https://github.com/nextml-code/ml-workflow

Pytorch project template and required tools
https://github.com/nextml-code/ml-workflow

Last synced: 1 day ago
JSON representation

Pytorch project template and required tools

• Host: GitHub
• URL: https://github.com/nextml-code/ml-workflow
• Owner: nextml-code
• License: apache-2.0
• Created: 2020-09-23T09:32:24.000Z (over 4 years ago)
• Default Branch: master
• Last Pushed: 2020-11-13T07:13:32.000Z (about 4 years ago)
• Last Synced: 2025-01-03T10:19:19.925Z (12 days ago)
• Language: Python
• Size: 412 KB
• Stars: 4
• Watchers: 3
• Forks: 0
• Open Issues: 4
• Metadata Files:
  • Readme: README.rst
  • License: LICENSE

Awesome Lists containing this project

README

        
===========

ML Workflow

===========

.. image:: https://badge.fury.io/py/ml-workflow.svg

       :target: https://badge.fury.io/py/ml-workflow

ML workflow contains our process of bringing a project to fruition as

efficiently as possible. This is subject to change as we iterate and improve.

This package implements tools and missing features to help bridge the gap

between frameworks and libraries that we utilize.

The main packages and tools that we build around are:

- pytorch

- ignite

- pytorch-datastream

- guild

See the `documentation `_

for more information.

Install in existing project

===========================

.. code-block::

    pip install ml-workflow

Create new project with MNIST template

======================================

.. code-block::

    mkdir new-project

    cd new-project

    virtualenv venv -p python3.8

    source venv/bin/activate

    pip install ml-workflow

    python -m workflow.setup_project

    pip install -r requirements.txt

    pip install -r dev_requirements.txt

    pip freeze > dev_requirements.txt

    # reactivate environment to find guild

    deactivate

    source venv/bin/activate

You can train a model and inspect the training with:

.. code-block::

    guild run prepare

    guild run train

    guild tensorboard

Development

===========

Prepare and run tests

.. code-block::

    git clone [email protected]:aiwizo/ml-workflow.git

    cd ml-workflow

    virtualenv venv --python python3.8

    source venv/bin/activate

    pip install -r requirements.txt

    pip install -r dev_requirements.txt

    pip install pytest

    python -m pytest

Test template

=============

.. code-block::

    ./setup_template.py

    ./test_template.py

Use development version in project

==================================

The following steps will create a link to the local directory and any changes 

made to the package there will directly carry over to your project environment.

.. code-block::

    cd path/to/my/project

    source venv/bin/activate

    cd path/to/work/area

    git clone [email protected]:aiwizo/ml-workflow.git

    cd ml-workflow

    pip install -e .