rapidata 2.41.3__py3-none-any.whl → 2.42.1__py3-none-any.whl

rapidata/rapidata_client/validation/validation_set_manager.py

@@ -47,7 +47,7 @@ class ValidationSetManager:
 
     def __init__(self, openapi_service: OpenAPIService) -> None:
         self.__openapi_service = openapi_service
-        self.rapid = RapidsManager()
+        self.rapid = RapidsManager(openapi_service)
         logger.debug("ValidationSetManager initialized")
 
     def _create_order_validation_set(
@@ -66,11 +66,13 @@ class ValidationSetManager:
                     Rapid(
                         asset=datapoint.asset,
                         payload=workflow._to_payload(datapoint),
-                        metadata=datapoint.metadata,
+                        context=datapoint.context,
+                        media_context=datapoint.media_context,
+                        data_type=datapoint.data_type,
                         settings=settings,
                     )
                 )
-            return self._submit(name=order_name, rapids=rapids, dimensions=None)
+            return self._submit(name=order_name, rapids=rapids, dimensions=[])
 
     def create_classification_set(
         self,
@@ -143,11 +145,6 @@ class ValidationSetManager:
             logger.debug("Creating classification rapids")
             rapids: list[Rapid] = []
             for i in range(len(datapoints)):
-                rapid_metadata = []
-                if contexts:
-                    rapid_metadata.append(PromptMetadata(contexts[i]))
-                if media_contexts:
-                    rapid_metadata.append(MediaAssetMetadata(media_contexts[i]))
                 rapids.append(
                     self.rapid.classification_rapid(
                         instruction=instruction,
@@ -155,7 +152,10 @@ class ValidationSetManager:
                         datapoint=datapoints[i],
                         truths=truths[i],
                         data_type=data_type,
-                        metadata=rapid_metadata,
+                        context=contexts[i] if contexts != None else None,
+                        media_context=(
+                            media_contexts[i] if media_contexts != None else None
+                        ),
                         explanation=explanations[i] if explanations != None else None,
                     )
                 )
@@ -231,18 +231,16 @@ class ValidationSetManager:
             logger.debug("Creating comparison rapids")
             rapids: list[Rapid] = []
             for i in range(len(datapoints)):
-                rapid_metadata = []
-                if contexts:
-                    rapid_metadata.append(PromptMetadata(contexts[i]))
-                if media_contexts:
-                    rapid_metadata.append(MediaAssetMetadata(media_contexts[i]))
                 rapids.append(
                     self.rapid.compare_rapid(
                         instruction=instruction,
                         truth=truths[i],
                         datapoint=datapoints[i],
                         data_type=data_type,
-                        metadata=rapid_metadata,
+                        context=contexts[i] if contexts != None else None,
+                        media_context=(
+                            media_contexts[i] if media_contexts != None else None
+                        ),
                         explanation=explanation[i] if explanation != None else None,
                     )
                 )
@@ -387,17 +385,15 @@ class ValidationSetManager:
             rapids = []
             rapids: list[Rapid] = []
             for i in range(len(datapoints)):
-                rapid_metadata = []
-                if contexts:
-                    rapid_metadata.append(PromptMetadata(contexts[i]))
-                if media_contexts:
-                    rapid_metadata.append(MediaAssetMetadata(media_contexts[i]))
                 rapids.append(
                     self.rapid.locate_rapid(
                         instruction=instruction,
                         truths=truths[i],
                         datapoint=datapoints[i],
-                        metadata=rapid_metadata,
+                        context=contexts[i] if contexts != None else None,
+                        media_context=(
+                            media_contexts[i] if media_contexts != None else None
+                        ),
                         explanation=explanation[i] if explanation != None else None,
                     )
                 )
@@ -466,17 +462,15 @@ class ValidationSetManager:
             logger.debug("Creating draw rapids")
             rapids: list[Rapid] = []
             for i in range(len(datapoints)):
-                rapid_metadata = []
-                if contexts:
-                    rapid_metadata.append(PromptMetadata(contexts[i]))
-                if media_contexts:
-                    rapid_metadata.append(MediaAssetMetadata(media_contexts[i]))
                 rapids.append(
                     self.rapid.draw_rapid(
                         instruction=instruction,
                         truths=truths[i],
                         datapoint=datapoints[i],
-                        metadata=rapid_metadata,
+                        context=contexts[i] if contexts != None else None,
+                        media_context=(
+                            media_contexts[i] if media_contexts != None else None
+                        ),
                         explanation=explanation[i] if explanation != None else None,
                     )
                 )
@@ -546,17 +540,15 @@ class ValidationSetManager:
             logger.debug("Creating timestamp rapids")
             rapids: list[Rapid] = []
             for i in range(len(datapoints)):
-                rapid_metadata = []
-                if contexts:
-                    rapid_metadata.append(PromptMetadata(contexts[i]))
-                if media_contexts:
-                    rapid_metadata.append(MediaAssetMetadata(media_contexts[i]))
                 rapids.append(
                     self.rapid.timestamp_rapid(
                         instruction=instruction,
                         truths=truths[i],
                         datapoint=datapoints[i],
-                        metadata=rapid_metadata,
+                        context=contexts[i] if contexts != None else None,
+                        media_context=(
+                            media_contexts[i] if media_contexts != None else None
+                        ),
                         explanation=explanation[i] if explanation != None else None,
                     )
                 )
@@ -587,7 +579,7 @@ class ValidationSetManager:
         self,
         name: str,
         rapids: list[Rapid],
-        dimensions: list[str] | None,
+        dimensions: list[str],
     ) -> RapidataValidationSet:
         logger.debug("Creating validation set")
         validation_set_id = (
@@ -611,16 +603,27 @@ class ValidationSetManager:
         with tracer.start_as_current_span("Adding rapids to validation set"):
             logger.debug("Adding rapids to validation set")
             failed_rapids = []
-            for rapid in tqdm(
-                rapids,
+
+            progress_bar = tqdm(
+                total=len(rapids),
                 desc="Uploading validation tasks",
                 disable=rapidata_config.logging.silent_mode,
-            ):
+            )
+
+            for rapid in rapids:
                 try:
                     validation_set.add_rapid(rapid)
-                except Exception:
+                    progress_bar.update(1)
+                except Exception as e:
+                    logger.error(
+                        "Failed to add rapid %s to validation set.\nError: %s",
+                        rapid.asset,
+                        str(e),
+                    )
                     failed_rapids.append(rapid.asset)
 
+            progress_bar.close()
+
             if failed_rapids:
                 logger.error(
                     "Failed to add %s datapoints to validation set: %s",

rapidata/rapidata_client/workflow/_ranking_workflow.py

@@ -5,13 +5,13 @@ from rapidata.api_client import (
 )
 from rapidata.api_client.models.compare_workflow_model import CompareWorkflowModel
 from rapidata.rapidata_client.workflow._base_workflow import Workflow
-from rapidata.rapidata_client.datapoints.metadata import PromptMetadata
-from rapidata.api_client.models.dataset_dataset_id_datapoints_post_request_metadata_inner import (
-    DatasetDatasetIdDatapointsPostRequestMetadataInner,
-)
 from rapidata.api_client import ComparePayload
 from rapidata.rapidata_client.datapoints._datapoint import Datapoint
 from rapidata.api_client.models.rapid_modality import RapidModality
+from rapidata.rapidata_client.datapoints.metadata import Metadata
+from rapidata.api_client.models.create_datapoint_from_files_model_metadata_inner import (
+    CreateDatapointFromFilesModelMetadataInner,
+)
 
 
 class RankingWorkflow(Workflow):
@@ -25,19 +25,11 @@ class RankingWorkflow(Workflow):
         elo_start: int = 1200,
         elo_k_factor: int = 40,
         elo_scaling_factor: int = 400,
-        context: str | None = None,
+        metadatas: list[Metadata] = [],
     ):
         super().__init__(type="CompareWorkflowConfig")
 
-        self.context = (
-            [
-                DatasetDatasetIdDatapointsPostRequestMetadataInner(
-                    PromptMetadata(context).to_model()
-                )
-            ]
-            if context
-            else None
-        )
+        self.metadatas = metadatas
 
         self.criteria = criteria
         self.total_comparison_budget = total_comparison_budget
@@ -67,7 +59,10 @@ class RankingWorkflow(Workflow):
             criteria=self.criteria,
             eloConfig=self.elo_config,
             pairMakerConfig=self.pair_maker_config,
-            metadata=self.context,
+            metadata=[
+                CreateDatapointFromFilesModelMetadataInner(metadata.to_model())
+                for metadata in self.metadatas
+            ],
         )
 
     def _to_payload(self, datapoint: Datapoint) -> ComparePayload:
@@ -77,7 +72,9 @@ class RankingWorkflow(Workflow):
         )
 
     def __str__(self) -> str:
-        return f"RankingWorkflow(criteria='{self.criteria}', context={self.context})"
+        return (
+            f"RankingWorkflow(criteria='{self.criteria}', metadatas={self.metadatas})"
+        )
 
     def __repr__(self) -> str:
-        return f"RankingWorkflow(criteria={self.criteria!r}, total_comparison_budget={self.total_comparison_budget!r}, random_comparisons_ratio={self.random_comparisons_ratio!r}, elo_start={self.elo_start!r}, elo_k_factor={self.elo_k_factor!r}, elo_scaling_factor={self.elo_scaling_factor!r}, context={self.context!r})"
+        return f"RankingWorkflow(criteria={self.criteria!r}, total_comparison_budget={self.total_comparison_budget!r}, random_comparisons_ratio={self.random_comparisons_ratio!r}, elo_start={self.elo_start!r}, elo_k_factor={self.elo_k_factor!r}, elo_scaling_factor={self.elo_scaling_factor!r}, metadatas={self.metadatas!r})"

rapidata/rapidata_client/workflow/_select_words_workflow.py

@@ -8,9 +8,6 @@ from rapidata.api_client.models.transcription_rapid_blueprint import (
 from rapidata.rapidata_client.workflow._base_workflow import Workflow
 from rapidata.api_client import TranscriptionPayload, TranscriptionWord
 from rapidata.rapidata_client.datapoints._datapoint import Datapoint
-from rapidata.rapidata_client.datapoints.metadata._select_words_metadata import (
-    SelectWordsMetadata,
-)
 from rapidata.api_client.models.rapid_modality import RapidModality
 
 
@@ -46,25 +43,15 @@ class SelectWordsWorkflow(Workflow):
 
     def _to_payload(self, datapoint: Datapoint) -> TranscriptionPayload:
         assert (
-            datapoint.metadata is not None
-        ), "SelectWordsWorkflow requires a metadata datapoint"
-
-        assert any(
-            isinstance(metadata, SelectWordsMetadata) for metadata in datapoint.metadata
-        ), "SelectWordsWorkflow requires a SelectWordsMetadata datapoint"
-
-        select_words_metadata = next(
-            metadata
-            for metadata in datapoint.metadata
-            if isinstance(metadata, SelectWordsMetadata)
-        )
+            datapoint.sentence is not None
+        ), "SelectWordsWorkflow requires a sentence datapoint"
 
         return TranscriptionPayload(
             _t="TranscriptionPayload",
             title=self._instruction,
             transcription=[
                 TranscriptionWord(word=word, wordIndex=i)
-                for i, word in enumerate(select_words_metadata.select_words.split())
+                for i, word in enumerate(datapoint.sentence.split())
             ],
         )
 

rapidata/service/openapi_service.py

@@ -1,12 +1,13 @@
 import subprocess
 from importlib.metadata import version, PackageNotFoundError
 
+from rapidata.api_client import CustomerRapidApi
 from rapidata.api_client.api.campaign_api import CampaignApi
+from rapidata.api_client.api.asset_api import AssetApi
 from rapidata.api_client.api.dataset_api import DatasetApi
 from rapidata.api_client.api.benchmark_api import BenchmarkApi
 from rapidata.api_client.api.order_api import OrderApi
 from rapidata.api_client.api.pipeline_api import PipelineApi
-from rapidata.api_client.api.rapid_api import RapidApi
 from rapidata.api_client.api.leaderboard_api import LeaderboardApi
 from rapidata.api_client.api.validation_set_api import ValidationSetApi
 from rapidata.api_client.api.workflow_api import WorkflowApi
@@ -94,6 +95,10 @@ class OpenAPIService:
     def order_api(self) -> OrderApi:
         return OrderApi(self.api_client)
 
+    @property
+    def asset_api(self) -> AssetApi:
+        return AssetApi(self.api_client)
+
     @property
     def dataset_api(self) -> DatasetApi:
         return DatasetApi(self.api_client)
@@ -103,8 +108,8 @@ class OpenAPIService:
         return ValidationSetApi(self.api_client)
 
     @property
-    def rapid_api(self) -> RapidApi:
-        return RapidApi(self.api_client)
+    def rapid_api(self) -> CustomerRapidApi:
+        return CustomerRapidApi(self.api_client)
 
     @property
     def campaign_api(self) -> CampaignApi:

{rapidata-2.41.3.dist-info → rapidata-2.42.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rapidata
-Version: 2.41.3
+Version: 2.42.1
 Summary: Rapidata package containing the Rapidata Python Client to interact with the Rapidata Web API in an easy way.
 License: Apache-2.0
 License-File: LICENSE