Subcategories
Packages
Calculates thermodynamic and transport properties of water, steam, ice, seawater, and heavy water according to IAPWS industrial standards (IF97, 95, 06, 08, 17).
Reduces noise in audio and time-domain signals using spectral gating, with both stationary and non-stationary algorithms and optional PyTorch-based implementations.
Pandarallel parallelizes pandas DataFrame operations across multiple CPU cores by replacing standard pandas methods with parallel equivalents, requiring only a one-line code change.
However, do not rely on it for production systems where you need active maintenance or compatibility with future pandas/Python releases.
Colour provides algorithms and datasets for colour science tasks including colour space conversions, chromatic adaptation, colour appearance models, and colour blindness simulation.
Install it if you need colour space conversions, colour appearance modelling, or spectral data handling.
Pint-Pandas adds physical unit support to pandas DataFrames, allowing you to store, manipulate, and compute with quantities that have units (like torque in lbf·ft or angular velocity in rpm) directly within DataFrame columns.
Python wrappers for SymEngine, a fast C++ symbolic manipulation library, enabling symbolic computation and algebraic operations in Python.
Optimizes PyTorch einsum operations and functions containing them by applying opt_einsum strategies through PyTorch FX compute graphs.
rapids-logger provides a CMake-based framework to generate project-specific logger classes that wrap spdlog without exposing its symbols, enabling safe coexistence of different spdlog versions in the same environment.
DEAP is an evolutionary computation framework for implementing genetic algorithms, genetic programming, evolution strategies, and multi-objective optimization with support for arbitrary data structures and parallelization.
Install it if you need a mature, flexible framework for genetic algorithms, genetic programming, or multi-objective optimization.
PyHDFE absorbs high-dimensional fixed effects from regression data using optimized algorithms, designed for integration into statistical projects rather than as a standalone regression interface.
Provides low-level Python and Cython bindings for HIP, enabling direct access to AMD GPU compute capabilities from Python code.
A Rust-based execution engine for Azure Machine Learning Data Prep that accelerates data loading, transformation, and writing operations in ML workflows.
Provides a reference abstract syntax tree (AST) for OpenQASM 3 quantum programs, with parsing and AST manipulation tools for building compiler passes in Python.
However, be aware that the API is explicitly unstable and subject to breaking changes as the OpenQASM 3 language itself evolves—pin your version carefully in…
fastdtw computes approximate Dynamic Time Warping (DTW) alignments between sequences with linear O(N) time and memory complexity, returning both the distance and alignment path.
Nevergrad is a gradient-free optimization library that finds optimal parameter values for functions without requiring derivatives, supporting continuous, discrete, and mixed variable spaces.
Install it if you're doing hyperparameter tuning, black-box optimization, or algorithm configuration without access to gradients.
STUMPY computes the matrix profile for time series data, enabling pattern discovery, anomaly detection, and other time series mining tasks through efficient algorithms optimized with Numba.
Install it if you need matrix profile computation or the downstream mining tasks it enables; skip it if your time series work is limited to forecasting or simple…
Provides native code extensions for the azureml-dataprep package, enabling optimized data preparation operations for Azure Machine Learning workflows.
Install it if azureml-dataprep is a dependency of your project.
Adds geospatial semantic functions to FabricDataFrame, enabling automatic discovery and application of location-aware operations based on detected latitude/longitude columns and metadata.
However, it is in beta status and has modest adoption (16 repository stars), so expect potential API changes and verify compatibility with your specific versions of…
Converts text between Traditional Chinese, Simplified Chinese, and Japanese Kanji, supporting character-level and phrase-level conversion with regional vocabulary variants.
libcudf-cu12 is a GPU-accelerated C++ library providing Apache Arrow-compliant data structures and fundamental algorithms for tabular data processing on NVIDIA CUDA 12 GPUs.
scikit-rf is an object-oriented Python library for RF and microwave engineering, providing tools to work with network parameters and Touchstone files.
Install it if you work with network parameters or Touchstone files in Python.
Provides Python clients to query the ENTSO-E API for European electricity transmission system data, returning results as raw XML/ZIP or parsed Pandas DataFrames.
PyIQA provides a PyTorch-based toolbox for computing image quality assessment metrics, supporting both full-reference and no-reference methods with GPU acceleration and calibration against official implementations.
Provides Python bindings to CityHash, a fast non-cryptographic hashing algorithm, specifically tuned for ClickHouse protocol compatibility.
Install only if you specifically need ClickHouse protocol compatibility; for general-purpose hashing, use the original python-cityhash package instead.
Nipype provides a uniform Python interface to neuroimaging software packages, allowing you to build reproducible analysis pipelines that combine tools from different packages and run them in parallel.
Install only if you have external neuroimaging software already available on your system.
Converts fitted Scikit-Learn pipelines to PMML (Predictive Model Markup Language) for deployment in other environments, supporting a wide range of estimators and transformers.
Provides C++ implementations of multiple tokenizers (SentencePiece, Tiktoken, Huggingface, Llama2.c, Tekken) with Python bindings for use in PyTorch-based LLM applications and ExecuTorch runners.
ExecuTorch exports and runs PyTorch models on mobile, embedded, and edge devices with support for hardware acceleration backends like XNNPACK, Core ML, and QNN.
Python SDK for connecting to and running computer vision models and workflows on a local or remote Inference server, enabling image and video stream processing with object detection, classification, segmentation, and custom model inference.
Install it if you need to programmatically interact with Inference from Python; if you only need the server itself, install inference-cli instead.
Provides panel regression, instrumental variable estimators, system regression, and factor asset pricing models that extend statsmodels for econometric and financial analysis.
YDF trains, evaluates, and interprets decision forest models (Random Forests, Gradient Boosted Trees, and others) through a Python wrapper around the C++ Yggdrasil Decision Forests library.
Solves unconstrained and linearly constrained nonlinear minimization problems without computing derivatives, using direct search methods that work well when objective function evaluations are expensive or noisy.
A Flask-based web dashboard for monitoring RQ job queues, workers, and jobs in real time, deployable as a standalone service or integrated into an existing Flask application.
Provides Python dataclass models for the Airbyte Protocol, enabling type-safe work with Airbyte data pipeline specifications with minimal performance overhead.
Install it if you need Airbyte protocol types; skip it otherwise.
PySCF is a Python framework for quantum chemistry simulations, providing tools to model molecular systems and compute electronic structure properties using methods like Hartree-Fock, density functional theory, and post-Hartree-Fock approaches.
UHI defines and documents a standard interface for histogram objects across libraries, providing type protocols for runtime checking and tools to adapt NumPy-style histogram tuples into a common format.
Computes stransverse mass (MT2) for particle physics calculations, supporting both symmetric and asymmetric cases with vectorized numpy array operations.
RMM provides a common interface for customizing GPU and host memory allocation, with implementations for device memory pooling, pinned host memory, and other allocation strategies.
cuDF is a GPU-accelerated DataFrame library that provides pandas-like APIs for tabular data processing on NVIDIA GPUs, with optional zero-code-change acceleration for existing pandas code via cudf.pandas.
Install only if you have an NVIDIA GPU with compatible CUDA 12 drivers and can manage the 18 runtime dependencies (including CUDA toolkit and GPU libraries).
Converts and computes statistics across sparse, dense, GPU, and distributed array formats with a unified interface.