Submarine is Cloud Native Machine Learning Platform.

Clone this repo:
  1. 6250f65 Preparing for 0.5.0 release by Kevin Su · 4 days ago master release-0.5.0-RC1
  2. d34b374 SUBMARINE-663. Use md5sum command in by Kevin Su · 5 days ago
  3. 2ff3339 SUBMARINE-662. Clarify duplicate experiment error by ByronHsu · 8 days ago
  4. 747b8cd SUBMARINE-660. Remove submarine-workbench/workbench-web/node/node by Kevin Su · 12 days ago release-0.5.0-RC0
  5. 1757975 SUBMARINE-656. Fix the wrong message when creating k8s object fail by Lisa · 14 days ago


Build Status License HitCount PyPI version

What is Apache Submarine?

Apache Submarine (Submarine for short) is an End-to-End Machine Learning PLATFORM to allow data scientists to create end-to-end machine learning workflows. To elaborate, on Submarine, data scientists can finish each stage in the ML model lifecycle, including data exploration, data pipeline creation, model training, serving, and monitoring.

Why Submarine?

Some open-source and commercial projects are trying to build an end-to-end ML platform. What's the vision of Submarine?


  1. Many platforms lack easy-to-use user interfaces (API, SDK, and IDE, etc.)
  2. In the same company, data scientists in different teams usually spend much time on developments of existing feature sets and models.
  3. Data scientists put emphasis on domain-specific tasks (e.g. Click-Through-Rate), but they need to implement their models from scratch with SDKs provided by existing platforms.
  4. Many platforms lack a unified workbench to manage each component in the ML lifecycle.

Theodore Levitt once said:

“People don’t want to buy a quarter-inch drill. They want a quarter-inch hole.”

Goals of Submarine

Model Training (Experiment)

  • Run/Track distributed training experiment on prem or cloud via easy-to-use UI/API/SDK.
  • Easy for data scientists to manage versions of experiment and dependencies of environment
  • Support popular machine learning frameworks, including TensorFlow, PyTorch, Horovod, and MXNet
  • Provide pre-defined template for data scientists to implement domain-specific tasks easily (e.g. using DeepFM template to build a CTR prediction model)
  • Support many compute resources (e.g. CPU and GPU, etc.)
  • Support Kubernetes and YARN
  • Pipeline is also on the backlog, we will look into pipeline for training in the future.

Notebook Service

  • Submarine aims to provide a notebook service (e.g. Jupyter notebook) which allows users to manage notebook instances running on the cluster.

Model Management (Serving/versioning/monitoring, etc.)

  • Model management for model-serving/versioning/monitoring is on the roadmap.

Easy-to-use User Interface

As mentioned above, Submarine attempts to provide Data-Scientist-friendly UI to make data scientists have a good user experience. Here're some examples.

Example: Submit a distributed Tensorflow experiment via Submarine Python SDK

Run a Tensorflow Mnist experiment

# New a submarine client of the submarine server submarine_client = submarine.ExperimentClient(host='http://localhost:8080') # The experiment's environment, could be Docker image or Conda environment based environment = EnvironmentSpec(image='apache/submarine:tf-dist-mnist-test-1.0') # Specify the experiment's name, framework it's using, namespace it will run in, # the entry point. It can also accept environment variables. etc. # For PyTorch job, the framework should be 'Pytorch'. experiment_meta = ExperimentMeta(name='mnist-dist', namespace='default', framework='Tensorflow', cmd='python /var/tf_dist_mnist/ --train_steps=100') # 1 PS task of 2 cpu, 1GB ps_spec = ExperimentTaskSpec(resources='cpu=2,memory=1024M', replicas=1) # 1 Worker task worker_spec = ExperimentTaskSpec(resources='cpu=2,memory=1024M', replicas=1) # Wrap up the meta, environment and task specs into an experiment. # For PyTorch job, the specs would be "Master" and "Worker". experiment_spec = ExperimentSpec(meta=experiment_meta, environment=environment, spec={'Ps':ps_spec, 'Worker': worker_spec}) # Submit the experiment to submarine server experiment = submarine_client.create_experiment(experiment_spec=experiment_spec) # Get the experiment ID id = experiment['experimentId']

Query a specific experiment


Wait for finish


Get the experiment's log


Get all running experiment


For a quick-start, see Submarine On K8s

Example: Submit a pre-defined experiment template job

Example: Submit an experiment via Submarine UI

(Available on 0.6.0, see Roadmap)

Architecture, Design and requirements

If you want to know more about Submarine's architecture, components, requirements and design doc, they can be found on Architecture-and-requirement

Detailed design documentation, implementation notes can be found at: Implementation notes

Apache Submarine Community

Read the Apache Submarine Community Guide

How to contribute Contributing Guide

Issue Tracking:

User Document

See User Guide Home Page

Developer Document

See Developer Guide Home Page


What to know more about what's coming for Submarine? Please check the roadmap out:


The Apache Submarine project is licensed under the Apache 2.0 License. See the LICENSE file for details.