking (machine learning technique).

Imbalanced Datasets

• imbalanced-learn (⭐6.2k) - Module to perform under-sampling and over-sampling with various techniques.
• imbalanced-algorithms (⭐220) - Python-based implementations of algorithms for learning on imbalanced data.

Random Forests

• rpforest (⭐211) - A forest of random projection trees.
• sklearn-random-bits-forest (⭐8) - Wrapper of the Random Bits Forest program written by (Wang et al., 2016).
• rgf_python (⭐364) - Python Wrapper of Regularized Greedy Forest.

Extreme Learning Machine

• Python-ELM (⭐514) - Extreme Learning Machine implementation in Python.
• Python Extreme Learning Machine (ELM) (⭐84) - A machine learning technique used for classification/regression tasks.
• hpelm (⭐175) - High-performance implementation of Extreme Learning Machines (fast randomized neural networks).

Kernel Methods

• pyFM (⭐896) - Factorization machines in python.
• fastFM (⭐1k) - A library for Factorization Machines.
• tffm (⭐784) - TensorFlow implementation of an arbitrary order Factorization Machine.
• liquidSVM (⭐57) - An implementation of SVMs.
• scikit-rvm (⭐205) - Relevance Vector Machine implementation using the scikit-learn API.
• ThunderSVM (⭐1.5k) - A fast SVM Library on GPUs and CPUs.

Gradient Boosting

• XGBoost (⭐24k) - Scalable, Portable, and Distributed Gradient Boosting.
• LightGBM (⭐15k) - A fast, distributed, high-performance gradient boosting.
• CatBoost (⭐7k) - An open-source gradient boosting on decision trees library.
• ThunderGBM (⭐653) - Fast GBDTs and Random Forests on GPUs.

Deep Learning

PyTorch

• PyTorch (⭐63k) - Tensors and Dynamic neural networks in Python with strong GPU acceleration.
• pytorch-lightning (⭐22k) - PyTorch Lightning is just organized PyTorch.
• torchvision (⭐13k) - Datasets, Transforms, and Models specific to Computer Vision.
• torchtext (⭐3.2k) - Data loaders and abstractions for text and NLP.
• torchaudio (⭐2k) - An audio library for PyTorch.
• ignite (⭐4.2k) - High-level library to help with training neural networks in PyTorch.
• skorch (⭐5k) - A scikit-learn compatible neural network library that wraps PyTorch.
• pytorch_geometric (⭐17k) - Geometric Deep Learning Extension Library for PyTorch.
• Catalyst (⭐3.1k) - High-level utils for PyTorch DL & RL research.
• pytorch_geometric_temporal (⭐1.9k) - Temporal Extension Library for PyTorch Geometric.
• ChemicalX (⭐610) - A PyTorch-based deep learning library for drug pair scoring.

TensorFlow

• TensorFlow (⭐171k) - Computation using data flow graphs for scalable machine learning by Google.
• TensorLayer (⭐7.1k) - Deep Learning and Reinforcement Learning Library for Researcher and Engineer.
• TFLearn (⭐9.6k) - Deep learning library featuring a higher-level API for TensorFlow.
• Sonnet (⭐9.5k) - TensorFlow-based neural network library.
• tensorpack (⭐6.3k) - A Neural Net Training Interface on TensorFlow.
• Polyaxon (⭐3.3k) - A platform that helps you build, manage and monitor deep learning models.
• NeuPy (⭐739) - NeuPy is a Python library for Artificial Neural Networks and Deep Learning (previously: ).
• tfdeploy (⭐350) - Deploy TensorFlow graphs for fast evaluation and export to TensorFlow-less environments running numpy.
• tensorflow-upstream (⭐626) - TensorFlow ROCm port.
• TensorFlow Fold (⭐1.8k) - Deep learning with dynamic computation graphs in TensorFlow.
• tensorlm (⭐63) - Wrapper library for text generation/language models at char and word level with RNN.
• TensorLight (⭐10) - A high-level framework for TensorFlow.
• Mesh TensorFlow (⭐1.4k) - Model Parallelism Made Easier.
• Ludwig (⭐8.8k) - A toolbox that allows one to train and test deep learning models without the need to write code.
• Keras - A high-level neural networks API running on top of TensorFlow.
• keras-contrib (⭐1.6k) - Keras community contributions.
• Hyperas (⭐2.2k) - Keras + Hyperopt: A straightforward wrapper for a convenient hyperparameter.
• Elephas (⭐1.6k) - Distributed Deep learning with Keras & Spark.
• Hera (⭐493) - Train/evaluate a Keras model, and get metrics streamed to a dashboard in your browser.
• Spektral (⭐2.2k) - Deep learning on graphs.
• qkeras (⭐442) - A quantization deep learning library.

MXNet

• MXNet (⭐20k) - Lightweight, Portable, Flexible Distributed/Mobile Deep Learning with Dynamic, Mutation-aware Dataflow Dep Scheduler.
• Gluon (⭐2.3k) - A clear, concise, simple yet powerful and efficient API for deep learning (now included in MXNet).
• MXbox (⭐31) - Simple, efficient, and flexible vision toolbox for the mxnet framework.
• gluon-cv (⭐5.5k) - Provides implementations of the state-of-the-art deep learning models in computer vision.
• gluon-nlp (⭐2.5k) - NLP made easy.
• Xfer (⭐247) - Transfer Learning library for Deep Neural Networks.
• MXNet (⭐29) - HIP Port of MXNet.

Others

• jax (⭐22k) - Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more.
• Tangent (⭐2.3k) - Source-to-Source Debuggable Derivatives in Pure Python.
• autograd (⭐6.2k) - Efficiently computes derivatives of numpy code.
• Myia (⭐456) - Deep Learning framework (pre-alpha).
• nnabla (⭐2.6k) - Neural Network Libraries by Sony.
• Caffe (⭐33k) - A fast open framework for deep learning.
• hipCaffe (⭐126) - The HIP port of Caffe.

DISCONTINUED PROJECTS (⭐1.8k)

Web Scraping

• BeautifulSoup: The easiest library to scrape static websites for beginners
• Scrapy: Fast and extensible scraping library. Can write rules and create customized scraper without touching the core
• Selenium: Use Selenium Python API to access all functionalities of Selenium WebDriver in an intuitive way like a real user.
• Pattern (⭐8.4k): High level scraping for well-establish websites such as Google, Twitter, and Wikipedia. Also has NLP, machine learning algorithms, and visualization
• twitterscraper (⭐2.2k): Efficient library to scrape Twitter

Data Manipulation

Data Frames

• pandas - Powerful Python data analysis toolkit.
• pandas_profiling (⭐10k) - Create HTML profiling reports from pandas DataFrame objects
• cuDF (⭐5.3k) - GPU DataFrame Library.
• blaze (⭐3.1k) - NumPy and pandas interface to Big Data.
• pandasql (⭐1.2k) - Allows you to query pandas DataFrames using SQL syntax.
• pandas-gbq (⭐357) - pandas Google Big Query.
• xpandas (⭐25) - Universal 1d/2d data containers with Transformers .functionality for data analysis by The Alan Turing Institute.
• pysparkling (⭐256) - A pure Python implementation of Apache Spark's RDD and DStream interfaces.
• Arctic (⭐2.9k) - High-performance datastore for time series and tick data.
• datatable (⭐1.7k) - Data.table for Python.
• koalas (⭐3.2k) - pandas API on Apache Spark.
• modin (⭐8.4k) - Speed up your pandas workflows by changing a single line of code.
• swifter (⭐2.2k) - A package that efficiently applies any function to a pandas dataframe or series in the fastest available manner.
• pandas_flavor (⭐272) - A package that allows writing your own flavor of Pandas easily.
• pandas-log (⭐211) - A package that allows providing feedback about basic pandas operations and finds both business logic and performance issues.
• vaex (⭐7.8k) - Out-of-Core DataFrames for Python, ML, visualize and explore big tabular data at a billion rows per second.
• xarray (⭐2.9k) - Xarray combines the best features of NumPy and pandas for multidimensional data selection by supplementing numerical axis labels with named dimensions for more intuitive, concise, and less error-prone indexing routines.
• sk-transformer (⭐4) - A collection of various pandas & scikit-learn compatible transformers for all kinds of preprocessing and feature engineering steps
• polars (⭐14k) - A fast multi-threaded, hybrid-out-of-core DataFrame library.

Pipelines

• pdpipe (⭐700) - Sasy pipelines for pandas DataFrames.
• SSPipe - Python pipe (|) operator with support for DataFrames and Numpy, and Pytorch.
• pandas-ply (⭐189) - Functional data manipulation for pandas.
• Dplython (⭐755) - Dplyr for Python.
• sklearn-pandas (⭐2.7k) - pandas integration with sklearn.
• Dataset (⭐184) - Helps you conveniently work with random or sequential batches of your data and define data processing.
• pyjanitor (⭐1.1k) - Clean APIs for data cleaning.
• meza (⭐404) - A Python toolkit for processing tabular data.
• Prodmodel (⭐54) - Build system for data science pipelines.
• dopanda (⭐447) - Hints and tips for using pandas in an analysis environment.
• Hamilton (⭐893) - A microframework for dataframe generation that applies Directed Acyclic Graphs specified by a flow of lazily evaluated Python functions.

Data-centric AI

• cleanlab (⭐5.3k) - The standard data-centric AI package for data quality and machine learning with messy, real-world data and labels.
• snorkel (⭐5.4k) - A system for quickly generating training data with weak supervision.
• dataprep (⭐1.5k) - Collect, clean, and visualize your data in Python with a few lines of code.

Synthetic Data

• ydata-synthetic (⭐891) - A package to generate synthetic tabular and time-series data leveraging the state-of-the-art generative models.

Feature Engineering

General

• Featuretools (⭐6.5k) - Automated feature engineering.
• Feature Engine (⭐1.3k) - Feature engineering package with sklearn-like functionality.
• skl-groups (⭐41) - A scikit-learn addon to operate on set/"group"-based features.
• Feature Forge (⭐380) - A set of tools for creating and testing machine learning features.
• few (⭐47) - A feature engineering wrapper for sklearn.
• scikit-mdr (⭐122) - A sklearn-compatible Python implementation of Multifactor Dimensionality Reduction (MDR) for feature construction.
• tsfresh (⭐7.1k) - Automatic extraction of relevant features from time series.
• dirty_cat (⭐672) - Machine learning on dirty tabular data (especially: string-based variables for classifcation and regression).
• NitroFE (⭐102) - Moving window features.

Feature Selection

• scikit-feature (⭐1.3k) - Feature selection repository in Python.
• boruta_py (⭐1.3k) - Implementations of the Boruta all-relevant feature selection method.
• BoostARoota (⭐190) - A fast xgboost feature selection algorithm.
• scikit-rebate (⭐375) - A scikit-learn-compatible Python implementation of ReBATE, a suite of Relief-based feature selection algorithms for Machine Learning.
• zoofs (⭐177) - A feature selection library based on evolutionary algorithms.

Visualization

General Purposes

• Matplotlib (⭐17k) - Plotting with Python.
• seaborn (⭐10k) - Statistical data visualization using matplotlib.
• prettyplotlib (⭐1.6k) - Painlessly create beautiful matplotlib plots.
• python-ternary (⭐617) - Ternary plotting library for Python with matplotlib.
• missingno (⭐3.5k) - Missing data visualization module for Python.
• chartify (⭐3.3k) - Python library that makes it easy for data scientists to create charts.
• physt (⭐120) - Improved histograms.

Interactive plots

• animatplot (⭐397) - A python package for animating plots built on matplotlib.
• plotly - A Python library that makes interactive and publication-quality graphs.
• Bokeh (⭐17k) - Interactive Web Plotting for Python.
• Altair - Declarative statistical visualization library for Python. Can easily do many data transformation within the code to create graph
• bqplot (⭐3.4k) - Plotting library for IPython/Jupyter notebooks
• pyecharts (⭐13k) - Migrated from Echarts (⭐54k), a charting and visualization library, to Python's interactive visual drawing library.

Map

• folium - Makes it easy to visualize data on an interactive open street map
• geemap (⭐2.6k) - Python package for interactive mapping with Google Earth Engine (GEE)

Automatic Plotting

• HoloViews (⭐2.4k) - Stop plotting your data - annotate your data and let it visualize itself.
• AutoViz (⭐1.2k): Visualize data automatically with 1 line of code (ideal for machine learning)
• SweetViz (⭐2.3k): Visualize and compare datasets, target values and associations, with one line of code.

NLP

• pyLDAvis (⭐1.7k): Visualize interactive topic model

Deployment

• fastapi - Modern, fast (high-performance), a web framework for building APIs with Python
• streamlit - Make it easy to deploy the machine learning model
• gradio (⭐13k) - Create UIs for your machine learning model in Python in 3 minutes.
• datapane - A collection of APIs to turn scripts and notebooks into interactive reports.
• binder - Enable sharing and execute Jupyter Notebooks

Model Explanation

• dalex (⭐1.2k) - moDel Agnostic Language for Exploration and explanation.
• Shapley (⭐190) - A data-driven framework to quantify the value of classifiers in a machine learning ensemble.
• Alibi (⭐1.9k) - Algorithms for monitoring and explaining machine learning models.
• anchor (⭐744) - Code for "High-Precision Model-Agnostic Explanations" paper.
• aequitas (⭐526) - Bias and Fairness Audit Toolkit.
• Contrastive Explanation (⭐42) - Contrastive Explanation (Foil Trees).
• yellowbrick (⭐3.9k) - Visual analysis and diagnostic tools to facilitate machine learning model selection.
• scikit-plot (⭐2.3k) - An intuitive library to add plotting functionality to scikit-learn objects.
• shap (⭐19k) - A unified approach to explain the output of any machine learning model.
• ELI5 (⭐2.6k) - A library for debugging/inspecting machine learning classifiers and explaining their predictions.
• Lime (⭐10k) - Explaining the predictions of any machine learning classifier.
• FairML (⭐341) - FairML is a python toolbox auditing the machine learning models for bias.
• L2X (⭐115) - Code for replicating the experiments in the paper Learning to Explain: An Information-Theoretic Perspective on Model Interpretation.
• PDPbox (⭐727) - Partial dependence plot toolbox.
• PyCEbox (⭐143) - Python Individual Conditional Expectation Plot Toolbox.
• Skater (⭐1.1k) - Python Library for Model Interpretation.
• model-analysis (⭐1.2k) - Model analysis tools for TensorFlow.
• themis-ml (⭐109) - A library that implements fairness-aware machine learning algorithms.
• treeinterpreter (⭐722) - Interpreting scikit-learn's decision tree and random forest predictions.
• AI Explainability 360 (⭐1.3k) - Interpretability and explainability of data and machine learning models.
• Auralisation (⭐39) - Auralisation of learned features in CNN (for audio).
• CapsNet-Visualization (⭐387) - A visualization of the CapsNet layers to better understand how it works.
• lucid (⭐4.5k) - A collection of infrastructure and tools for research in neural network interpretability.
• Netron (⭐21k) - Visualizer for deep learning and machine learning models (no Python code, but visualizes models from most Python Deep Learning frameworks).
• FlashLight - Visualization Tool for your NeuralNetwork.
• tensorboard-pytorch (⭐7.5k) - Tensorboard for PyTorch (and chainer, mxnet, numpy, ...).
• mxboard (⭐327) - Logging MXNet data for visualization in TensorBoard.

Reinforcement Learning

• OpenAI Gym (⭐30k) - A toolkit for developing and comparing reinforcement learning algorithms.
• Coach (⭐2.2k) - Easy experimentation with state-of-the-art Reinforcement Learning algorithms.
• garage (⭐1.6k) - A toolkit for reproducible reinforcement learning research.
• OpenAI Baselines (⭐14k) - High-quality implementations of reinforcement learning algorithms.
• Stable Baselines (⭐3.7k) - A set of improved implementations of reinforcement learning algorithms based on OpenAI Baselines.
• RLlib - Scalable Reinforcement Learning.
• Horizon (⭐3.3k) - A platform for Applied Reinforcement Learning.
• TF-Agents (⭐2.4k) - A library for Reinforcement Learning in TensorFlow.
• TensorForce (⭐3.2k) - A TensorFlow library for applied reinforcement learning.
• TRFL (⭐3.1k) - TensorFlow Reinforcement Learning.
• Dopamine (⭐10k) - A research framework for fast prototyping of reinforcement learning algorithms.
• keras-rl (⭐5.4k) - Deep Reinforcement Learning for Keras.
• ChainerRL (⭐1.1k) - A deep reinforcement learning library built on top of Chainer.

Probabilistic Methods

• pyro (⭐7.8k) - A flexible, scalable deep probabilistic programming library built on PyTorch.
• pomegranate (⭐3k) - Probabilistic and graphical models for Python.
• ZhuSuan - Bayesian Deep Learning.
• PyMC (⭐7.3k) - Bayesian Stochastic Modelling in Python.
• InferPy (⭐141) - Deep Probabilistic Modelling Made Easy.
• GPflow - Gaussian processes in TensorFlow.
• PyStan (⭐243) - Bayesian inference using the No-U-Turn sampler (Python interface).
• sklearn-bayes (⭐486) - Python package for Bayesian Machine Learning with scikit-learn API.
• pgmpy (⭐2.3k) - A python library for working with Probabilistic Graphical Models.
• skpro (⭐112) - Supervised domain-agnostic prediction framework for probabilistic modelling by The Alan Turing Institute.
• PtStat (⭐109) - Probabilistic Programming and Statistical Inference in PyTorch.
• PyVarInf (⭐343) - Bayesian Deep Learning methods with Variational Inference for PyTorch.
• emcee (⭐1.3k) - The Python ensemble sampling toolkit for affine-invariant MCMC.
• hsmmlearn (⭐71) - A library for hidden semi-Markov models with explicit durations.
• pyhsmm (⭐528) - Bayesian inference in HSMMs and HMMs.
• GPyTorch (⭐3k) - A highly efficient and modular implementation of Gaussian Processes in PyTorch.
• MXFusion (⭐100) - Modular Probabilistic Programming on MXNet.
• sklearn-crfsuite (⭐420) - A scikit-learn-inspired API for CRFsuite.

Genetic Programming

• gplearn (⭐1.3k) - Genetic Programming in Python.
• DEAP (⭐5k) - Distributed Evolutionary Algorithms in Python.
• karoo_gp (⭐149) - A Genetic Programming platform for Python with GPU support.
• monkeys (⭐117) - A strongly-typed genetic programming framework for Python.
• sklearn-genetic (⭐265) - Genetic feature selection module for scikit-learn.

Optimization

• Optuna (⭐7.6k) - A hyperparameter optimization framework.
• Spearmint (⭐1.5k) - Bayesian optimization.
• BoTorch (⭐2.5k) - Bayesian optimization in PyTorch.
• scikit-opt (⭐3.8k) - Heuristic Algorithms for optimization.
• sklearn-genetic-opt (⭐192) - Hyperparameters tuning and feature selection using evolutionary algorithms.
• SMAC3 (⭐796) - Sequential Model-based Algorithm Configuration.
• Optunity (⭐402) - Is a library containing various optimizers for hyperparameter tuning.
• hyperopt (⭐6.6k) - Distributed Asynchronous Hyperparameter Optimization in Python.
• hyperopt-sklearn (⭐1.4k) - Hyper-parameter optimization for sklearn.
• sklearn-deap (⭐715) - Use evolutionary algorithms instead of gridsearch in scikit-learn.
• sigopt_sklearn (⭐73) - SigOpt wrappers for scikit-learn methods.
• Bayesian Optimization (⭐6.6k) - A Python implementation of global optimization with gaussian processes.
• SafeOpt (⭐113) - Safe Bayesian Optimization.
• scikit-optimize (⭐2.5k) - Sequential model-based optimization with a scipy.optimize interface.
• Solid (⭐564) - A comprehensive gradient-free optimization framework written in Python.
• PySwarms (⭐1k) - A research toolkit for particle swarm optimization in Python.
• Platypus (⭐442) - A Free and Open Source Python Library for Multiobjective Optimization.
• GPflowOpt (⭐258) - Bayesian Optimization using GPflow.
• POT (⭐1.8k) - Python Optimal Transport library.
• Talos (⭐1.6k) - Hyperparameter Optimization for Keras Models.
• nlopt (⭐1.4k) - Library for nonlinear optimization (global and local, constrained or unconstrained).
• OR-Tools - An open-source software suite for optimization by Google; provides a unified programming interface to a half dozen solvers: SCIP, GLPK, GLOP, CP-SAT, CPLEX, and Gurobi.

Time Series

• sktime (⭐6.1k) - A unified framework for machine learning with time series.
• darts (⭐5.4k) - A python library for easy manipulation and forecasting of time series.
• statsforecast (⭐2.2k) - Lightning fast forecasting with statistical and econometric models.
• mlforecast (⭐229) - Scalable machine learning-based time series forecasting.
• neuralforecast (⭐1.2k) - Scalable machine learning-based time series forecasting.
• tslearn (⭐2.4k) - Machine learning toolkit dedicated to time-series data.
• tick (⭐415) - Module for statistical learning, with a particular emphasis on time-dependent modeling.
• greykite (⭐1.7k) - A flexible, intuitive, and fast forecasting library next.
• Prophet (⭐16k) - Automatic Forecasting Procedure.
• PyFlux (⭐2k) - Open source time series library for Python.
• bayesloop (⭐126) - Probabilistic programming framework that facilitates objective model selection for time-varying parameter models.
• luminol (⭐1.1k) - Anomaly Detection and Correlation library.
• dateutil - Powerful extensions to the standard datetime module
• maya (⭐3.4k) - makes it very easy to parse a string and for changing timezones
• Chaos Genius (⭐552) - ML powered analytics engine for outlier/anomaly detection and root cause analysis

Natural Language Processing

• spaCy - Industrial-Strength Natural Language Processing.
• NLTK (⭐12k) - Modules, data sets, and tutorials supporting research and development in Natural Language Processing.
• CLTK (⭐763) - The Classical Language Toolkik.
• gensim - Topic Modelling for Humans.
• pyMorfologik (⭐18) - Python binding for Morfologik.
• skift (⭐233) - Scikit-learn wrappers for Python fastText.
• Phonemizer (⭐798) - Simple text-to-phonemes converter for multiple languages.
• flair (⭐13k) - Very simple framework for state-of-the-art NLP.

Computer Audition

• librosa (⭐5.7k) - Python library for audio and music analysis.
• Yaafe (⭐232) - Audio features extraction.
• aubio (⭐2.9k) - A library for audio and music analysis.
• Essentia (⭐2.3k) - Library for audio and music analysis, description, and synthesis.
• LibXtract (⭐216) - A simple, portable, lightweight library of audio feature extraction functions.
• Marsyas (⭐371) - Music Analysis, Retrieval, and Synthesis for Audio Signals.
• muda (⭐214) - A library for augmenting annotated audio data.
• madmom (⭐1k) - Python audio and music signal processing library.

Computer Vision

• OpenCV (⭐67k) - Open Source Computer Vision Library.
• scikit-image (⭐5.2k) - Image Processing SciKit (Toolbox for SciPy).
• imgaug (⭐13k) - Image augmentation for machine learning experiments.
• imgaug_extension - Additional augmentations for imgaug.
• Augmentor (⭐4.9k) - Image augmentation library in Python for machine learning.
• albumentations (⭐12k) - Fast image augmentation library and easy-to-use wrapper around other libraries.

Statistics

• pandas_summary (⭐460) - Extension to pandas dataframes describe function.
• Pandas Profiling (⭐10k) - Create HTML profiling reports from pandas DataFrame objects.
• statsmodels (⭐8.2k) - Statistical modeling and econometrics in Python.
• stockstats (⭐1.1k) - Supply a wrapper StockDataFrame based on the pandas.DataFrame with inline stock statistics/indicators support.
• weightedcalcs (⭐98) - A pandas-based utility to calculate weighted means, medians, distributions, standard deviations, and more.
• scikit-posthocs (⭐270) - Pairwise Multiple Comparisons Post-hoc Tests.
• Alphalens (⭐2.6k) - Performance analysis of predictive (alpha) stock factors.

Distributed Computing

• Horovod (⭐13k) - Distributed training framework for TensorFlow, Keras, PyTorch, and Apache MXNet.
• PySpark - Exposes the Spark programming model to Python.
• Veles (⭐897) - Distributed machine learning platform.
• Jubatus (⭐702) - Framework and Library for Distributed Online Machine Learning.
• DMTK (⭐2.8k) - Microsoft Distributed Machine Learning Toolkit.
• PaddlePaddle (⭐20k) - PArallel Distributed Deep LEarning.
• dask-ml (⭐838) - Distributed and parallel machine learning.
• Distributed (⭐1.4k) - Distributed computation in Python.

Experimentation

• mlflow (⭐14k) - Open source platform for the machine learning lifecycle.
• Neptune - A lightweight ML experiment tracking, results visualization, and management tool.
• dvc (⭐11k) - Data Version Control | Git for Data & Models | ML Experiments Management.
• envd (⭐1.5k) - 🏕️ machine learning development environment for data science and AI/ML engineering teams.
• Sacred (⭐4k) - A tool to help you configure, organize, log, and reproduce experiments.
• Ax (⭐2k) - Adaptive Experimentation Platform.

Data Validation

• great_expectations (⭐8k) - Always know what to expect from your data.
• pandera (⭐2k) - A lightweight, flexible, and expressive statistical data testing library.
• deepchecks (⭐2.4k) - Validation & testing of ML models and data during model development, deployment, and production.
• evidently (⭐3.1k) - Evaluate and monitor ML models from validation to production.
• TensorFlow Data Validation (⭐699) - Library for exploring and validating machine learning data.

Evaluation

• recmetrics (⭐476) - Library of useful metrics and plots for evaluating recommender systems.
• Metrics (⭐1.6k) - Machine learning evaluation metric.
• sklearn-evaluation (⭐0) - Model evaluation made easy: plots, tables, and markdown reports.
• AI Fairness 360 (⭐2k) - Fairness metrics for datasets and ML models, explanations, and algorithms to mitigate bias in datasets and models.

Computations

• numpy - The fundamental package needed for scientific computing with Python.
• Dask (⭐11k) - Parallel computing with task scheduling.
• bottleneck (⭐865) - Fast NumPy array functions written in C.
• CuPy (⭐6.7k) - NumPy-like API accelerated with CUDA.
• scikit-tensor (⭐396) - Python library for multilinear algebra and tensor factorizations.
• numdifftools (⭐201) - Solve automatic numerical differentiation problems in one or more variables.
• quaternion (⭐540) - Add built-in support for quaternions to numpy.
• adaptive (⭐752) - Tools for adaptive and parallel samping of mathematical functions.
• NumExpr (⭐1.9k) - A fast numerical expression evaluator for NumPy that comes with an integrated computing virtual machine to speed calculations up by avoiding memory allocation for intermediate results.

Spatial Analysis

• GeoPandas (⭐3.6k) - Python tools for geographic data.
• PySal (⭐1.1k) - Python Spatial Analysis Library.

Quantum Computing

• qiskit - Qiskit is an open-source SDK for working with quantum computers at the level of circuits, algorithms, and application modules.
• cirq (⭐3.7k) - A python framework for creating, editing, and invoking Noisy Intermediate Scale Quantum (NISQ) circuits.
• PennyLane (⭐1.7k) - Quantum machine learning, automatic differentiation, and optimization of hybrid quantum-classical computations.
• QML (⭐176) - A Python Toolkit for Quantum Machine Learning.

Conversion

• sklearn-porter (⭐1.2k) - Transpile trained scikit-learn estimators to C, Java, JavaScript, and others.
• ONNX (⭐14k) - Open Neural Network Exchange.
• MMdnn (⭐5.7k) - A set of tools to help users inter-operate among different deep learning frameworks.

Related Resources

• Best of Machine Learning Libraries - A curated list of best Machine Learning libraries in Python.

Contributing

Contributions are welcome! 😎
Read the contribution guideline.

License

This work is licensed under the Creative Commons Attribution 4.0 International License - CC BY 4.0