TensorFlowOnSpark and tensorflow_scala
These are complements: TensorFlowOnSpark provides distributed training infrastructure on Spark clusters while tensorflow_scala provides the language bindings to write TensorFlow code in Scala, and they can be used together to build Scala-based distributed ML pipelines.
About TensorFlowOnSpark
yahoo/TensorFlowOnSpark
TensorFlowOnSpark brings TensorFlow programs to Apache Spark clusters.
This project helps data scientists and machine learning engineers run their TensorFlow deep learning models on large Apache Spark and Hadoop clusters. You can take your existing TensorFlow code and run it in a distributed fashion, processing vast datasets managed by Spark. It's designed for professionals building and deploying machine learning solutions at scale.
About tensorflow_scala
eaplatanios/tensorflow_scala
TensorFlow API for the Scala Programming Language
This is a tool for machine learning engineers and data scientists who primarily use Scala for their development. It allows you to build, train, and deploy machine learning models, especially neural networks, using the TensorFlow framework directly within Scala. You can input raw data, define your model architecture, train it, and then analyze the results, similar to how Python users work with TensorFlow.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work