medigan.select_model package
Submodules
medigan.select_model.matched_entry module
MatchedEntry class that represents one match of a key value pair of a model’s config against a search query.
- class medigan.select_model.matched_entry.MatchedEntry(key: str, value, matching_element: Optional[str] = None)[source]
Bases:
object
MatchedEntry class: One target key-value pair that matches with a model’s selection config.
- Parameters
key (str) – string that represents the matched key in model selection dict
value – represents the key’s matched value in the model selection dict
matching_element (str) – string that was used to match the search value
- key
string that represents the matched key in model selection dict
- Type
str
- value
represents the key’s matched value in the model selection dict
- matching_element
string that was used to match the search value
- Type
str
medigan.select_model.model_match_candidate module
ModelMatchCandidate class that holds data to evaluate if a generative model matches against a search query.
- class medigan.select_model.model_match_candidate.ModelMatchCandidate(model_id: str, target_values: list, target_values_operator: str = 'AND', is_case_sensitive: bool = False, are_keys_also_matched: bool = False, is_match: bool = False)[source]
Bases:
object
ModelMatchCandidate class: A prospectively matching model given the target values as model search params.
- Parameters
model_id (str) – The generative model’s unique id
target_values (list) – list of target values used to evaluate if a ModelMatchCandidate instance is a match
target_values_operator (str) – the operator indicating the relationship between values in the evaluation of ModelMatchCandidate instances. Should be either “AND”, “OR”, or “XOR”.
is_case_sensitive (bool) – flag indicating whether the matching of values (and) keys should be case-sensitive
are_keys_also_matched (bool) – flag indicating whether, apart from values, keys should also be matched
is_match (bool) – flag indicating whether the ModelMatchCandidate instance is a match
- model_id
The generative model’s unique id
- Type
str
- target_values
list of target values used to evaluate if a ModelMatchCandidate instance is a match
- Type
list
- target_values_operator
the operator indicating the relationship between values in the evaluation of ModelMatchCandidate instances. Should be either “AND”, “OR”, or “XOR”.
- Type
str
- is_case_sensitive
flag indicating whether the matching of values (and) keys should be case-sensitive
- Type
bool
- are_keys_also_matched
flag indicating whether, apart from values, keys should also be matched
- Type
bool
- matched_entries
contains iteratively added MatchedEntry class instances. Each of the MatchedEntry instances indicates a match between one of the user specified values in self.target_values and the selection config keys or values of the model of this ModelMatchCandidate.
- Type
list
- is_match
flag indicating whether the ModelMatchCandidate instance is a match
- Type
bool
- add_matched_entry(matched_entry: medigan.select_model.matched_entry.MatchedEntry) None [source]
Add a MatchedEntry instance to the matched_entries list.
- check_if_is_match() bool [source]
Evaluates whether the model represented by this instance is a match given search values and operator.
The matching element from each MatchEntry of this instance (‘self.matched_entries’) are retrieved. To be a match, this instance (‘self’) needs to fulfill the requirement of the operator, which can be of value ‘AND’, or ‘OR’, or ‘XOR’. For example, the default ‘AND’ requires that each search value (‘self.target_values’) has at least one corresponding MatchEntry, while in ‘OR’ only one of the search values needs to have been matched by a corresponding MatchedEntry.
- Returns
flag that, only if True, indicates that this instance is a match given the search values and operator.
- Return type
bool
medigan.select_model.model_selector module
Model selection class that describes, finds, compares, and ranks generative models.
- class medigan.select_model.model_selector.ModelSelector(config_manager: Optional[medigan.config_manager.ConfigManager] = None)[source]
Bases:
object
ModelSelector class: Given a config dict, gets, searches, and ranks matching models.
- Parameters
config_manager (ConfigManager) – Provides the config dictionary, based on which models are selected and compared.
- config_manager
Provides the config dictionary, based on which models are selected and compared.
- Type
- model_selection_dicts
Contains a dictionary for each model id that consists of the model_id and the selection config of that model
- Type
list
- find_matching_models_by_values(values: list, target_values_operator: str = 'AND', are_keys_also_matched: bool = False, is_case_sensitive: bool = False) list [source]
Search for values (and keys) in model configs and return a list of each matching ModelMatchCandidate.
Uses self.recursive_search_for_values to recursively populate each ModelMatchCandidate with MatchedEntry instances. After populating, each ModelMatchCandidate is evaluated based on the provided target_values_operator and values list using ModelMatchCandidate.check_if_is_match.
- Parameters
values (list) – list of values used to search and find models corresponding to these values
target_values_operator (str) – the operator indicating the relationship between values in the evaluation of model search results. Should be either “AND”, “OR”, or “XOR”.
are_keys_also_matched (bool) – flag indicating whether, apart from values, the keys in the model config should also be searchable
is_case_sensitive (bool) – flag indicating whether the search for values (and) keys in the model config should be case-sensitive.
- Returns
a list of ModelMatchCandidate class instances each of which was successfully matched against the search values.
- Return type
list
- find_models_and_rank(values: list, target_values_operator: str = 'AND', are_keys_also_matched: bool = False, is_case_sensitive: bool = False, metric: str = 'SSIM', order: str = 'asc') list [source]
Search for values (and keys) in model configs, rank results and return sorted list of model dicts.
- Parameters
values (list) – list of values used to search and find models corresponding to these values
target_values_operator (str) – the operator indicating the relationship between values in the evaluation of model search results. Should be either “AND”, “OR”, or “XOR”.
are_keys_also_matched (bool) – flag indicating whether, apart from values, the keys in the model config should also be searchable
is_case_sensitive (bool) – flag indicating whether the search for values (and) keys in the model config should be case-sensitive.
metric (str) – The key in the selection dict that corresponds to the metric of interest
order (str) – the sorting order of the ranked results. Should be either “asc” (ascending) or “desc” (descending)
- Returns
a list of the searched and matched model dictionaries containing metric and model_id, sorted by metric.
- Return type
list
- get_models_by_key_value_pair(key1: str, value1: str, is_case_sensitive: bool = False) list [source]
Get and return a list of model_id dicts that contain the specified key value pair in their selection config.
The key param can contain ‘.’ (dot) separations to allow for retrieval of nested config keys such as ‘execution.generator.name’. If key1 points to a list, any value in the list that matches value1` is accepted.
- Parameters
key1 (str`) – The key in the selection dict
value1 (str) – The value in the selection dict that corresponds to key1
is_case_sensitive (bool) – flag to evaluate keys and values with case sensitivity if set to True
- Returns
a list of the dictionaries each containing a model’s model_id and the found key-value pair in the models config
- Return type
list
- get_selection_criteria_by_id(model_id: str, is_model_id_removed: bool = True) dict [source]
Get and return the selection config dict for a specific model_id.
- Parameters
model_id (str) – The generative model’s unique id
is_model_id_removed (bool) – flag to to remove the model_id from first level of each dictionary.
- Returns
a dictionary corresponding to the selection config of a model
- Return type
dict
- get_selection_criteria_by_ids(model_ids: Optional[list] = None, are_model_ids_removed: bool = True) list [source]
Get and return a list of selection config dicts for each of the specified model_ids.
- Parameters
model_ids (list) – A list of generative models’ unique ids
are_model_ids_removed (bool) – flag to remove the model_ids from first level of dictionary.
- Returns
a list of dictionaries each corresponding to the selection config of a model
- Return type
list
- get_selection_keys(model_id: Optional[str] = None) list [source]
Get and return all first level keys from the selection config dict for a specific model_id.
- Parameters
model_id (str) – The generative model’s unique id
- Returns
a list containing the keys as strings of the selection config of the model_id.
- Return type
list
- get_selection_values_for_key(key: str, model_id: Optional[str] = None) list [source]
Get and return the value of a specified key of the selection dict in the config for a specific model_id.
The key param can contain ‘.’ (dot) separations to allow for retrieval of nested config keys such as ‘execution.generator.name’
- Parameters
key (str) – The key in the selection dict
model_id (str) – The generative model’s unique id
- Returns
a list of the values that correspond to the key in the selection config of the model_id.
- Return type
list
- rank_models_by_performance(model_ids: Optional[list] = None, metric: str = 'SSIM', order: str = 'asc') list [source]
Rank model based on a provided metric and return sorted list of model dicts.
The metric param can contain ‘.’ (dot) separations to allow for retrieval via nested metric config keys such as ‘downstream_task.CLF.accuracy’. If the value found for the metric key is of type list, then the largest value in the list is used for ranking if order is descending, while the smallest value is used if order is ascending.
- Parameters
model_ids (list) – only evaluate the model_ids in this list. If none, evaluate all available model_ids
metric (str) – The key in the selection dict that corresponds to the metric of interest
order (str) – the sorting order of the ranked results. Should be either “asc” (ascending) or “desc” (descending)
- Returns
a list of model dictionaries containing metric and model_id, sorted by metric.
- Return type
list
- recursive_search_for_values(search_dict: dict, model_match_candidate: medigan.select_model.model_match_candidate.ModelMatchCandidate)[source]
Do a recursive search to match values in the search_dict with values (and keys) in a model’s config.
The provided ModelMatchCandidate instance is recursively populated with MatchedEntry instances. A MatchedEntry instance contains a key-value pair found in the model’s config that matches with one search term of interest.
The search terms of interest are stored in ModelMatchCandidate.target_values. The model’s selection config is provided in the ‘search_dict’.
To traverse the search_dict, the value for each key in the search_dict is retrieved.
If that value is of type dictionary, the function calls itself with that nested dictionary as new search_dict.
If that value is of type list, each value in the list is compared with each search term of interest in ModelMatchCandidate.target_values.
If the value of the search_dict is of another type (i.e. str), it is compared with each search term of interest in ModelMatchCandidate.target_values.
- Parameters
search_dict (dict) – contains keys and values from a model’s config that are matched against a set of search values.
model_match_candidate (ModelMatchCandidate) – a class instance representing a model to be prepared for evaluation (populated with MatchedEntry objects), as to whether it is a match given its search values (self.target_values).
- Returns
a ModelMatchCandidate class instance that has been populated with MatchedEntry class instances.
- Return type
list