3.26.1

CHANGELOG.md

@@ -1,6 +1,22 @@
 # Changelog
 
-# Version 3.26.0 (2022-08-12)
+# Version 3.26.1 (2022-08-23)
+### Changed
+* `ModelRun.get_config()`
+    * Modifies get_config to return un-nested Model Run config
+### Added
+* `ModelRun.update_config()`
+    * Updates model run training metadata
+* `ModelRun.reset_config()`
+    * Resets model run training metadata
+* `ModelRun.get_config()`
+    * Fetches model run training metadata
+
+### Changed
+* `Model.create_model_run()`
+    * Add training metadata config as a model run creation param
+
+# Version 3.26.0 (2022-08-15)
 ## Added
 * `Batch.delete()` which will delete an existing `Batch`
 * `Batch.delete_labels()`  which will delete all `Label`’s created after a `Project`’s mode has been set to batch.
@@ -663,7 +679,3 @@ a `Label`. Default value is 0.0.
 
 ## Version 2.2 (2019-10-18)
 Changelog not maintained before version 2.2.
-
-### Changed
-* `Model.create_model_run()`
-    * Add training metadata config as a model run creation param

docs/source/conf.py

@@ -21,7 +21,7 @@
 copyright = '2021, Labelbox'
 author = 'Labelbox'
 
-release = '3.25.1'
+release = '3.26.1'
 
 # -- General configuration ---------------------------------------------------
 

labelbox/__init__.py

@@ -1,5 +1,5 @@
 name = "labelbox"
-__version__ = "3.26.0"
+__version__ = "3.26.1"
 
 from labelbox.client import Client
 from labelbox.schema.project import Project

labelbox/data/serialization/labelbox_v1/label.py

@@ -146,6 +146,7 @@ class LBV1Label(BaseModel):
     skipped: Optional[bool] = Extra('Skipped')
     media_type: Optional[str] = Extra('media_type')
     data_split: Optional[str] = Extra('Data Split')
+    global_key: Optional[str] = Extra('Global Key')
 
     def to_common(self) -> Label:
         if isinstance(self.label, list):

labelbox/data/serialization/labelbox_v1/objects.py

@@ -22,6 +22,8 @@ class LBV1ObjectBase(LBV1Feature):
     instanceURI: Optional[str] = None
     classifications: List[Union[LBV1Text, LBV1Radio, LBV1Dropdown,
                                 LBV1Checklist]] = []
+    page: Optional[int] = None
+    unit: Optional[str] = None
 
     def dict(self, *args, **kwargs) -> Dict[str, Any]:
         res = super().dict(*args, **kwargs)
@@ -262,7 +264,7 @@ def from_common(cls, text_entity: TextEntity,
 class LBV1Objects(BaseModel):
     objects: List[Union[LBV1Line, LBV1Point, LBV1Polygon, LBV1Rectangle,
                         LBV1TextEntity, LBV1Mask, LBV1TIPoint, LBV1TILine,
-                        LBV1TIPolygon, LBV1TIRectangle]]
+                        LBV1TIPolygon, LBV1TIRectangle,]]
 
     def to_common(self) -> List[ObjectAnnotation]:
         objects = [
@@ -285,6 +287,8 @@ def to_common(self) -> List[ObjectAnnotation]:
                                  'color': obj.color,
                                  'feature_id': obj.feature_id,
                                  'value': obj.value,
+                                 'page': obj.page,
+                                 'unit': obj.unit,
                              }) for obj in self.objects
         ]
         return objects

labelbox/schema/data_row.py

@@ -17,6 +17,7 @@ class DataRow(DbObject, Updateable, BulkDeletable):
 
     Attributes:
         external_id (str): User-generated file name or identifier
+        global_key (str): User-generated globally unique identifier
         row_data (str): Paths to local files are uploaded to Labelbox's server.
             Otherwise, it's treated as an external URL.
         updated_at (datetime)
@@ -33,6 +34,7 @@ class DataRow(DbObject, Updateable, BulkDeletable):
         attachments (Relationship) `ToMany` relationship with AssetAttachment
     """
     external_id = Field.String("external_id")
+    global_key = Field.String("global_key")
     row_data = Field.String("row_data")
     updated_at = Field.DateTime("updated_at")
     created_at = Field.DateTime("created_at")

labelbox/schema/model_run.py

@@ -42,7 +42,7 @@ class Status(Enum):
         FAILED = "FAILED"
 
     def upsert_labels(self, label_ids, timeout_seconds=60):
-        """ Adds data rows and labels to a model run
+        """ Adds data rows and labels to a Model Run
         Args:
             label_ids (list): label ids to insert
             timeout_seconds (float): Max waiting time, in seconds.
@@ -75,7 +75,7 @@ def upsert_labels(self, label_ids, timeout_seconds=60):
                                      timeout_seconds=timeout_seconds)
 
     def upsert_data_rows(self, data_row_ids, timeout_seconds=60):
-        """ Adds data rows to a model run without any associated labels
+        """ Adds data rows to a Model Run without any associated labels
         Args:
             data_row_ids (list): data row ids to add to mea
             timeout_seconds (float): Max waiting time, in seconds.
@@ -167,7 +167,7 @@ def model_run_data_rows(self):
             ['annotationGroups', 'pageInfo', 'endCursor'])
 
     def delete(self):
-        """ Deletes specified model run.
+        """ Deletes specified Model Run.
 
         Returns:
             Query execution success.
@@ -178,10 +178,10 @@ def delete(self):
         self.client.execute(query_str, {ids_param: str(self.uid)})
 
     def delete_model_run_data_rows(self, data_row_ids: List[str]):
-        """ Deletes data rows from model runs.
+        """ Deletes data rows from Model Runs.
 
         Args:
-            data_row_ids (list): List of data row ids to delete from the model run.
+            data_row_ids (list): List of data row ids to delete from the Model Run.
         Returns:
             Query execution success.
         """
@@ -262,6 +262,56 @@ def update_status(self,
             },
             experimental=True)
 
+    @experimental
+    def update_config(self, config: Dict[str, Any]) -> Dict[str, Any]:
+        """ 
+         Updates the Model Run's training metadata config
+         Args: 
+             config (dict): A dictionary of keys and values
+         Returns: 
+             Model Run id and updated training metadata
+         """
+        data: Dict[str, Any] = {'config': config}
+        res = self.client.execute(
+            """mutation updateModelRunConfigPyApi($modelRunId: ID!, $data: UpdateModelRunConfigInput!){
+                updateModelRunConfig(modelRun: {id : $modelRunId}, data: $data){trainingMetadata}
+            }
+        """, {
+                'modelRunId': self.uid,
+                'data': data
+            },
+            experimental=True)
+        return res["updateModelRunConfig"]
+
+    @experimental
+    def reset_config(self) -> Dict[str, Any]:
+        """ 
+         Resets Model Run's training metadata config
+         Returns: 
+             Model Run id and reset training metadata
+         """
+        res = self.client.execute(
+            """mutation resetModelRunConfigPyApi($modelRunId: ID!){
+                resetModelRunConfig(modelRun: {id : $modelRunId}){trainingMetadata}
+            }
+        """, {'modelRunId': self.uid},
+            experimental=True)
+        return res["resetModelRunConfig"]
+
+    @experimental
+    def get_config(self) -> Dict[str, Any]:
+        """ 
+         Gets Model Run's training metadata 
+         Returns: 
+             training metadata as a dictionary 
+         """
+        res = self.client.execute("""query ModelRunPyApi($modelRunId: ID!){
+                modelRun(where: {id : $modelRunId}){trainingMetadata}
+            }
+        """, {'modelRunId': self.uid},
+                                  experimental=True)
+        return res["modelRun"]["trainingMetadata"]
+
     @experimental
     def export_labels(
         self,

labelbox/schema/project.py

@@ -93,6 +93,22 @@ class QueueMode(Enum):
         Dataset = "Dataset"
 
     def update(self, **kwargs):
+        """ Updates this project with the specified attributes
+
+        Args:
+            kwargs: a dictionary containing attributes to be upserted
+
+        Note that the quality setting cannot be changed after a project has been created. The quality mode
+            for a project is inferred through the following attributes:
+            Benchmark:
+                auto_audit_number_of_labels = 1
+                auto_audit_percentage = 1.0
+            Consensus:
+                auto_audit_number_of_labels > 1
+                auto_audit_percentage <= 1.0
+            Attempting to switch between benchmark and consensus modes is an invalid operation and will result
+            in an error.
+        """
         mode: Optional[Project.QueueMode] = kwargs.pop("queue_mode", None)
         if mode:
             self._update_queue_mode(mode)

tests/data/assets/labelbox_v1/pdf_export.json

@@ -0,0 +1,130 @@
+[{
+    "ID": "cl6xnzi4a7ldn0729381g7104",
+    "DataRow ID": "cl6xnv9h61fv0085yhtoq06ht",
+    "Labeled Data": "https://storage.labelbox.com/ckcz6bubudyfi0855o1dt1g9s%2F4cef4e08-e13d-8a5e-fbbf-c7624babb490-Airbnb_%20Labelbox%20-%20Focus%20on%20Workforce%20-%20Labelbox%20Labeling%20Operations%20(1).pdf?Expires=1661971050348&KeyName=labelbox-assets-key-3&Signature=JK6ral5CXF7T9Q5LaQqKvJy5A2A",
+    "Label": {
+        "objects": [{
+            "featureId": "cl6xnzjpq0dmr07yocs2vfot8",
+            "schemaId": "cl6xnuwt95lqq07330tbb3mfd",
+            "color": "#1CE6FF",
+            "title": "boxy",
+            "value": "boxy",
+            "bbox": {
+                "top": 144.68,
+                "left": 107.84,
+                "height": 441.6,
+                "width": 9.48
+            },
+            "page": 0,
+            "unit": "POINTS",
+            "instanceURI": "https://api.labelbox.com/masks/feature/cl6xnzjpq0dmr07yocs2vfot8?token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJ1c2VySWQiOiJja2NjOWZtbXc0aGNkMDczOHFpeWM2YW54Iiwib3JnYW5pemF0aW9uSWQiOiJja2N6NmJ1YnVkeWZpMDg1NW8xZHQxZzlzIiwiaWF0IjoxNjYwNzYxNDUwLCJleHAiOjE2NjMzNTM0NTB9.X4-j6zee8o685PUrL9C6oC2m6TayKuJQHhN8iLgG8kI"
+        }, {
+            "featureId": "cl6xnzjpq0dms07yobwv68gxf",
+            "schemaId": "cl6xnuwt95lqq07330tbb3mfd",
+            "color": "#1CE6FF",
+            "title": "boxy",
+            "value": "boxy",
+            "bbox": {
+                "top": 162.73,
+                "left": 32.45,
+                "height": 388.17,
+                "width": 101.66
+            },
+            "page": 4,
+            "unit": "POINTS",
+            "instanceURI": "https://api.labelbox.com/masks/feature/cl6xnzjpq0dms07yobwv68gxf?token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJ1c2VySWQiOiJja2NjOWZtbXc0aGNkMDczOHFpeWM2YW54Iiwib3JnYW5pemF0aW9uSWQiOiJja2N6NmJ1YnVkeWZpMDg1NW8xZHQxZzlzIiwiaWF0IjoxNjYwNzYxNDUwLCJleHAiOjE2NjMzNTM0NTB9.X4-j6zee8o685PUrL9C6oC2m6TayKuJQHhN8iLgG8kI"
+        }, {
+            "featureId": "cl6xnzjpq0dmt07yo8pp45gru",
+            "schemaId": "cl6xnuwt95lqq07330tbb3mfd",
+            "color": "#1CE6FF",
+            "title": "boxy",
+            "value": "boxy",
+            "bbox": {
+                "top": 223.26,
+                "left": 251.42,
+                "height": 457.04,
+                "width": 186.78
+            },
+            "page": 7,
+            "unit": "POINTS",
+            "instanceURI": "https://api.labelbox.com/masks/feature/cl6xnzjpq0dmt07yo8pp45gru?token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJ1c2VySWQiOiJja2NjOWZtbXc0aGNkMDczOHFpeWM2YW54Iiwib3JnYW5pemF0aW9uSWQiOiJja2N6NmJ1YnVkeWZpMDg1NW8xZHQxZzlzIiwiaWF0IjoxNjYwNzYxNDUwLCJleHAiOjE2NjMzNTM0NTB9.X4-j6zee8o685PUrL9C6oC2m6TayKuJQHhN8iLgG8kI"
+        }, {
+            "featureId": "cl6xnzjpq0dmu07yo2qik0en4",
+            "schemaId": "cl6xnuwt95lqq07330tbb3mfd",
+            "color": "#1CE6FF",
+            "title": "boxy",
+            "value": "boxy",
+            "bbox": {
+                "top": 32.52,
+                "left": 218.17,
+                "height": 231.73,
+                "width": 110.56
+            },
+            "page": 6,
+            "unit": "POINTS",
+            "instanceURI": "https://api.labelbox.com/masks/feature/cl6xnzjpq0dmu07yo2qik0en4?token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJ1c2VySWQiOiJja2NjOWZtbXc0aGNkMDczOHFpeWM2YW54Iiwib3JnYW5pemF0aW9uSWQiOiJja2N6NmJ1YnVkeWZpMDg1NW8xZHQxZzlzIiwiaWF0IjoxNjYwNzYxNDUwLCJleHAiOjE2NjMzNTM0NTB9.X4-j6zee8o685PUrL9C6oC2m6TayKuJQHhN8iLgG8kI"
+        }, {
+            "featureId": "cl6xnzjpq0dmv07yo7phz7ofz",
+            "schemaId": "cl6xnuwt95lqq07330tbb3mfd",
+            "color": "#1CE6FF",
+            "title": "boxy",
+            "value": "boxy",
+            "bbox": {
+                "top": 117.39,
+                "left": 4.25,
+                "height": 456.92,
+                "width": 164.83
+            },
+            "page": 7,
+            "unit": "POINTS",
+            "instanceURI": "https://api.labelbox.com/masks/feature/cl6xnzjpq0dmv07yo7phz7ofz?token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJ1c2VySWQiOiJja2NjOWZtbXc0aGNkMDczOHFpeWM2YW54Iiwib3JnYW5pemF0aW9uSWQiOiJja2N6NmJ1YnVkeWZpMDg1NW8xZHQxZzlzIiwiaWF0IjoxNjYwNzYxNDUwLCJleHAiOjE2NjMzNTM0NTB9.X4-j6zee8o685PUrL9C6oC2m6TayKuJQHhN8iLgG8kI"
+        }, {
+            "featureId": "cl6xnzjpq0dmw07yofocp6uf6",
+            "schemaId": "cl6xnuwt95lqq07330tbb3mfd",
+            "color": "#1CE6FF",
+            "title": "boxy",
+            "value": "boxy",
+            "bbox": {
+                "top": 82.13,
+                "left": 217.28,
+                "height": 279.76,
+                "width": 82.43
+            },
+            "page": 8,
+            "unit": "POINTS",
+            "instanceURI": "https://api.labelbox.com/masks/feature/cl6xnzjpq0dmw07yofocp6uf6?token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJ1c2VySWQiOiJja2NjOWZtbXc0aGNkMDczOHFpeWM2YW54Iiwib3JnYW5pemF0aW9uSWQiOiJja2N6NmJ1YnVkeWZpMDg1NW8xZHQxZzlzIiwiaWF0IjoxNjYwNzYxNDUwLCJleHAiOjE2NjMzNTM0NTB9.X4-j6zee8o685PUrL9C6oC2m6TayKuJQHhN8iLgG8kI"
+        }, {
+            "featureId": "cl6xnzjpq0dmx07yo0qh40z0n",
+            "schemaId": "cl6xnuwt95lqq07330tbb3mfd",
+            "color": "#1CE6FF",
+            "title": "boxy",
+            "value": "boxy",
+            "bbox": {
+                "top": 298.12,
+                "left": 83.34,
+                "height": 203.83,
+                "width": 0.38
+            },
+            "page": 3,
+            "unit": "POINTS",
+            "instanceURI": "https://api.labelbox.com/masks/feature/cl6xnzjpq0dmx07yo0qh40z0n?token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJ1c2VySWQiOiJja2NjOWZtbXc0aGNkMDczOHFpeWM2YW54Iiwib3JnYW5pemF0aW9uSWQiOiJja2N6NmJ1YnVkeWZpMDg1NW8xZHQxZzlzIiwiaWF0IjoxNjYwNzYxNDUwLCJleHAiOjE2NjMzNTM0NTB9.X4-j6zee8o685PUrL9C6oC2m6TayKuJQHhN8iLgG8kI"
+        }],
+        "classifications": [],
+        "relationships": []
+    },
+    "Created By": "jtso@labelbox.com",
+    "Project Name": "PDF MAL Test",
+    "Created At": "2022-08-17T18:37:18.000Z",
+    "Updated At": "2022-08-17T18:37:20.073Z",
+    "Seconds to Label": 15.003,
+    "External ID": "Airbnb_ Labelbox - Focus on Workforce - Labelbox Labeling Operations (1).pdf",
+    "Global Key": null,
+    "Agreement": -1,
+    "Benchmark Agreement": -1,
+    "Benchmark ID": null,
+    "Dataset Name": "PDF ",
+    "Reviews": [],
+    "View Label": "https://editor.labelbox.com?project=cl6xntneb7t28072bggdydv7a&label=cl6xnzi4a7ldn0729381g7104",
+    "Has Open Issues": 0,
+    "Skipped": false
+}]

tests/data/serialization/labelbox_v1/test_document.py

@@ -0,0 +1,45 @@
+import json
+from typing import Dict, Any
+
+from labelbox.data.serialization.labelbox_v1.converter import LBV1Converter
+
+IGNORE_KEYS = [
+    "Data Split", "media_type", "DataRow Metadata", "Media Attributes"
+]
+
+
+def round_dict(data: Dict[str, Any]) -> Dict[str, Any]:
+    for key in data:
+        if isinstance(data[key], float):
+            data[key] = int(data[key])
+        elif isinstance(data[key], dict):
+            data[key] = round_dict(data[key])
+    return data
+
+
+def test_pdf():
+    """
+    Tests an export from a pdf document with only bounding boxes
+    """
+    payload = json.load(
+        open('tests/data/assets/labelbox_v1/pdf_export.json', 'r'))
+    collection = LBV1Converter.deserialize(payload)
+    serialized = next(LBV1Converter.serialize(collection))
+
+    payload = payload[0]  # only one document in the export
+
+    serialized = {k: v for k, v in serialized.items() if k not in IGNORE_KEYS}
+
+    assert serialized.keys() == payload.keys()
+    for key in payload.keys():
+        if key == 'Label':
+            serialized_no_classes = [{
+                k: v for k, v in dic.items() if k != 'classifications'
+            } for dic in serialized[key]['objects']]
+            serialized_round = [
+                round_dict(dic) for dic in serialized_no_classes
+            ]
+            payload_round = [round_dict(dic) for dic in payload[key]['objects']]
+            assert payload_round == serialized_round
+        else:
+            assert serialized[key] == payload[key]