Sub-classes plus generatedBy fix

fairscape_models/__init__.py

@@ -1,8 +1,11 @@
+from fairscape_models.activity import Activity
+from fairscape_models.digital_object import DigitalObject
 from fairscape_models.annotation import Annotation
 from fairscape_models.biochem_entity import BioChemEntity
 from fairscape_models.computation import Computation
 from fairscape_models.dataset import Dataset
 from fairscape_models.software import Software
+from fairscape_models.mlmodel import MLModel
 from fairscape_models.fairscape_base import IdentifierValue, IdentifierPropertyValue
 from fairscape_models.medical_condition import MedicalCondition
 from fairscape_models.schema import Schema

fairscape_models/activity.py

@@ -0,0 +1,16 @@
+from pydantic import BaseModel, Field, ConfigDict
+from typing import Optional, List
+
+from fairscape_models.fairscape_base import IdentifierValue
+
+class Activity(BaseModel):
+    """Base class for Activity types (Computation, Annotation, Experiment)"""
+    guid: str = Field(alias="@id")
+    name: str
+    metadataType: Optional[str] = Field(default=None, alias="@type")
+    description: str = Field(min_length=10)
+    associatedPublication: Optional[str] = Field(default=None)
+    generated: Optional[List[IdentifierValue]] = Field(default=[])
+    isPartOf: Optional[List[IdentifierValue]] = Field(default=[])
+
+    model_config = ConfigDict(extra="allow")

fairscape_models/annotation.py

@@ -1,19 +1,12 @@
-from pydantic import BaseModel, Field, ConfigDict
+from pydantic import Field, ConfigDict
 from typing import Optional, List
 
 from fairscape_models.fairscape_base import IdentifierValue, ANNOTATION_TYPE
+from fairscape_models.activity import Activity
 
-class Annotation(BaseModel):
-    guid: str = Field(alias="@id")
-    name: str
+class Annotation(Activity):
     metadataType: Optional[str] = Field(default="https://w3id.org/EVI#Annotation", alias="@type")
     additionalType: Optional[str] = Field(default=ANNOTATION_TYPE)
     createdBy: str
-    description: str = Field(min_length=10)
     dateCreated: str
-    associatedPublication: Optional[str] = Field(default=None)
     usedDataset: Optional[List[IdentifierValue]] = Field(default=[])
-    generated: Optional[List[IdentifierValue]] = Field(default=[])
-    isPartOf: Optional[List[IdentifierValue]] = Field(default=[])
-
-    model_config = ConfigDict(extra="allow")

fairscape_models/computation.py

@@ -1,23 +1,16 @@
-from pydantic import BaseModel, Field, ConfigDict
+from pydantic import Field, ConfigDict
 from typing import Optional, List, Union
 
 from fairscape_models.fairscape_base import IdentifierValue, COMPUTATION_TYPE
+from fairscape_models.activity import Activity
 
-class Computation(BaseModel):
-    guid: str = Field(alias="@id")
-    name: str
-    metadataType: Optional[str] = Field(default="https://w3id.org/EVI#Computation",alias="@type")
+class Computation(Activity):
+    metadataType: Optional[str] = Field(default="https://w3id.org/EVI#Computation", alias="@type")
     additionalType: Optional[str] = Field(default=COMPUTATION_TYPE)
     runBy: str
-    description: str = Field(min_length=10)
     dateCreated: str
-    associatedPublication: Optional[str] = Field(default=None)
     additionalDocumentation: Optional[str] = Field(default=None)
     command: Optional[Union[List[str], str]] = Field(default=None)
     usedSoftware: Optional[List[IdentifierValue]] = Field(default=[])
     usedMLModel: Optional[List[IdentifierValue]] = Field(default=[])
     usedDataset: Optional[List[IdentifierValue]] = Field(default=[])
-    generated: Optional[List[IdentifierValue]] = Field(default=[])
-    isPartOf: Optional[List[IdentifierValue]] = Field(default=[])
-
-    model_config = ConfigDict(extra="allow")

fairscape_models/dataset.py

@@ -1,30 +1,19 @@
-from pydantic import BaseModel, Field, ConfigDict, AliasChoices, field_validator
+from pydantic import Field, ConfigDict, AliasChoices
 from typing import Optional, List, Union
 
 from fairscape_models.fairscape_base import IdentifierValue, DATASET_TYPE
+from fairscape_models.digital_object import DigitalObject
 
-class Dataset(BaseModel):
-    guid: str = Field(alias="@id")
-    name: str
-    metadataType: Optional[str] = Field(default="https://w3id.org/EVI#Dataset",alias="@type")
+class Dataset(DigitalObject):
+    metadataType: Optional[str] = Field(default="https://w3id.org/EVI#Dataset", alias="@type")
     additionalType: Optional[str] = Field(default=DATASET_TYPE)
-    author: Union[str, List[str]]
     datePublished: str = Field(...)
-    version: str = Field(default="0.1.0")
-    description: str = Field(min_length=10)
     keywords: List[str] = Field(...)
-    associatedPublication: Optional[Union[str,List[str]]] = Field(default=None)
-    additionalDocumentation: Optional[str] = Field(default=None)
     fileFormat: str = Field(alias="format")
     dataSchema: Optional[IdentifierValue] = Field(
-        validation_alias=AliasChoices('evi:Schema', 'EVI:Schema', 'schema','evi:schema'),
+        validation_alias=AliasChoices('evi:Schema', 'EVI:Schema', 'schema', 'evi:schema'),
         serialization_alias='evi:Schema',
         default=None
     )
     generatedBy: Optional[Union[IdentifierValue, List[IdentifierValue]]] = Field(default=[])
-    derivedFrom: Optional[List[IdentifierValue]] = Field(default=[])
-    usedByComputation: Optional[List[IdentifierValue]] = Field(default=[])
-    contentUrl: Optional[Union[str, List[str]]] = Field(default=None)
-    isPartOf: Optional[List[IdentifierValue]] = Field(default=[])
-
-    model_config = ConfigDict(extra="allow")
+    derivedFrom: Optional[List[IdentifierValue]] = Field(default=[])

fairscape_models/digital_object.py

@@ -0,0 +1,20 @@
+from pydantic import BaseModel, Field, ConfigDict
+from typing import Optional, List, Union
+
+from fairscape_models.fairscape_base import IdentifierValue
+
+class DigitalObject(BaseModel):
+    """Base class for DigitalObject types (Dataset, Software, MLModel)"""
+    guid: str = Field(alias="@id")
+    name: str
+    metadataType: Optional[str] = Field(default=None, alias="@type")
+    author: Union[str, List[str]]
+    description: str = Field(min_length=10)
+    version: str = Field(default="0.1.0")
+    associatedPublication: Optional[Union[str, List[str]]] = Field(default=None)
+    additionalDocumentation: Optional[str] = Field(default=None)
+    contentUrl: Optional[Union[str, List[str]]] = Field(default=None)
+    isPartOf: Optional[List[IdentifierValue]] = Field(default=[])
+    usedByComputation: Optional[List[IdentifierValue]] = Field(default=[])
+
+    model_config = ConfigDict(extra="allow")

fairscape_models/experiment.py

@@ -1,21 +1,15 @@
-from pydantic import BaseModel, Field, ConfigDict
-from typing import Optional, List, Union
+from pydantic import Field, ConfigDict
+from typing import Optional, List
 from fairscape_models.fairscape_base import IdentifierValue
+from fairscape_models.activity import Activity
 
-class Experiment(BaseModel):
-    guid: str = Field(alias="@id")
-    name: str
+class Experiment(Activity):
     metadataType: Optional[str] = Field(default="https://w3id.org/EVI#Experiment", alias="@type")
-    experimentType: str  
+    experimentType: str
     runBy: str
-    description: str = Field(min_length=10)
-    datePerformed: str 
-    associatedPublication: Optional[str] = Field(default=None)
-    protocol: Optional[str] = Field(default=None) 
-    usedInstrument: Optional[List[IdentifierValue]] = Field(default=[]) 
-    usedSample: Optional[List[IdentifierValue]] = Field(default=[]) 
+    datePerformed: str
+    protocol: Optional[str] = Field(default=None)
+    usedInstrument: Optional[List[IdentifierValue]] = Field(default=[])
+    usedSample: Optional[List[IdentifierValue]] = Field(default=[])
     usedTreatment: Optional[List[IdentifierValue]] = Field(default=[])
-    usedStain: Optional[List[IdentifierValue]] = Field(default=[])
-    generated: Optional[List[IdentifierValue]] = Field(default=[])
-    isPartOf: Optional[List[IdentifierValue]] = Field(default=[])
-    model_config = ConfigDict(extra="allow")
+    usedStain: Optional[List[IdentifierValue]] = Field(default=[])

fairscape_models/fairscape_base.py

@@ -20,6 +20,7 @@
 DATASET_TYPE = "Dataset"
 DATASET_CONTAINER_TYPE = "DatasetContainer"
 SOFTWARE_TYPE = "Software"
+MLMODEL_TYPE = "MLModel"
 COMPUTATION_TYPE = "Computation"
 ANNOTATION_TYPE = "Annotation"
 ROCRATE_TYPE = "ROCrate"
@@ -58,6 +59,7 @@
 class ClassType(str, Enum):
     DATASET = 'Dataset'
     SOFTWARE = 'Software'
+    MLMODEL = 'MLModel'
     COMPUTATION = 'Computation'
     ANNOTATION = 'Annotation'
     SCHEMA = 'Schema'

fairscape_models/mlmodel.py

@@ -0,0 +1,14 @@
+from pydantic import Field, ConfigDict
+from typing import Optional, List, Union
+
+from fairscape_models.fairscape_base import IdentifierValue, MLMODEL_TYPE
+from fairscape_models.digital_object import DigitalObject
+
+class MLModel(DigitalObject):
+    metadataType: Optional[str] = Field(default="https://w3id.org/EVI#MLModel", alias="@type")
+    additionalType: Optional[str] = Field(default=MLMODEL_TYPE)
+    dateModified: Optional[str] = Field(default=None)
+    fileFormat: str = Field(alias="format")
+    modelTask: Optional[str] = Field(default=None)
+    modelArchitecture: Optional[str] = Field(default=None)
+    trainedOn: Optional[List[IdentifierValue]] = Field(default=[])

fairscape_models/rocrate.py

@@ -8,8 +8,11 @@
 from fairscape_models.biochem_entity import BioChemEntity
 from fairscape_models.medical_condition import MedicalCondition
 from fairscape_models.computation import Computation
+from fairscape_models.annotation import Annotation
+from fairscape_models.experiment import Experiment
 from fairscape_models.dataset import Dataset
 from fairscape_models.software import Software
+from fairscape_models.mlmodel import MLModel
 from fairscape_models.patient import Patient
 
 class GenericMetadataElem(BaseModel):
@@ -127,7 +130,10 @@ class ROCrateV1_2(BaseModel):
     metadataGraph: List[Union[
         Dataset,
         Software,
+        MLModel,
         Computation,
+        Annotation,
+        Experiment,
         ROCrateMetadataElem,
         ROCrateMetadataFileElem,
         Schema,
@@ -146,7 +152,10 @@ def validate_metadata_graph(cls, values: Dict[str, Any]) -> Dict[str, Any]:
         type_map = {
             "Dataset": Dataset,
             "Software": Software,
+            "MLModel": MLModel,
             "Computation": Computation,
+            "Annotation": Annotation,
+            "Experiment": Experiment,
             "CreativeWork": ROCrateMetadataFileElem,
             "Schema": Schema,
             "BioChemEntity": BioChemEntity,
@@ -206,8 +215,6 @@ def cleanIdentifierList(identifier_list):
             """Helper to clean a list of identifiers"""
             if identifier_list is None:
                 return
-            if not isinstance(identifier_list, list):
-                return
             for item in identifier_list:
                 if hasattr(item, 'guid') and isinstance(item.guid, str) and "ark:" in item.guid:
                     cleanGUID(item)
@@ -241,6 +248,11 @@ def cleanIdentifierUnion(identifier_union):
             if isinstance(elem, Software):
                 cleanIdentifierList(elem.usedByComputation)
 
+            if isinstance(elem, MLModel):
+                cleanIdentifierList(elem.usedByComputation)
+
+                cleanIdentifierList(elem.trainedOn)
+
             if isinstance(elem, Computation):
 
                 cleanIdentifierList(elem.usedDataset)
@@ -249,6 +261,26 @@ def cleanIdentifierUnion(identifier_union):
 
                 cleanIdentifierList(elem.usedSoftware)
 
+                cleanIdentifierList(elem.usedMLModel)
+
+            if isinstance(elem, Annotation):
+
+                cleanIdentifierList(elem.usedDataset)
+
+                cleanIdentifierList(elem.generated)
+
+            if isinstance(elem, Experiment):
+
+                cleanIdentifierList(elem.usedInstrument)
+
+                cleanIdentifierList(elem.usedSample)
+
+                cleanIdentifierList(elem.usedTreatment)
+
+                cleanIdentifierList(elem.usedStain)
+
+                cleanIdentifierList(elem.generated)
+
     def getCrateMetadata(self)-> ROCrateMetadataElem:
         """ Filter the Metadata Graph for the Metadata Element Describing the Toplevel ROCrate
 
@@ -316,7 +348,49 @@ def getComputations(self) -> List[Computation]:
         :rtype List[fairscape_mds.models.rocrate.Computation]
         """
         filterResults = list(filter(
-            lambda x: isinstance(x, Computation), 
+            lambda x: isinstance(x, Computation),
+            self.metadataGraph
+        ))
+
+        return filterResults
+
+    def getAnnotations(self) -> List[Annotation]:
+        """ Filter the Metadata Graph for Annotation Elements
+
+        :param self
+        :return: All Annotation metadata records within the ROCrate
+        :rtype List[fairscape_mds.models.rocrate.Annotation]
+        """
+        filterResults = list(filter(
+            lambda x: isinstance(x, Annotation),
+            self.metadataGraph
+        ))
+
+        return filterResults
+
+    def getExperiments(self) -> List[Experiment]:
+        """ Filter the Metadata Graph for Experiment Elements
+
+        :param self
+        :return: All Experiment metadata records within the ROCrate
+        :rtype List[fairscape_mds.models.rocrate.Experiment]
+        """
+        filterResults = list(filter(
+            lambda x: isinstance(x, Experiment),
+            self.metadataGraph
+        ))
+
+        return filterResults
+
+    def getMLModels(self) -> List[MLModel]:
+        """ Filter the Metadata Graph for MLModel Elements
+
+        :param self
+        :return: All MLModel metadata records within the ROCrate
+        :rtype List[fairscape_mds.models.rocrate.MLModel]
+        """
+        filterResults = list(filter(
+            lambda x: isinstance(x, MLModel),
             self.metadataGraph
         ))
 
@@ -354,13 +428,18 @@ def getMedicalConditions(self) -> List[MedicalCondition]:
 
 
     def getEVIElements(self) -> List[Union[
-        Computation, 
-        Dataset, 
-        Software, 
+        Computation,
+        Annotation,
+        Experiment,
+        Dataset,
+        Software,
+        MLModel,
         Schema,
         BioChemEntity,
         MedicalCondition
         ]]:
         """ Query the metadata graph for elements which require minting identifiers
         """
-        return self.getDatasets() + self.getSoftware() + self.getComputations() + self.getSchemas()
+        return (self.getDatasets() + self.getSoftware() + self.getMLModels() +
+                self.getComputations() + self.getAnnotations() + self.getExperiments() +
+                self.getSchemas())

fairscape_models/software.py

@@ -1,22 +1,11 @@
-from pydantic import BaseModel, Field, ConfigDict
-from typing import Optional, List, Union
+from pydantic import Field, ConfigDict
+from typing import Optional, List
 
 from fairscape_models.fairscape_base import IdentifierValue, SOFTWARE_TYPE
+from fairscape_models.digital_object import DigitalObject
 
-class Software(BaseModel): 
-    guid: str = Field(alias="@id")
-    name: str
+class Software(DigitalObject):
     metadataType: Optional[str] = Field(default="https://w3id.org/EVI#Software", alias="@type")
     additionalType: Optional[str] = Field(default=SOFTWARE_TYPE)
-    author: str = Field(min_length=4)
     dateModified: Optional[str]
-    version: str = Field(default="0.1.0")
-    description: str =  Field(min_length=10)
-    associatedPublication: Optional[str] = Field(default=None)
-    additionalDocumentation: Optional[str] = Field(default=None)
     fileFormat: str = Field(title="fileFormat", alias="format")
-    usedByComputation: Optional[List[IdentifierValue]] = Field(default=[])
-    contentUrl: Optional[str] = Field(default=None)
-    isPartOf: Optional[List[IdentifierValue]] = Field(default=[])
-
-    model_config = ConfigDict(extra="allow")

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "fairscape-models"
-version = "1.0.15"
+version = "1.0.19"
 description = "Fairscape pydantic models"
 readme = "README.md"
 authors = [