# microsoft/SynapseML

Simple and Distributed Machine Learning Python Library porting ML algorithms for Spark

Repository: https://github.com/microsoft/SynapseML
Canonical: https://ross.abutalabs.com/products/synapseml
Homepage: http://aka.ms/spark
Language: Scala
License: MIT
License Family: permissive
Topics: spark, pyspark, azure, scala, microsoft, ml, machine-learning, databricks, cognitive-services, lightgbm, http, model-deployment, deep-learning, ai, apache-spark, data-science, synapse, big-data, onnx, opencv
Last push: 2026-08-26T08:23:45+00:00

## Health v2 (maintenance only)
Score: 88/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 66, longevity 100
- inputs: {"age_days": 3376, "days_push": 7, "days_rel": 148, "gap_med": 64, "n_releases_24m": 8}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 5240, forks 866 (observed 2026-08-28T04:09:13.542640+00:00)

## What it is
SynapseML (formerly MMLSpark) is an open-source machine learning library built on Apache Spark that provides simple, composable, distributed APIs for tasks like text analytics, vision, and anomaly detection. It shares the SparkML/MLlib API so models integrate seamlessly into existing Spark workflows across Python, R, Scala, Java, and .NET.

## Use cases
- train machine learning models on large distributed Spark clusters
- run distributed LightGBM training on Spark DataFrames
- integrate Azure Cognitive Services into Spark pipelines
- build scalable text analytics and NLP pipelines
- perform distributed deep learning and ONNX model inference
- detect anomalies at scale in big data workflows
- deploy and evaluate ML models on multi-node clusters

## When to choose
- you already use Apache Spark and want ML pipelines with the same API
- you need distributed training of algorithms like LightGBM on big data
- you want to combine cloud AI services with Spark workflows
- you need multi-language access (Python, R, Scala, Java, .NET) to ML tools

## When to avoid
- you work on a single machine with small datasets where scikit-learn suffices
- you need a lightweight library without JVM/Spark dependencies
- your project does not use Spark or a distributed computing framework

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, nlp, computer-vision, etl, data-science
- domain: machine-learning, big-data, data-science, deep-learning, artificial-intelligence
- platform: jvm, python, cloud
- tags: apache-spark, pyspark, distributed-ml, lightgbm, onnx, databricks, azure, mmlspark, spark

## Member repositories
- microsoft/SynapseML (main) score 88

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:13.542640+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:59:37.593474+00:00, confidence not recorded.
  - readme: https://github.com/microsoft/SynapseML (fetched 2026-08-28T04:09:13.542640+00:00, sha 5afd4b83cfff)
- Data as of 2026-08-30T08:39:29.467469+00:00.