Metadata-Version: 2.1
Name: mml-suggest
Version: 0.3.1
Summary: This is the MML suggest plugin, providing methods to compile pipelines from previous knowledge.
Home-page: https://git.dkfz.de/imsy/ise/mml
Author: Patrick Godau
Author-email: patrick.godau@dkfz-heidelberg.de
License: MIT
Classifier: Natural Language :: English
Classifier: Development Status :: 4 - Beta
Classifier: Environment :: Console
Classifier: Intended Audience :: Developers
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Topic :: Scientific/Engineering :: Image Recognition
Classifier: Topic :: Scientific/Engineering :: Information Analysis
Classifier: Topic :: Scientific/Engineering :: Image Processing
Classifier: Topic :: Software Development :: Libraries :: Application Frameworks
Classifier: Topic :: Software Development :: Version Control :: Git
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.8
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Environment :: GPU
Classifier: Intended Audience :: Science/Research
Classifier: Typing :: Typed
Classifier: License :: OSI Approved :: MIT License
Requires-Python: >=3.8
Description-Content-Type: text/markdown
License-File: LICENSE.txt
Requires-Dist: mml-core>=1.0.0
Requires-Dist: mml-similarity

# MML Suggest plugin

This plugin provides a method to compile pipeline blueprints for new tasks. 

# Install

> Note: this plugin depends on another plugin (mml-similarity) which is installed automatically - though 
> this has no side effects.

```commandline
pip install mml-suggest
```


# Usage

To leverage existing models and task distances these must be reused:

```commandline
mml suggest pivot.name=my_target_task +reuse.fed=my_distance_project reuse.models=[some_proj,another_proj,...]
```

This will compile a pipeline blueprint for `my_target_task` based on the `fed` task similarities from 
`my_distance_project` leveraging all models previously trained in the specified projects.

To use the blueprint call 

```commandline
mml train pivot.name=my_target_task reuse.blueprint=my_blueprint_proj#the_blueprint_number mode.use_blueprint=true
```
