[PTDT-2863]: Feature schema attributes

Author: Tim-Kerr

feature-schema-attributes-testing.py

@@ -0,0 +1,106 @@
+import os
+from libs.labelbox.src.labelbox import Client
+from libs.labelbox.src.labelbox.schema.ontology import OntologyBuilder, Tool
+from libs.labelbox.src.labelbox.schema.tool_building.classification import Classification, Option
+from libs.labelbox.src.labelbox.schema.tool_building.types import FeatureSchemaAttribute
+from libs.labelbox.src.labelbox.schema.media_type import MediaType
+from libs.labelbox.src.labelbox.schema.ontology_kind import OntologyKind
+import json
+
+
+# client = Client(
+#     api_key=os.environ.get('STAGE_API_KEY'),
+#     endpoint="https://app.lb-stage.xyz/api/_gql/graphql",
+#     rest_endpoint="https://app.lb-stage.xyz/api/api/v1")
+
+client = Client(
+    api_key=os.environ.get('LOCALHOST_API_KEY'),
+    endpoint="http://localhost:8080/graphql",
+    rest_endpoint="http://localhost:3000/api/api/v1")
+
+builder = OntologyBuilder(
+
+  tools=[
+    Tool(
+      name="Auto OCR",
+      tool=Tool.Type.BBOX,
+      attributes=[
+        FeatureSchemaAttribute(
+          attributeName="auto-ocr",
+          attributeValue="true"
+        )
+      ],
+      classifications=[
+        Classification(
+          name="Auto ocr text class value",
+          instructions="This is an auto OCR text value classification",
+          class_type=Classification.Type.TEXT,
+          scope=Classification.Scope.GLOBAL,
+          attributes=[
+            FeatureSchemaAttribute(
+              attributeName="auto-ocr-text-value",
+              attributeValue="true"
+            )
+          ]
+        )
+      ]
+    )
+  ]
+)
+
+# client.create_ontology("Auto OCR ontology", builder.asdict(), media_type=MediaType.Document)
+
+builder = OntologyBuilder(
+  classifications=[
+    Classification(
+      name="prompt message scope text classification",
+      instructions="This is a prompt message scoped text classification",
+      class_type=Classification.Type.TEXT,
+      scope=Classification.Scope.INDEX,
+      attributes=[
+        FeatureSchemaAttribute(
+          attributeName="prompt-message-scope",
+          attributeValue="true"
+        )
+      ]
+    )
+  ]
+)
+
+# client.create_ontology('MMC Ontology with prompt message scope class', builder.asdict(), media_type=MediaType.Conversational, ontology_kind=OntologyKind.ModelEvaluation)
+
+builder = OntologyBuilder(
+  classifications=[
+    Classification(
+      name="Requires connection checklist classification",
+      instructions="This is a requires connection checklist classification",
+      class_type=Classification.Type.CHECKLIST,
+      scope=Classification.Scope.GLOBAL,
+      attributes=[
+        FeatureSchemaAttribute(
+          attributeName="required-connection",
+          attributeValue="true"
+        )
+      ],
+      options=[
+        Option(
+          value='First option'
+        ),
+        Option(
+          value='Second option'
+        )
+      ]
+    )
+  ]
+);
+
+# client.create_ontology('Image ontology with requires connection classes', builder.asdict(), media_type=MediaType.Image)
+
+
+feature_schema = client.upsert_feature_schema(Tool(name='Auto OCR from upsert feature schema', tool=Tool.Type.BBOX, attributes=[FeatureSchemaAttribute(attributeName='auto-ocr', attributeValue='true')]).asdict())
+fetched_feature_schema = client.get_feature_schema(feature_schema.uid)
+feature_schemas_with_name = client.get_feature_schemas('Auto OCR')
+
+# Iterate over the feature schemas
+for schema in feature_schemas_with_name:
+    print(schema)

libs/labelbox/src/labelbox/client.py

@@ -1073,6 +1073,7 @@ def rootSchemaPayloadToFeatureSchema(client, payload):
             ["rootSchemaNodes", "nextCursor"],
         )
 
+# TODO tkerr:
     def create_ontology_from_feature_schemas(
         self,
         name,
@@ -1242,6 +1243,9 @@ def upsert_feature_schema(self, feature_schema: Dict) -> FeatureSchema:
             endpoint, json={"normalized": json.dumps(feature_schema)}
         )
 
+        print('json.dumps(feature_schema)')
+        print(json.dumps(feature_schema))
+
         if response.status_code == requests.codes.ok:
             return self.get_feature_schema(response.json()["schemaId"])
         else:
@@ -1328,6 +1332,7 @@ def get_unused_feature_schemas(self, after: str = None) -> List[str]:
                 + str(response.json()["message"])
             )
 
+# TODO tkerr: 
     def create_ontology(
         self,
         name,

libs/labelbox/src/labelbox/schema/ontology.py

@@ -25,6 +25,8 @@
 from labelbox.schema.tool_building.tool_type_mapping import (
     map_tool_type_to_tool_cls,
 )
+from labelbox.schema.tool_building.types import FeatureSchemaAttributes
+
 
 
 class DeleteFeatureFromOntologyResult:
@@ -43,7 +45,6 @@ class FeatureSchema(DbObject):
     color = Field.String("name")
     normalized = Field.Json("normalized")
 
-
 @dataclass
 class Tool:
     """
@@ -73,6 +74,7 @@ class Tool:
         classifications: (list)
         schema_id: (str)
         feature_schema_id: (str)
+        attributes: (list)
     """
 
     class Type(Enum):
@@ -95,7 +97,7 @@ class Type(Enum):
     classifications: List[Classification] = field(default_factory=list)
     schema_id: Optional[str] = None
     feature_schema_id: Optional[str] = None
-
+    attributes: Optional[FeatureSchemaAttributes] = None
     @classmethod
     def from_dict(cls, dictionary: Dict[str, Any]) -> Dict[str, Any]:
         return cls(
@@ -109,6 +111,7 @@ def from_dict(cls, dictionary: Dict[str, Any]) -> Dict[str, Any]:
                 for c in dictionary["classifications"]
             ],
             color=dictionary["color"],
+            attributes=dictionary.get("attributes", None),
         )
 
     def asdict(self) -> Dict[str, Any]:
@@ -122,6 +125,7 @@ def asdict(self) -> Dict[str, Any]:
             ],
             "schemaNodeId": self.schema_id,
             "featureSchemaId": self.feature_schema_id,
+            "attributes": self.attributes,
         }
 
     def add_classification(self, classification: Classification) -> None:

libs/labelbox/src/labelbox/schema/tool_building/classification.py

@@ -5,9 +5,10 @@
 
 from lbox.exceptions import InconsistentOntologyException
 
-from labelbox.schema.tool_building.types import FeatureSchemaId
+from labelbox.schema.tool_building.types import FeatureSchemaId, FeatureSchemaAttributes
 
 
+# TODO tkerr: Update all these tools & classifications to use attributes
 @dataclass
 class Classification:
     """
@@ -42,6 +43,7 @@ class Classification:
         schema_id: (str)
         feature_schema_id: (str)
         scope: (str)
+        attributes: (list)
     """
 
     class Type(Enum):
@@ -70,6 +72,7 @@ class UIMode(Enum):
     ui_mode: Optional[UIMode] = (
         None  # How this classification should be answered (e.g. hotkeys / autocomplete, etc)
     )
+    attributes: Optional[FeatureSchemaAttributes] = None
 
     def __post_init__(self):
         if self.name is None:
@@ -88,9 +91,13 @@ def __post_init__(self):
         else:
             if self.instructions is None:
                 self.instructions = self.name
+        if self.attributes is not None:
+            warnings.warn('Attributes are an experimental feature and may change in the future.')
 
     @classmethod
     def from_dict(cls, dictionary: Dict[str, Any]) -> "Classification":
+        print('attributes:')
+        print(dictionary.get("attributes", None))
         return cls(
             class_type=Classification.Type(dictionary["type"]),
             name=dictionary["name"],
@@ -103,6 +110,7 @@ def from_dict(cls, dictionary: Dict[str, Any]) -> "Classification":
             schema_id=dictionary.get("schemaNodeId", None),
             feature_schema_id=dictionary.get("featureSchemaId", None),
             scope=cls.Scope(dictionary.get("scope", cls.Scope.GLOBAL)),
+            attributes=FeatureSchemaAttributes(dictionary.get("attributes", None)),
         )
 
     def asdict(self, is_subclass: bool = False) -> Dict[str, Any]:
@@ -118,6 +126,7 @@ def asdict(self, is_subclass: bool = False) -> Dict[str, Any]:
             "options": [o.asdict() for o in self.options],
             "schemaNodeId": self.schema_id,
             "featureSchemaId": self.feature_schema_id,
+            "attributes": self.attributes if self.attributes is not None else None,
         }
         if (
             self.class_type == self.Type.RADIO

libs/labelbox/src/labelbox/schema/tool_building/types.py

@@ -1,6 +1,14 @@
-from typing import Annotated
+from typing import Annotated, List
+from pydantic import Field, BaseModel
+from typing import TypedDict
 
-from pydantic import Field
+
+class FeatureSchemaAttribute(TypedDict):
+    attributeName: str
+    attributeValue: str
+
+FeatureSchemaAttriubte = Annotated[FeatureSchemaAttribute, Field()]
 
 FeatureSchemaId = Annotated[str, Field(min_length=25, max_length=25)]
 SchemaId = Annotated[str, Field(min_length=25, max_length=25)]
+FeatureSchemaAttributes = Annotated[List[FeatureSchemaAttribute], Field(default_factory=list)]

libs/labelbox/tests/integration/test_ontology.py

@@ -225,6 +225,7 @@ def feature_schema_cat_normalized(name_for_read):
         "name": name_for_read,
         "color": "black",
         "classifications": [],
+        "attributes": [] # TODO tkerr: Finish
     }
 
 

requirements-dev.lock

@@ -6,6 +6,7 @@
 #   features: []
 #   all-features: true
 #   with-sources: false
+#   generate-hashes: false
 
 -e file:libs/labelbox
 -e file:libs/lbox-clients
@@ -133,7 +134,7 @@ nbconvert==7.16.4
 nbformat==5.10.4
     # via nbclient
     # via nbconvert
-numpy==1.24.4
+numpy==2.0.2
     # via labelbox
     # via opencv-python-headless
     # via pandas
@@ -147,7 +148,7 @@ packaging==24.1
     # via pytest-cases
     # via pytest-rerunfailures
     # via sphinx
-pandas==2.0.3
+pandas==2.2.3
 pandocfilters==1.5.1
     # via nbconvert
 parso==0.8.4

requirements.lock

@@ -6,6 +6,7 @@
 #   features: []
 #   all-features: true
 #   with-sources: false
+#   generate-hashes: false
 
 -e file:libs/labelbox
 -e file:libs/lbox-clients
@@ -54,7 +55,7 @@ mypy==1.10.1
     # via labelbox
 mypy-extensions==1.0.0
     # via mypy
-numpy==1.24.4
+numpy==2.0.2
     # via labelbox
     # via opencv-python-headless
     # via shapely