medigan.contribute_model package

Submodules

medigan.contribute_model.base_model_uploader module

Base Model uploader class that uploads models to medigan associated data storage services.

class medigan.contribute_model.base_model_uploader.BaseModelUploader(model_id: str, metadata: dict)[source]

Bases: object

BaseModelUploader class: Uploads a user’s model and metadata to third party storage to allow its inclusion into the medigan library.

Parameters

  • model_id (str) – The generative model’s unique id

  • metadata (dict) – The model’s corresponding metadata

model_id

The generative model’s unique id

Type

str

metadata

The model’s corresponding metadata

Type

dict

push()[source]

medigan.contribute_model.github_model_uploader module

Github Model uploader class that uploads the metadata of a new model to the medigan github repository.

class medigan.contribute_model.github_model_uploader.GithubModelUploader(model_id: str, access_token: str)[source]

Bases: medigan.contribute_model.base_model_uploader.BaseModelUploader

GithubModelUploader class: Pushes the metadata of a user’s model to the medigan repo, where it creates a dedicated github issue.

Parameters

  • model_id (str) – The generative model’s unique id

  • access_token (str) – a personal access token linked to your github user account, used as means of authentication

model_id

The generative model’s unique id

Type

str

access_token

a personal access token linked to your github user account, used as means of authentication

Type

str

Update package_link in the model’s metadata if current package_link does not containing a valid url.

Parameters

  • metadata (dict) – The model’s corresponding metadata

  • package_link (str) – the new package link to used to replace the old one

  • is_update_forced (bool) – flag to update metadata even though metadata already contains a valid url in its package_link

Returns

Returns the updated metadata dict with replaced package_link if replacement was applicable.

Return type

dict

push(metadata: dict, package_link: Optional[str] = None, creator_name: str = 'n.a.', creator_affiliation: str = 'n.a.', model_description: str = 'n.a.')[source]

Upload the model’s metadata inside a github issue to the medigan github repository.

To add your model to medigan, your metadata will be reviewed on Github and added to medigan’s official model metadata

The medigan repository issues page: https://github.com/RichardObi/medigan/issues

Get your Github access token here: https://github.com/settings/tokens

Parameters

  • metadata (dict) – The model’s corresponding metadata

  • package_link – a package link

  • creator_name (str) – the creator name that will appear on the corresponding github issue

  • creator_affiliation (str) – the creator affiliation that will appear on the corresponding github issue

  • model_description (list) – the model_description that will appear on the corresponding github issue

Returns

Returns the url pointing to the corresponding issue on github

Return type

str

medigan.contribute_model.model_contributor module

Model contributor class that tests models, creates metadata entries, uploads and contributes them to medigan.

class medigan.contribute_model.model_contributor.ModelContributor(model_id: str, init_py_path: str)[source]

Bases: object

ModelContributor class: Contributes a user’s local model to the public medigan library

Parameters

  • model_id (str) – The generative model’s unique id

  • init_py_path (str) – The path to the local model’s __init__.py file needed for importing and running this model.

model_id

The generative model’s unique id

Type

str

init_py_path

The path to the local model’s __init__.py file needed for importing and running this model.

Type

str

package_path

Path as string to the generative model’s python package

Type

str

package_name

Name of the model’s python package i.e. the name of the model’s zip file and unzipped package folder

Type

str

metadata_file_path

Path as string to the generative model’s metadata file e.g. default is relative path to package root.

Type

str

zenodo_model_uploader

An instance of the ZenodoModelUploader class

Type

str

github_model_uploader

An instance of the GithubModelUploader class.

Type

str

add_metadata_from_file(metadata_file_path) → dict[source]

Read and parse the metadata of a local model, identified by model_id, from a metadata file in json format.

Parameters

  • model_id (str) – The generative model’s unique id

  • metadata_file_path (str) – the path pointing to the metadata file

Returns

Returns a dict containing the contents of parsed metadata json file.

Return type

dict

add_metadata_from_input(model_weights_name: Optional[str] = None, model_weights_extension: Optional[str] = None, generate_method_name: Optional[str] = None, dependencies: list = [], fill_more_fields_interactively: bool = True, output_path: str = 'config')[source]

Create a metadata dict for a local model, identified by model_id, given the necessary minimum metadata contents.

Parameters

  • model_id (str) – The generative model’s unique id

  • model_weights_name (str) – the name of the checkpoint file containing the model’s weights

  • model_weights_extension (str) – the extension (e.g. .pt) of the checkpoint file containing the model’s weights

  • generate_method_name (str) – the name of the sample generation method inside the models __init__.py file

  • dependencies (list) – the list of dependencies that need to be installed via pip to run the model.

  • fill_more_fields_interactively (bool) – flag indicating whether a user will be interactively asked via command line for further input to fill out missing metadata content

  • output_path (str) – the path where the created metadata json file will be stored.

Returns

Returns a dict containing the contents of the metadata json file.

Return type

dict

is_value_for_key_already_set(key: str, metadata: dict, nested_key) → bool[source]

Check if the value of a key in a metadata dictionary is already set and e.g. not an empty string, dict or list.

Parameters

  • key (str) – The key in the currently traversed part of the model’s metadata dictionary

  • metadata (dict) – The currently traversed part of the model’s metadata dictionary

  • nested_key (str) – the nested_key indicates which subpart of the model’s metadata we are currently traversing

Returns

Flag indicating whether a value exists for the key in the dict

Return type

bool

load_metadata_template() → dict[source]

Loads and parses (json to dict) a default medigan metadata template.

Returns

Returns the metadata template as dict

Return type

dict

push_to_github(access_token: str, package_link: Optional[str] = None, creator_name: str = '', creator_affiliation: str = '', model_description: str = '')[source]

Upload the model’s metadata inside a github issue to the medigan github repository.

To add your model to medigan, your metadata will be reviewed on Github and added to medigan’s official model metadata

The medigan repository issues page: https://github.com/RichardObi/medigan/issues

Get your Github access token here: https://github.com/settings/tokens

Parameters

  • access_token (str) – a personal access token linked to your github user account, used as means of authentication

  • package_link – a package link

  • creator_name (str) – the creator name that will appear on the corresponding github issue

  • creator_affiliation (str) – the creator affiliation that will appear on the corresponding github issue

  • model_description (list) – the model_description that will appear on the corresponding github issue

Returns

Returns the url pointing to the corresponding issue on github

Return type

str

push_to_zenodo(access_token: str, creator_name: str, creator_affiliation: str, model_description: str = '')[source]

Upload the model files as zip archive to a public Zenodo repository where the model will be persistently stored.

Get your Zenodo access token here: https://zenodo.org/account/settings/applications/tokens/new/ (Enable scopes deposit:actions and deposit:write)

Parameters

  • access_token (str) – a personal access token in Zenodo linked to a user account for authentication

  • creator_name (str) – the creator name that will appear on the corresponding Zenodo model upload homepage

  • creator_affiliation (str) – the creator affiliation that will appear on the corresponding Zenodo model upload homepage

  • model_description (list) – the model_description that will appear on the corresponding Zenodo model upload homepage

Returns

Returns the url pointing to the corresponding Zenodo model upload homepage

Return type

str

validate_and_update_model_weights_path() → dict[source]

Check if the model files can be found in the package_path or based on the path_to_metadata.

Ideally, the user provided package_path and the path_to_metadata should both point to the same model package containing weights, config, license, etc. Here we check both of these paths to find the model weights.

Returns

Returns the metadata after updating the path to the model’s checkpoint’s weights

Return type

dict

validate_init_py_path(init_py_path) → bool[source]

Asserts whether the init_py_path exists and points to a valid __init__.py correct file.

Parameters

init_py_path (str) – The path to the local model’s __init__.py file needed for importing and running this model.

validate_local_model_import()[source]

Check if the model package in the package_path can be imported as python library using importlib.

validate_model_id(model_id: str, max_chars: int = 30, min_chars: int = 13) → bool[source]

Asserts if the model_id is in the correct format and has a valid length

Parameters

  • model_id (str) – The generative model’s unique id

  • max_chars (int) – the maximum of chars allowed in the model_id

  • min_chars (int) – the minimum of chars allowed in the model_id

Returns

Returns flag indicating whether the model_id is correctly formatted.

Return type

bool

medigan.contribute_model.zenodo_model_uploader module

Zenodo Model uploader class that uploads models to medigan associated data storage on Zenodo.

class medigan.contribute_model.zenodo_model_uploader.ZenodoModelUploader(model_id, access_token)[source]

Bases: medigan.contribute_model.base_model_uploader.BaseModelUploader

ZenodoModelUploader class: Uploads a user’s model via API to Zenodo, here it is permanently stored with DOI.

Parameters

  • model_id (str) – The generative model’s unique id

  • access_token (str) – a personal access token in Zenodo linked to a user account for authentication

model_id

The generative model’s unique id

Type

str

access_token

a personal access token in Zenodo linked to a user account for authentication

Type

str

create_upload_description(metadata: dict, model_description: str = '') → str[source]

Create a string containing the textual description that will accompany the upload model files.

The string contains tags and a text retrieved from the description subsection of the model metadata.

Parameters

  • metadata (dict) – The model’s corresponding metadata

  • model_description (str) – the model_description that will appear on the corresponding Zenodo model upload homepage

Returns

Returns the textual description of the model upload

Return type

str

create_upload_json_data(creator_name: str, creator_affiliation: str, description: str = '') → dict[source]

Create some descriptive data in dict format to be uploaded and stored alongside the model files.

Parameters

  • creator_name (str) – the creator name that will appear on the corresponding Zenodo model upload homepage

  • creator_affiliation (str) – the creator affiliation that will appear on the corresponding Zenodo model upload homepage

  • description (str) – the model_description that will appear on the corresponding Zenodo model upload homepage

Returns

Returns the descriptive data in dictionary structure describing the model upload.

Return type

dict

empty_upload() → dict[source]

Upload an empty placeholder entry to Zenodo as is required to retrieve a deposition_id and bucket_url.

deposition_id and bucket_url aare needed for file upload and publishing in the subsequent upload steps.

Returns

Returns the response retrieved via the Zenodo API

Return type

dict

locate_or_create_model_zip_file(package_path: str, package_name: str) -> (<class 'str'>, <class 'str'>)[source]

If not possible to locate, create a zipped python package of the model.

Parameters

  • package_path (str) – Path as string to the generative model’s python package containing an __init__.py file

  • package_name (str) – Name of the model’s python package i.e. the name of the model’s zip file and unzipped package folder

Returns

Returns a tuple containing two strings: The filename and the file_path of and to the zipped python package

Return type

tuple

publish(deposition_id: str) → dict[source]

Publish a zenodo upload.

This makes the upload official, as it will then be publicly accessible and persistently stored on Zenodo with associated DOI.

Parameters

deposition_id (str) – The deposition id assigned by Zenodo to the uploaded model file

Returns

Returns the response retrieved via the Zenodo API

Return type

dict

push(metadata: dict, package_path: str, package_name: str, creator_name: str, creator_affiliation: str, model_description: str = '')[source]

Upload the model files as zip archive to a public Zenodo repository where the model will be persistently stored.

Get your Zenodo access token here: https://zenodo.org/account/settings/applications/tokens/new/ (Enable scopes deposit:actions and deposit:write)

Parameters

  • metadata (dict) – The model’s corresponding metadata

  • package_path (dict) – The path to the packaged model files

  • package_name (dict) – The name of the packaged model files

  • creator_name (str) – the creator name that will appear on the corresponding Zenodo model upload homepage

  • creator_affiliation (str) – the creator affiliation that will appear on the corresponding Zenodo model upload homepage

  • model_description (list) – the model_description that will appear on the corresponding Zenodo model upload homepage

Returns

Returns the url pointing to the corresponding Zenodo model upload homepage

Return type

str

upload(file_path: str, filename: str, bucket_url: str) → dict[source]

Upload a file to Zenodo entry of the uploaded model files.

Parameters

  • file_path (str) – The path of the file that is uploaded to Zenodo

  • filename (str) – The name of the file that is uploaded to Zenodo

  • bucket_url (str) – The bucket url used in the PUT request to upload the data file.

Returns

Returns the response retrieved via the Zenodo API

Return type

dict

upload_descriptive_data(deposition_id: str, data: dict) → dict[source]

Upload textual descriptive data to be associated and added to the Zenodo entry of the uploaded model files.

Parameters

  • deposition_id (str) – The deposition id assigned by Zenodo to the uploaded model file

  • data (dict) – The descriptive information that will to be uploaded to Zenodo and associated with the desposition_id

Returns

Returns the response retrieved via the Zenodo API

Return type

dict

Module contents