awesome-repositories.com
Blog
awesome-repositories.com

Descoperă cele mai bune repository-uri open source cu căutare AI.

ExploreazăCăutări recomandateAlternative open-sourceSoftware self-hostedBlogHartă site
ProiectDespreCum realizăm clasamentulPresăServer MCP
LegalConfidențialitateTermeni
© 2026 Bringes Technology SRL·VAT RO45896025·hello@awesome-repositories.com
·

2 repository-uri

Awesome GitHub RepositoriesFeature Materialization Pipelines

Reading batch features from offline stores and writing them to online stores using pluggable distributed compute engines.

Distinct from Distributed Computing Engines: Distinct from Distributed Computing Engines: focuses on the specific pipeline for feature materialization, not general distributed computing.

Explore 2 awesome GitHub repositories matching data & databases · Feature Materialization Pipelines. Refine with filters or upvote what's useful.

Awesome Feature Materialization Pipelines GitHub Repositories

Găsește cele mai bune repo-uri cu AI.Vom căuta cele mai potrivite repository-uri folosind AI.
  • feast-dev/feastAvatar feast-dev

    feast-dev/feast

    6,727Vezi pe GitHub↗

    Feast is an open-source feature store for machine learning that provides a central platform for defining, storing, and serving features across both training and inference workflows. It operates as a declarative system where feature definitions are written as code in Python files, synchronized to a central registry, and made available for low-latency online retrieval or point-in-time correct historical joins for training datasets. The project abstracts storage behind a pluggable architecture, allowing offline and online backends to be swapped without changing retrieval logic, and coordinates ma

    Reads batch features from offline stores and writes them to online stores using pluggable distributed compute engines like Spark or Ray.

    Pythonbig-datadata-engineeringdata-quality
    Vezi pe GitHub↗6,727
  • datalinkdc/dinkyAvatar DataLinkDC

    DataLinkDC/dinky

    3,740Vezi pe GitHub↗

    Dinky is a real-time data platform for developing, deploying, and operating streaming applications based on Apache Flink. It functions as a SQL streaming IDE and a real-time data pipeline orchestrator, providing a web-based environment for writing and verifying queries with integrated logic plan visualization and lineage tracking. The platform acts as a distributed cluster manager, allowing the registration, monitoring, and administration of multiple processing clusters from a centralized interface. It also serves as a change data capture integration tool, synchronizing real-time database cha

    Leverages Apache Flink as a distributed processing engine to execute real-time data pipelines and SQL jobs.

    Javadatalakedatawarehouseflink
    Vezi pe GitHub↗3,740
  1. Home
  2. Data & Databases
  3. Data Engineering
  4. Distributed Compute Frameworks
  5. Distributed Computing Engines
  6. Feature Materialization Pipelines

Explorează sub-etichetele

  • AWS Lambda MaterializationRunning batch feature materialization on AWS Lambda by reading from the offline store and writing to the online store. **Distinct from Feature Materialization Pipelines:** Distinct from Feature Materialization Pipelines: focuses on using AWS Lambda as the compute engine for materialization, not general pipeline orchestration.
  • Batch Feature Materialization1 sub-tagComputes and writes the latest batch feature values into an online store for low-latency serving. **Distinct from Feature Materialization Pipelines:** Distinct from Feature Materialization Pipelines: focuses on the batch computation and write step specifically, not the full pipeline orchestration.
  • Feature View ExecutionExecutes feature view logic including transformations, aggregations, and UDFs to generate features. **Distinct from Feature Materialization Pipelines:** Distinct from Feature Materialization Pipelines: focuses on executing the feature view's transformation logic, not the pipeline that moves data between stores.
  • Flink Execution EnginesRuns batch materialization and historical retrieval pipelines using Apache Flink via the PyFlink Table API. **Distinct from Feature Materialization Pipelines:** Distinct from Feature Materialization Pipelines: specifies Flink as the compute engine for executing the pipeline, not the pipeline abstraction itself.
  • Incremental MaterializationsLoading only new or updated feature data into the online store from the last materialization point, avoiding full recomputation. **Distinct from Feature Materialization Pipelines:** Distinct from Feature Materialization Pipelines: focuses on the incremental update strategy rather than the general pipeline for reading and writing batch features.
  • Snowflake Feature MaterializationRunning batch feature materialization on a Snowflake Warehouse using Python UDFs for serialization. **Distinct from Feature Materialization Pipelines:** Distinct from Feature Materialization Pipelines: focuses on using Snowflake as the compute engine for materialization, not general pipeline orchestration.
  • Spark-Based MaterializationRuns batch materialization operations on a Spark cluster to transform and write large volumes of feature data. **Distinct from Feature Materialization Pipelines:** Distinct from Feature Materialization Pipelines: specifically targets Spark as the compute engine for large-scale materialization, not the general pipeline concept.