Metadata-Version: 2.1
Name: mir_ml_utils
Version: 0.0.1
Summary: Various utilities for machine learning training in Python 3. for the mir project
Project-URL: Homepage, https://github.com/Navalmartin/mir_ml_utils
Project-URL: Bug Tracker, https://github.com/Navalmartin/mir_ml_utils/issues
Author-email: Alexandros Giavaras <a.giavaras@gmail.com>
License-File: LICENSE
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Requires-Python: >=3.9
Description-Content-Type: text/markdown

[![mir (Marine Inspection by Remote)](https://github.com/pockerman/mir/actions/workflows/python-app.yml/badge.svg)](https://github.com/pockerman/mir/actions/workflows/python-app.yml)

# mir-ml-engine

The ML engine for the mir (Marine Inspection by Remote) app.

## Dependencies

- PyTorch
- Numpy
- Pillow
- matplolib

- sphinx (if you want to build the documentation)

The project also gets benefited from <a href="https://pypi.org/project/easyfsl/">easyfsl</a> and
<a href="https://github.com/sicara/easy-few-shot-learning">easy-few-shot-learning</a>. However,
you don't need to install this as a fork is maintained within mir-engine

### Testing

There are additional dependencies if you want to run the tests

- pytest
- coverage
- flake8

In order to test locally, you can run the

```commandline
local_test_pipeline.sh
```

The script first runs ```flake8``` and then ```pytest``` with coverage. Finally, it produces
a report ```coverage_report.txt```.

## Installation 




