label-studio-sdk 1.0.17__tar.gz → 1.0.18__tar.gz

{label_studio_sdk-1.0.17 → label_studio_sdk-1.0.18}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: label-studio-sdk
-Version: 1.0.17
+Version: 1.0.18
 Summary: 
 Requires-Python: >=3.9,<4
 Classifier: Intended Audience :: Developers

{label_studio_sdk-1.0.17 → label_studio_sdk-1.0.18}/pyproject.toml

@@ -3,7 +3,7 @@ name = "label-studio-sdk"
 
 [tool.poetry]
 name = "label-studio-sdk"
-version = "1.0.17"
+version = "1.0.18"
 description = ""
 readme = "README.md"
 authors = []

{label_studio_sdk-1.0.17 → label_studio_sdk-1.0.18}/src/label_studio_sdk/converter/converter.py

@@ -481,6 +481,9 @@ class Converter(object):
         then the from_name "my_output_tag_0" should match it, and we should return "my_output_tag_{{idx}}".
         """
 
+        best_match = None
+        best_match_len = 0
+
         for tag_name, tag_info in self._schema.items():
             if tag_name == from_name:
                 return tag_name
@@ -492,10 +495,13 @@ class Converter(object):
             for variable, regex in tag_info["regex"].items():
                 tag_name_pattern = tag_name_pattern.replace(variable, regex)
 
-            if re.compile(tag_name_pattern).match(from_name):
-                return tag_name
+            # In some cases there are tags with same prefix - we need to find the best or longest matching pattern
+            if r := re.compile(tag_name_pattern).match(from_name):
+                if match_len := len(tag_name_pattern) > best_match_len:
+                    best_match = tag_name
+                    best_match_len = len(tag_name_pattern)
 
-        return None
+        return best_match if best_match else None
 
     def annotation_result_from_task(self, task):
         has_annotations = "completions" in task or "annotations" in task

{label_studio_sdk-1.0.17 → label_studio_sdk-1.0.18}/src/label_studio_sdk/projects/client.py

@@ -207,7 +207,8 @@ class ProjectsClient:
             Project color in HEX format
 
         control_weights : typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]]
-            Dict of weights for each control tag in metric calculation. Each control tag (e.g. label or choice) will have its own key in control weight dict with weight for each label and overall weight. For example, if a bounding box annotation with a control tag named my_bbox should be included with 0.33 weight in agreement calculation, and the first label Car should be twice as important as Airplane, then you need to specify: {'my_bbox': {'type': 'RectangleLabels', 'labels': {'Car': 1.0, 'Airplane': 0.5}, 'overall': 0.33}
+            Dict of weights for each control tag in metric calculation. Each control tag (e.g. label or choice) will have its own key in control weight dict with weight for each label and overall weight. For example, if a bounding box annotation with a control tag named my_bbox should be included with 0.33 weight in agreement calculation, and the first label Car should be twice as important as Airplane, then you need to specify: \{'my_bbox': \{'type': 'RectangleLabels', 'labels': \{'Car': 1.0, 'Airplane': 0.5}, 'overall': 0.33}}
+
 
         workspace : typing.Optional[int]
             Workspace ID
@@ -440,7 +441,8 @@ class ProjectsClient:
             Project color in HEX format
 
         control_weights : typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]]
-            Dict of weights for each control tag in metric calculation. Each control tag (e.g. label or choice) will have its own key in control weight dict with weight for each label and overall weight. For example, if a bounding box annotation with a control tag named my_bbox should be included with 0.33 weight in agreement calculation, and the first label Car should be twice as important as Airplane, then you need to specify: {'my_bbox': {'type': 'RectangleLabels', 'labels': {'Car': 1.0, 'Airplane': 0.5}, 'overall': 0.33}
+            Dict of weights for each control tag in metric calculation. Each control tag (e.g. label or choice) will have its own key in control weight dict with weight for each label and overall weight. For example, if a bounding box annotation with a control tag named my_bbox should be included with 0.33 weight in agreement calculation, and the first label Car should be twice as important as Airplane, then you need to specify: \{'my_bbox': \{'type': 'RectangleLabels', 'labels': \{'Car': 1.0, 'Airplane': 0.5}, 'overall': 0.33}}
+
 
         workspace : typing.Optional[int]
             Workspace ID
@@ -890,7 +892,8 @@ class AsyncProjectsClient:
             Project color in HEX format
 
         control_weights : typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]]
-            Dict of weights for each control tag in metric calculation. Each control tag (e.g. label or choice) will have its own key in control weight dict with weight for each label and overall weight. For example, if a bounding box annotation with a control tag named my_bbox should be included with 0.33 weight in agreement calculation, and the first label Car should be twice as important as Airplane, then you need to specify: {'my_bbox': {'type': 'RectangleLabels', 'labels': {'Car': 1.0, 'Airplane': 0.5}, 'overall': 0.33}
+            Dict of weights for each control tag in metric calculation. Each control tag (e.g. label or choice) will have its own key in control weight dict with weight for each label and overall weight. For example, if a bounding box annotation with a control tag named my_bbox should be included with 0.33 weight in agreement calculation, and the first label Car should be twice as important as Airplane, then you need to specify: \{'my_bbox': \{'type': 'RectangleLabels', 'labels': \{'Car': 1.0, 'Airplane': 0.5}, 'overall': 0.33}}
+
 
         workspace : typing.Optional[int]
             Workspace ID
@@ -1147,7 +1150,8 @@ class AsyncProjectsClient:
             Project color in HEX format
 
         control_weights : typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]]
-            Dict of weights for each control tag in metric calculation. Each control tag (e.g. label or choice) will have its own key in control weight dict with weight for each label and overall weight. For example, if a bounding box annotation with a control tag named my_bbox should be included with 0.33 weight in agreement calculation, and the first label Car should be twice as important as Airplane, then you need to specify: {'my_bbox': {'type': 'RectangleLabels', 'labels': {'Car': 1.0, 'Airplane': 0.5}, 'overall': 0.33}
+            Dict of weights for each control tag in metric calculation. Each control tag (e.g. label or choice) will have its own key in control weight dict with weight for each label and overall weight. For example, if a bounding box annotation with a control tag named my_bbox should be included with 0.33 weight in agreement calculation, and the first label Car should be twice as important as Airplane, then you need to specify: \{'my_bbox': \{'type': 'RectangleLabels', 'labels': \{'Car': 1.0, 'Airplane': 0.5}, 'overall': 0.33}}
+
 
         workspace : typing.Optional[int]
             Workspace ID

{label_studio_sdk-1.0.17 → label_studio_sdk-1.0.18}/src/label_studio_sdk/projects/types/projects_create_response.py

@@ -78,7 +78,7 @@ class ProjectsCreateResponse(UniversalBaseModel):
 
     control_weights: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = pydantic.Field(default=None)
     """
-    Dict of weights for each control tag in metric calculation. Each control tag (e.g. label or choice) will have its own key in control weight dict with weight for each label and overall weight. For example, if a bounding box annotation with a control tag named my_bbox should be included with 0.33 weight in agreement calculation, and the first label Car should be twice as important as Airplane, then you need to specify: {'my_bbox': {'type': 'RectangleLabels', 'labels': {'Car': 1.0, 'Airplane': 0.5}, 'overall': 0.33}
+    Dict of weights for each control tag in metric calculation. Each control tag (e.g. label or choice) will have its own key in control weight dict with weight for each label and overall weight. For example, if a bounding box annotation with a control tag named my_bbox should be included with 0.33 weight in agreement calculation, and the first label Car should be twice as important as Airplane, then you need to specify: \{'my_bbox': \{'type': 'RectangleLabels', 'labels': \{'Car': 1.0, 'Airplane': 0.5}, 'overall': 0.33}}
     """
 
     if IS_PYDANTIC_V2:

{label_studio_sdk-1.0.17 → label_studio_sdk-1.0.18}/src/label_studio_sdk/projects/types/projects_update_response.py

@@ -83,7 +83,7 @@ class ProjectsUpdateResponse(UniversalBaseModel):
 
     control_weights: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = pydantic.Field(default=None)
     """
-    Dict of weights for each control tag in metric calculation. Each control tag (e.g. label or choice) will have its own key in control weight dict with weight for each label and overall weight. For example, if a bounding box annotation with a control tag named my_bbox should be included with 0.33 weight in agreement calculation, and the first label Car should be twice as important as Airplane, then you need to specify: {'my_bbox': {'type': 'RectangleLabels', 'labels': {'Car': 1.0, 'Airplane': 0.5}, 'overall': 0.33}
+    Dict of weights for each control tag in metric calculation. Each control tag (e.g. label or choice) will have its own key in control weight dict with weight for each label and overall weight. For example, if a bounding box annotation with a control tag named my_bbox should be included with 0.33 weight in agreement calculation, and the first label Car should be twice as important as Airplane, then you need to specify: \{'my_bbox': \{'type': 'RectangleLabels', 'labels': \{'Car': 1.0, 'Airplane': 0.5}, 'overall': 0.33}}
     """
 
     if IS_PYDANTIC_V2:

{label_studio_sdk-1.0.17 → label_studio_sdk-1.0.18}/src/label_studio_sdk/prompts/client.py

@@ -397,6 +397,7 @@ class PromptsClient:
     def batch_predictions(
         self,
         *,
+        num_predictions: typing.Optional[int] = None,
         modelrun_id: typing.Optional[int] = OMIT,
         results: typing.Optional[typing.Sequence[PromptsBatchPredictionsRequestResultsItem]] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
@@ -406,6 +407,9 @@ class PromptsClient:
 
         Parameters
         ----------
+        num_predictions : typing.Optional[int]
+            Number of predictions being sent
+
         modelrun_id : typing.Optional[int]
             Model Run ID to associate the prediction with
 
@@ -431,6 +435,9 @@ class PromptsClient:
         _response = self._client_wrapper.httpx_client.request(
             "api/model-run/batch-predictions",
             method="POST",
+            params={
+                "num_predictions": num_predictions,
+            },
             json={
                 "modelrun_id": modelrun_id,
                 "results": convert_and_respect_annotation_metadata(
@@ -462,6 +469,7 @@ class PromptsClient:
     def batch_failed_predictions(
         self,
         *,
+        num_failed_predictions: typing.Optional[int] = None,
         modelrun_id: typing.Optional[int] = OMIT,
         failed_predictions: typing.Optional[
             typing.Sequence[PromptsBatchFailedPredictionsRequestFailedPredictionsItem]
@@ -473,6 +481,9 @@ class PromptsClient:
 
         Parameters
         ----------
+        num_failed_predictions : typing.Optional[int]
+            Number of failed predictions being sent
+
         modelrun_id : typing.Optional[int]
             Model Run ID where the failed predictions came from
 
@@ -498,6 +509,9 @@ class PromptsClient:
         _response = self._client_wrapper.httpx_client.request(
             "api/model-run/batch-failed-predictions",
             method="POST",
+            params={
+                "num_failed_predictions": num_failed_predictions,
+            },
             json={
                 "modelrun_id": modelrun_id,
                 "failed_predictions": convert_and_respect_annotation_metadata(
@@ -933,6 +947,7 @@ class AsyncPromptsClient:
     async def batch_predictions(
         self,
         *,
+        num_predictions: typing.Optional[int] = None,
         modelrun_id: typing.Optional[int] = OMIT,
         results: typing.Optional[typing.Sequence[PromptsBatchPredictionsRequestResultsItem]] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
@@ -942,6 +957,9 @@ class AsyncPromptsClient:
 
         Parameters
         ----------
+        num_predictions : typing.Optional[int]
+            Number of predictions being sent
+
         modelrun_id : typing.Optional[int]
             Model Run ID to associate the prediction with
 
@@ -975,6 +993,9 @@ class AsyncPromptsClient:
         _response = await self._client_wrapper.httpx_client.request(
             "api/model-run/batch-predictions",
             method="POST",
+            params={
+                "num_predictions": num_predictions,
+            },
             json={
                 "modelrun_id": modelrun_id,
                 "results": convert_and_respect_annotation_metadata(
@@ -1006,6 +1027,7 @@ class AsyncPromptsClient:
     async def batch_failed_predictions(
         self,
         *,
+        num_failed_predictions: typing.Optional[int] = None,
         modelrun_id: typing.Optional[int] = OMIT,
         failed_predictions: typing.Optional[
             typing.Sequence[PromptsBatchFailedPredictionsRequestFailedPredictionsItem]
@@ -1017,6 +1039,9 @@ class AsyncPromptsClient:
 
         Parameters
         ----------
+        num_failed_predictions : typing.Optional[int]
+            Number of failed predictions being sent
+
         modelrun_id : typing.Optional[int]
             Model Run ID where the failed predictions came from
 
@@ -1050,6 +1075,9 @@ class AsyncPromptsClient:
         _response = await self._client_wrapper.httpx_client.request(
             "api/model-run/batch-failed-predictions",
             method="POST",
+            params={
+                "num_failed_predictions": num_failed_predictions,
+            },
             json={
                 "modelrun_id": modelrun_id,
                 "failed_predictions": convert_and_respect_annotation_metadata(

{label_studio_sdk-1.0.17 → label_studio_sdk-1.0.18}/src/label_studio_sdk/types/project.py

@@ -154,7 +154,7 @@ class Project(UniversalBaseModel):
 
     control_weights: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = pydantic.Field(default=None)
     """
-    Dict of weights for each control tag in metric calculation. Each control tag (e.g. label or choice) will have it's own key in control weight dict with weight for each label and overall weight.For example, if bounding box annotation with control tag named my_bbox should be included with 0.33 weight in agreement calculation, and the first label Car should be twice more important than Airplaine, then you have to need the specify: {'my_bbox': {'type': 'RectangleLabels', 'labels': {'Car': 1.0, 'Airplaine': 0.5}, 'overall': 0.33}
+    Dict of weights for each control tag in metric calculation. Each control tag (e.g. label or choice) will have its own key in control weight dict with weight for each label and overall weight. For example, if a bounding box annotation with a control tag named my_bbox should be included with 0.33 weight in agreement calculation, and the first label Car should be twice as important as Airplane, then you need to specify: \{'my_bbox': \{'type': 'RectangleLabels', 'labels': \{'Car': 1.0, 'Airplane': 0.5}, 'overall': 0.33}}
     """
 
     parsed_label_config: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = pydantic.Field(default=None)