Determined is an all-in-one deep learning platform, compatible with PyTorch and TensorFlow.

It takes care of:

• Distributed training for faster results.
• Hyperparameter tuning for obtaining the best models.
• Resource management for cutting cloud GPU costs.
• Experiment tracking for analysis and reproducibility.

How Determined Works

The main components of Determined are the Python library, the command line interface (CLI), and the Web UI.

Python Library

Use the Python library to make your existing PyTorch or Tensorflow code compatible with Determined.

You can do this by organizing your code into one of the class-based APIs:

from determined.pytorch import PyTorchTrial

class YourExperiment(PyTorchTrial):
  def __init__(self, context):
    ...

Or by using just the functions you want, via the Core API:

import determined as det

with det.core.init() as core_context:
    ...

Command Line Interface (CLI)

You can use the CLI to:

• Start a Determined cluster locally:

det deploy local cluster-up

• Launch Determined on cloud services, such as Amazon Web Services (AWS) or Google Cloud Platform (GCP):

det deploy aws up

• Train your models:

det experiment create gpt.yaml .

Configure everything from distributed training to hyperparameter tuning using YAML files:

resources:
  slots_per_trial: 8
  priority: 1
hyperparameters:
  learning_rate:
    type: double
    minval: .0001
    maxval: 1.0
searcher:
  name: adaptive_asha
  metric: validation_loss
  smaller_is_better: true

Web UI

Use the Web UI to view loss curves, hyperparameter plots, code and configuration snapshots, model registries, cluster utilization, debugging logs, performance profiling reports, and more.

Installation

To install the CLI:

pip install determined

Then use det deploy to start the Determined cluster locally, or on cloud services like AWS and GCP.

For installation details, visit the the cluster deployment guide for your environment:

• Local (on-prem)
• AWS
• GCP
• Kubernetes
• Slurm/PBS

Examples

Get familiar with Determined by exploring the 30+ examples in the examples folder and the determined-examples repo.

Documentation

• Documentation
• Quick Start Guide
• Tutorials:
  • PyTorch MNIST Tutorial
  • TensorFlow Keras MNIST Tutorial
• User Guides:
  • Core API
  • PyTorch API
  • Keras API
  • DeepSpeed API

Community

If you need help, want to file a bug report, or just want to keep up-to-date with the latest news about Determined, please join the Determined community!

• Slack is the best place to ask questions about Determined and get support. Click here to join our Slack.
• You can also follow us on YouTube and Twitter.
• You can also join the community mailing list to ask questions about the project and receive announcements.
• To report a bug, open an issue on GitHub.
• To report a security issue, email [email protected].

Contributing

Contributor's Guide

License

Apache V2