arize-phoenix 11.27.0__py3-none-any.whl → 11.29.0__py3-none-any.whl

{arize_phoenix-11.27.0.dist-info → arize_phoenix-11.29.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: arize-phoenix
-Version: 11.27.0
+Version: 11.29.0
 Summary: AI Observability and Evaluation
 Project-URL: Documentation, https://arize.com/docs/phoenix/
 Project-URL: Issues, https://github.com/Arize-ai/phoenix/issues
@@ -50,7 +50,8 @@ Requires-Dist: python-multipart
 Requires-Dist: scikit-learn
 Requires-Dist: scipy
 Requires-Dist: sqlalchemy[asyncio]<3,>=2.0.4
-Requires-Dist: sqlean-py>=3.45.1
+Requires-Dist: sqlean-py<3.50,>=3.45.1; platform_system == 'Windows'
+Requires-Dist: sqlean-py>=3.45.1; platform_system != 'Windows'
 Requires-Dist: starlette
 Requires-Dist: strawberry-graphql==0.270.1
 Requires-Dist: tqdm

{arize_phoenix-11.27.0.dist-info → arize_phoenix-11.29.0.dist-info}/RECORD

@@ -6,7 +6,7 @@ phoenix/exceptions.py,sha256=n2L2KKuecrdflB9MsCdAYCiSEvGJptIsfRkXMoJle7A,169
 phoenix/py.typed,sha256=AbpHGcgLb-kRsJGnwFEktk7uzpZOCcBY74-YBdrKVGs,1
 phoenix/services.py,sha256=ngkyKGVatX3cO2WJdo2hKdaVKP-xJCMvqthvga6kJss,5196
 phoenix/settings.py,sha256=2kHfT3BNOVd4dAO1bq-syEQbHSG8oX2-7NhOwK2QREk,896
-phoenix/version.py,sha256=ofjFoTgUco9LdD4reUgJGjEcFX2RzUAJpVKpDorgoms,24
+phoenix/version.py,sha256=dmpBux9ABrJi8fLg-MbszS9TOYZ3loPQG7TMhFmf_iE,24
 phoenix/core/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 phoenix/core/embedding_dimension.py,sha256=zKGbcvwOXgLf-yrJBpQyKtd-LEOPRKHnUToyAU8Owis,87
 phoenix/core/model.py,sha256=qBFraOtmwCCnWJltKNP18DDG0mULXigytlFsa6YOz6k,4837
@@ -27,13 +27,13 @@ phoenix/db/pg_config.py,sha256=h6mB7qF7t4Zk6VGvAiyefHGVu74o-yJynaWzeE39k9Y,6001
 phoenix/db/insertion/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 phoenix/db/insertion/constants.py,sha256=8wifm7X-1XvroZ__R2Gc96NsgLhTDn0zXl4lehlXtcA,70
 phoenix/db/insertion/dataset.py,sha256=KeUtfSnznmkDDlGs6KM5URduZtIo8I951tybamwoLsc,7216
-phoenix/db/insertion/document_annotation.py,sha256=rLL35nxcBXh2O1RaY-R_nucER6wVPZBVJ3XkQQnmiQE,6498
+phoenix/db/insertion/document_annotation.py,sha256=Lo6pPEUNNBYFjN_PPKqX9aZdchyIOgX3wPZe7vPVwHA,6540
 phoenix/db/insertion/evaluation.py,sha256=XIlDQMx9FFu_hO1TEao8_ScgKETw61AoeGezsmngXrY,6876
 phoenix/db/insertion/helpers.py,sha256=wA4f_B8DI1pfNvX7R5luG1MklwX6rtCrvv_SOcjtJGo,4106
 phoenix/db/insertion/span.py,sha256=UB4UR8QlkJdVzbJ7_-S6FXIHYjyKZSOEVzCOvwGw7cY,8414
-phoenix/db/insertion/span_annotation.py,sha256=EqjT7OcCUO69PAw3shkOSQunxWtLNpToO7gtqlK8dbY,5682
-phoenix/db/insertion/trace_annotation.py,sha256=0wIrAEDMYDvwRbd7zYAGIcnZtdjoPEINP51m28Wk1VI,5722
-phoenix/db/insertion/types.py,sha256=hOwtuYeKA7rjRXT2bfsD_SwYnkrIpxdVirQ50jq1fws,8253
+phoenix/db/insertion/span_annotation.py,sha256=WA9538nPNIgRucRkd2S1zCC2TUMlJPvgMm1cCnH50Y4,5724
+phoenix/db/insertion/trace_annotation.py,sha256=UKmAsSRd1mAwdGWu-i2aEGC3Yu16tqfbpzqMGtJnkVU,5764
+phoenix/db/insertion/types.py,sha256=nIdqq-9vqciHMP4UrX0pmKnVUSfxfnO8l0ECPz0kiSs,8694
 phoenix/db/migrations/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 phoenix/db/migrations/env.py,sha256=tFO3ceuCx9Es5_2w_BXclaQMmNQKNX21b1UEV5mYdeo,3387
 phoenix/db/migrations/script.py.mako,sha256=MEqL-2qATlST9TAOeYgscMn1uy6HUS9NFvDgl93dMj8,635
@@ -120,14 +120,14 @@ phoenix/server/api/utils.py,sha256=quCBRcusc6PUq9tJq7M8PgwFZp7nXgVAxtbw8feribY,8
 phoenix/server/api/dataloaders/__init__.py,sha256=ddiX1BdbyGkPTzMZNo-hkF_2kqIquelBUFvQejnAJYk,6834
 phoenix/server/api/dataloaders/annotation_configs_by_project.py,sha256=_Nfiug9o01JimU3Z0LpZJ0uaMCjchXomyt_dYAxPFRY,1178
 phoenix/server/api/dataloaders/annotation_summaries.py,sha256=0b23-bucBKyL25RWb2QzCNJjQzrq403qMmHKUVD5W4M,14377
-phoenix/server/api/dataloaders/average_experiment_run_latency.py,sha256=GLFoFAbztOH-0FVzzZ77mATIO63UcjB50j3qXiNi-tE,1811
+phoenix/server/api/dataloaders/average_experiment_run_latency.py,sha256=EgRkx_2QIzdOZe4D-0kd3B9HMoAibU9DZDotUh2CWFE,1945
 phoenix/server/api/dataloaders/dataset_example_revisions.py,sha256=xF7M2dg3UmjhdCrscnztCIBBI0cg3RF48IIqvilpc18,4623
 phoenix/server/api/dataloaders/dataset_example_spans.py,sha256=z_MFquqAcJ9wat7BBp7MVeJ9BYuu2EZEdaog52iWDno,1390
 phoenix/server/api/dataloaders/document_evaluation_summaries.py,sha256=9fdROnzp-mymggHwNvpRkCk93LUFxxLy55-j3HP_2HY,5565
 phoenix/server/api/dataloaders/document_evaluations.py,sha256=KCnCItJ2DQOCHvxFp1KK2AStPN1akGlAGOEzEAu6c6I,1246
 phoenix/server/api/dataloaders/document_retrieval_metrics.py,sha256=37EcAW7oYQuWYHMDHb0wcqbWj9lhSskvzDO7NJbT5Js,4136
-phoenix/server/api/dataloaders/experiment_annotation_summaries.py,sha256=fFEjpJzUOhu_cJKQ-YnwEvtn8NDl6bPs-moERe-Bp04,2767
-phoenix/server/api/dataloaders/experiment_error_rates.py,sha256=V4U_y16LwfBDksTZ4QP1dDALGSsQAQ1KcttKRE6lXto,1920
+phoenix/server/api/dataloaders/experiment_annotation_summaries.py,sha256=Jfurk2bBqt3RTq1ydqZ4B4H_0mTyW6XMeIImShT1aQs,5643
+phoenix/server/api/dataloaders/experiment_error_rates.py,sha256=06IZF07qt2y167DBM49QkSNdnphPArhcsgYFcunaL-U,1992
 phoenix/server/api/dataloaders/experiment_run_annotations.py,sha256=uJ--9Ue4tnmZYH9Zy-Cj2Y2t3xzs2xUqN8Y6EGe1MCo,1296
 phoenix/server/api/dataloaders/experiment_run_counts.py,sha256=j_7229IL705p_TycxIMYylhSgLHsAn91TiGPYXtOuQ8,1617
 phoenix/server/api/dataloaders/experiment_sequence_number.py,sha256=zM_f78fnqhppLtevrx9iISQSN7w_BNeXT9CoX8jYgAI,1534
@@ -259,15 +259,15 @@ phoenix/server/api/routers/v1/__init__.py,sha256=ngLMPjC7lgZxgKy_Is33KxTRnMzSqy2
 phoenix/server/api/routers/v1/annotation_configs.py,sha256=xp5lJmKYlRsINCUrRD9-lTAElw2v4hdFndS5BWrxICA,16048
 phoenix/server/api/routers/v1/annotations.py,sha256=fVl2qeh_ZbWXGvFBTZgeL7aGkkINIScdjuyxnOoSzNM,6817
 phoenix/server/api/routers/v1/datasets.py,sha256=9iPORLmbOrPKgUUcRDMs6ZczSIz7hvc6bngJy3IbdR0,38331
-phoenix/server/api/routers/v1/evaluations.py,sha256=_I0X01J8EZmTlfAGkvMkZ05IB3x691xzUmHV05ov26Y,12914
+phoenix/server/api/routers/v1/evaluations.py,sha256=aBrPO-xCAWyTxydaHq7W2wQFm65k89uVR-H3VWsd6WQ,13062
 phoenix/server/api/routers/v1/experiment_evaluations.py,sha256=DZ3UK9OoYKElpRcEER7559-KiAqWr-1IXpZ27FbfP3k,5249
 phoenix/server/api/routers/v1/experiment_runs.py,sha256=LZeCQWQIEOZ9jK5Gp_C4JbiYY6AmnnWe85cVcvdkCLE,7107
 phoenix/server/api/routers/v1/experiments.py,sha256=hIBecGACzGZEgl93ap_JV52pUv-Ij03QJMRxQhBlktI,20611
 phoenix/server/api/routers/v1/models.py,sha256=p3gJN-9SWiUYTUTft4bZMsZVCBNTb4nN1Foy68eRZzQ,1997
 phoenix/server/api/routers/v1/projects.py,sha256=XR6uJxHXXtC1q8LNyS9W6iaj440sv1OKCu-OSBfxEys,12824
 phoenix/server/api/routers/v1/prompts.py,sha256=chRYcLkOYDJdJfVZVukVTUyIRnLPvsJCg41CuPxOIU8,26695
-phoenix/server/api/routers/v1/spans.py,sha256=la_bfxrVPGY-N8qvk9o9ICrGiwl1B6-jbZ0uDjR5QXc,49345
-phoenix/server/api/routers/v1/traces.py,sha256=Skn0N_L4ZjoJ7x76PBrqvbKPFiAk8xSe1yxfiOaQ0Gc,11285
+phoenix/server/api/routers/v1/spans.py,sha256=roDDE0RDBGSCircPgzwYWhwwyK33IaW7YliWN-lwxWw,49385
+phoenix/server/api/routers/v1/traces.py,sha256=ho6SXJ0R3g1ROxgtd7hNPo8QSZnA85oG0dgWDokHM2w,11365
 phoenix/server/api/routers/v1/users.py,sha256=eO8zMtGU33Td2_G1l9D7Z0a4CG1CwBUCj_Z9z2uk7wg,12089
 phoenix/server/api/routers/v1/utils.py,sha256=oXIOGPzPTkE0ZWUTRCoRIQQ7wTzoSwtWFaUSjlGBqts,4960
 phoenix/server/api/types/Annotation.py,sha256=gsl8CwjIbDUbZRj4d9USwZ_w_Tkz4i7zuZh9ftV80jA,1132
@@ -284,7 +284,7 @@ phoenix/server/api/types/CostBreakdown.py,sha256=yw9dlb0blGIB_dWNP8yEvDHJztHjpiV
 phoenix/server/api/types/CreateDatasetPayload.py,sha256=R-6zCmuD0f76RU9Giu78xwTHlASQs6Aq8yzvX1Kxc3g,140
 phoenix/server/api/types/CronExpression.py,sha256=R7oxuSSX_eTUHQWaoaSueQqWDmkkHr5dBKRN6q-6ROk,331
 phoenix/server/api/types/DataQualityMetric.py,sha256=Aieg3bHeBFaAf4mqeRcH1zT04sXAtQD8ATSHJt7FaBQ,1538
-phoenix/server/api/types/Dataset.py,sha256=StVJmOE996Citau11JtFTmcgLqvN9IeZsHAbe-Y1gkg,12933
+phoenix/server/api/types/Dataset.py,sha256=23dst_glr7kFNC62-q6D9H2hJgrfZnGe7V-Bg72SJgg,15303
 phoenix/server/api/types/DatasetExample.py,sha256=_9byxGpXfYb-hmFMUJeG7Bw1wsRKSJaHwF2IPAbFpFw,3115
 phoenix/server/api/types/DatasetExampleRevision.py,sha256=c-jWR6dTguEZTm54IMlFr0Ic84I3nefyDnZb7nF5hnI,874
 phoenix/server/api/types/DatasetValues.py,sha256=7VbCOLlzOXpZN80-zYF2UGuafRcPsZF-8WQNc0YsKFc,1119
@@ -303,7 +303,7 @@ phoenix/server/api/types/EvaluationSummary.py,sha256=vILYejnfPvMwWEXOwhQZsANvYe3
 phoenix/server/api/types/Event.py,sha256=iYt_Jx1Roioo0vZ0iPeJTHcTu6NSm4ilVMJ-IMUHAKk,3970
 phoenix/server/api/types/EventMetadata.py,sha256=-J0tYF9eZTHwCjwxQHY7Gckr2_MNW5OoWT1mydweZNM,635
 phoenix/server/api/types/ExampleRevisionInterface.py,sha256=gV3Gt9-3Oi5wjaVtepC6nOt3FzTzZFD1KebNnqiw56E,294
-phoenix/server/api/types/Experiment.py,sha256=8Hd-8-4Rcym95sPHIBLyTfMuRdyl50YxlKO_fMvdNAA,7830
+phoenix/server/api/types/Experiment.py,sha256=eN0NhUVdcjJyMbyAGoltv30mhunatWS6i2gG97xu5TU,7758
 phoenix/server/api/types/ExperimentAnnotationSummary.py,sha256=Uk3JtxIrsMoZT5tqc4nJdUOM3XegVzjUyoV3pkjNotE,256
 phoenix/server/api/types/ExperimentComparison.py,sha256=PXFcB0e8aaJ391yRsuRJr9_dvTZI1RAzF93oC_-HtxU,461
 phoenix/server/api/types/ExperimentRun.py,sha256=_fcwDLuURV0yviOlkjWAgJJwcCPdz-xGR6VX3UKf73s,6541
@@ -392,10 +392,10 @@ phoenix/server/static/apple-touch-icon-76x76.png,sha256=CT_xT12I0u2i0WU8JzBZBuOQ
 phoenix/server/static/apple-touch-icon.png,sha256=fOfpjqGpWYbJ0eAurKsyoZP1EAs6ZVooBJ_SGk2ZkDs,3801
 phoenix/server/static/favicon.ico,sha256=bY0vvCKRftemZfPShwZtE93DiiQdaYaozkPGwNFr6H8,34494
 phoenix/server/static/modernizr.js,sha256=mvK-XtkNqjOral-QvzoqsyOMECXIMu5BQwSVN_wcU9c,2564
-phoenix/server/static/.vite/manifest.json,sha256=f81ZT-s_X796lx4IwpiRFFg00JHq2_CI-G-1_tXxmSE,2328
-phoenix/server/static/assets/components-BC-SDWO9.js,sha256=_i4WPHkV2y5M3u6arDbZrRwF-qa9oWlXsK5q5CpvKsY,664485
-phoenix/server/static/assets/index-CjTh13rR.js,sha256=SSBxvFh8I1jiscNFNZzK69sDapeHOrhjRjhg11cF45Y,63396
-phoenix/server/static/assets/pages-D19Jzk-2.js,sha256=KEIJNicB4MfHx29SVywYIktUZBnVmeP3mLMM31uVpxM,1265049
+phoenix/server/static/.vite/manifest.json,sha256=-rNvMLAChq7_eDpTSPXWf7ujX9-KJZOw6I_ONqi1eBY,2328
+phoenix/server/static/assets/components-dCdVienD.js,sha256=a8fwhvVLGgLiQmj6R7zuXEdvLLpormuCrJePqF8ss5k,664533
+phoenix/server/static/assets/index-Bp44T8N2.js,sha256=kaA9zlqpA50LK3xdjN2K3tPAjq6roX3cy6OyCyBS49k,63396
+phoenix/server/static/assets/pages-CA4bKhm9.js,sha256=plhzt6_SL9xsAOva_dTqDk-0yWfre4t_nvg_VJcQeX0,1268001
 phoenix/server/static/assets/vendor-CqDb5u4o.css,sha256=zIyFiNJKxMaQk8AvtLgt1rR01oO10d1MFndSDKH9Clw,5517
 phoenix/server/static/assets/vendor-RdRDaQiR.js,sha256=oTxLetZZXJ20yoKNAYExto9V73y8X5zjddWV46K9CWM,2595492
 phoenix/server/static/assets/vendor-arizeai-DsYDNOqt.js,sha256=0HIkPJXbKTh85nqphdAXYeStRzdaim0IQxRXiXxa21U,121514
@@ -442,9 +442,9 @@ phoenix/utilities/project.py,sha256=auVpARXkDb-JgeX5f2aStyFIkeKvGwN9l7qrFeJMVxI,
 phoenix/utilities/re.py,sha256=6YyUWIkv0zc2SigsxfOWIHzdpjKA_TZo2iqKq7zJKvw,2081
 phoenix/utilities/span_store.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 phoenix/utilities/template_formatters.py,sha256=gh9PJD6WEGw7TEYXfSst1UR4pWWwmjxMLrDVQ_CkpkQ,2779
-arize_phoenix-11.27.0.dist-info/METADATA,sha256=QY0OG7N7HMD-gi2_15QJThyeIXch1CDC86bSt9S8vQA,31634
-arize_phoenix-11.27.0.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-arize_phoenix-11.27.0.dist-info/entry_points.txt,sha256=Pgpn8Upxx9P8z8joPXZWl2LlnAlGc3gcQoVchb06X1Q,94
-arize_phoenix-11.27.0.dist-info/licenses/IP_NOTICE,sha256=JBqyyCYYxGDfzQ0TtsQgjts41IJoa-hiwDrBjCb9gHM,469
-arize_phoenix-11.27.0.dist-info/licenses/LICENSE,sha256=HFkW9REuMOkvKRACuwLPT0hRydHb3zNg-fdFt94td18,3794
-arize_phoenix-11.27.0.dist-info/RECORD,,
+arize_phoenix-11.29.0.dist-info/METADATA,sha256=N5M1wqlrq9TDt1tg8K9INOEqK1VBdOtVqyGPmR28e34,31733
+arize_phoenix-11.29.0.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+arize_phoenix-11.29.0.dist-info/entry_points.txt,sha256=Pgpn8Upxx9P8z8joPXZWl2LlnAlGc3gcQoVchb06X1Q,94
+arize_phoenix-11.29.0.dist-info/licenses/IP_NOTICE,sha256=JBqyyCYYxGDfzQ0TtsQgjts41IJoa-hiwDrBjCb9gHM,469
+arize_phoenix-11.29.0.dist-info/licenses/LICENSE,sha256=HFkW9REuMOkvKRACuwLPT0hRydHb3zNg-fdFt94td18,3794
+arize_phoenix-11.29.0.dist-info/RECORD,,

phoenix/db/insertion/document_annotation.py

@@ -63,7 +63,7 @@ class DocumentAnnotationQueueInserter(
         session: AsyncSession,
         *insertions: Insertables.DocumentAnnotation,
     ) -> list[DocumentAnnotationDmlEvent]:
-        records = [dict(as_kv(ins.row)) for ins in insertions]
+        records = [{**dict(as_kv(ins.row)), "updated_at": ins.row.updated_at} for ins in insertions]
         stmt = self._insert_on_conflict(*records).returning(self.table.id)
         ids = tuple([_ async for _ in await session.stream_scalars(stmt)])
         return [DocumentAnnotationDmlEvent(ids)]
@@ -99,7 +99,7 @@ class DocumentAnnotationQueueInserter(
 
         for p in parcels:
             if (anno := existing_annos.get(_key(p))) is not None:
-                if p.received_at <= anno.updated_at:
+                if p.item.updated_at <= anno.updated_at:
                     to_discard.append(p)
                 else:
                     to_insert.append(

phoenix/db/insertion/span_annotation.py

@@ -57,7 +57,7 @@ class SpanAnnotationQueueInserter(
         session: AsyncSession,
         *insertions: Insertables.SpanAnnotation,
     ) -> list[SpanAnnotationDmlEvent]:
-        records = [dict(as_kv(ins.row)) for ins in insertions]
+        records = [{**dict(as_kv(ins.row)), "updated_at": ins.row.updated_at} for ins in insertions]
         stmt = self._insert_on_conflict(*records).returning(self.table.id)
         ids = tuple([_ async for _ in await session.stream_scalars(stmt)])
         return [SpanAnnotationDmlEvent(ids)]
@@ -92,7 +92,7 @@ class SpanAnnotationQueueInserter(
 
         for p in parcels:
             if (anno := existing_annos.get(_key(p))) is not None:
-                if p.received_at <= anno.updated_at:
+                if p.item.updated_at <= anno.updated_at:
                     to_discard.append(p)
                 else:
                     to_insert.append(

phoenix/db/insertion/trace_annotation.py

@@ -56,7 +56,7 @@ class TraceAnnotationQueueInserter(
         session: AsyncSession,
         *insertions: Insertables.TraceAnnotation,
     ) -> list[TraceAnnotationDmlEvent]:
-        records = [dict(as_kv(ins.row)) for ins in insertions]
+        records = [{**dict(as_kv(ins.row)), "updated_at": ins.row.updated_at} for ins in insertions]
         stmt = self._insert_on_conflict(*records).returning(self.table.id)
         ids = tuple([_ async for _ in await session.stream_scalars(stmt)])
         return [TraceAnnotationDmlEvent(ids)]
@@ -91,7 +91,7 @@ class TraceAnnotationQueueInserter(
 
         for p in parcels:
             if (anno := existing_annos.get(_key(p))) is not None:
-                if p.received_at <= anno.updated_at:
+                if p.item.updated_at <= anno.updated_at:
                     to_discard.append(p)
                 else:
                     to_insert.append(

phoenix/db/insertion/types.py

@@ -174,6 +174,7 @@ class QueueInserter(ABC, Generic[_PrecursorT, _InsertableT, _RowT, _DmlEventT]):
 class Precursors(ABC):
     @dataclass(frozen=True)
     class SpanAnnotation:
+        updated_at: datetime
         span_id: str
         obj: models.SpanAnnotation
 
@@ -182,6 +183,7 @@ class Precursors(ABC):
             span_rowid: int,
         ) -> Insertables.SpanAnnotation:
             return Insertables.SpanAnnotation(
+                updated_at=self.updated_at,
                 span_id=self.span_id,
                 obj=self.obj,
                 span_rowid=span_rowid,
@@ -189,6 +191,7 @@ class Precursors(ABC):
 
     @dataclass(frozen=True)
     class TraceAnnotation:
+        updated_at: datetime
         trace_id: str
         obj: models.TraceAnnotation
 
@@ -197,6 +200,7 @@ class Precursors(ABC):
             trace_rowid: int,
         ) -> Insertables.TraceAnnotation:
             return Insertables.TraceAnnotation(
+                updated_at=self.updated_at,
                 trace_id=self.trace_id,
                 obj=self.obj,
                 trace_rowid=trace_rowid,
@@ -204,6 +208,7 @@ class Precursors(ABC):
 
     @dataclass(frozen=True)
     class DocumentAnnotation:
+        updated_at: datetime
         span_id: str
         document_position: int
         obj: models.DocumentAnnotation
@@ -213,6 +218,7 @@ class Precursors(ABC):
             span_rowid: int,
         ) -> Insertables.DocumentAnnotation:
             return Insertables.DocumentAnnotation(
+                updated_at=self.updated_at,
                 span_id=self.span_id,
                 document_position=self.document_position,
                 obj=self.obj,
@@ -223,6 +229,7 @@ class Precursors(ABC):
 class Insertables(ABC):
     @dataclass(frozen=True)
     class SpanAnnotation(Precursors.SpanAnnotation):
+        updated_at: datetime
         span_rowid: int
         identifier: str = ""
 
@@ -230,10 +237,12 @@ class Insertables(ABC):
         def row(self) -> models.SpanAnnotation:
             obj = copy(self.obj)
             obj.span_rowid = self.span_rowid
+            obj.updated_at = self.updated_at
             return obj
 
     @dataclass(frozen=True)
     class TraceAnnotation(Precursors.TraceAnnotation):
+        updated_at: datetime
         trace_rowid: int
         identifier: str = ""
 
@@ -241,10 +250,12 @@ class Insertables(ABC):
         def row(self) -> models.TraceAnnotation:
             obj = copy(self.obj)
             obj.trace_rowid = self.trace_rowid
+            obj.updated_at = self.updated_at
             return obj
 
     @dataclass(frozen=True)
     class DocumentAnnotation(Precursors.DocumentAnnotation):
+        updated_at: datetime
         span_rowid: int
         identifier: str = ""
 
@@ -252,4 +263,5 @@ class Insertables(ABC):
         def row(self) -> models.DocumentAnnotation:
             obj = copy(self.obj)
             obj.span_rowid = self.span_rowid
+            obj.updated_at = self.updated_at
             return obj

phoenix/server/api/dataloaders/average_experiment_run_latency.py

@@ -23,32 +23,28 @@ class AverageExperimentRunLatencyDataLoader(DataLoader[Key, Result]):
 
     async def _load_fn(self, keys: list[Key]) -> list[Result]:
         experiment_ids = keys
-        resolved_experiment_ids = (
-            select(models.Experiment.id)
-            .where(models.Experiment.id.in_(set(experiment_ids)))
-            .subquery()
-        )
-        query = (
+        average_repetition_latency_ms = (
             select(
-                resolved_experiment_ids.c.id,
+                models.ExperimentRun.experiment_id.label("experiment_id"),
                 func.avg(
                     func.extract("epoch", models.ExperimentRun.end_time)
                     - func.extract("epoch", models.ExperimentRun.start_time)
-                ),
+                ).label("average_repetition_latency_ms"),
             )
-            .outerjoin_from(
-                from_=resolved_experiment_ids,
-                target=models.ExperimentRun,
-                onclause=resolved_experiment_ids.c.id == models.ExperimentRun.experiment_id,
-            )
-            .group_by(resolved_experiment_ids.c.id)
+            .select_from(models.ExperimentRun)
+            .where(models.ExperimentRun.experiment_id.in_(experiment_ids))
+            .group_by(models.ExperimentRun.dataset_example_id, models.ExperimentRun.experiment_id)
+            .subquery()
         )
+        query = select(
+            average_repetition_latency_ms.c.experiment_id,
+            func.avg(average_repetition_latency_ms.c.average_repetition_latency_ms).label(
+                "average_run_latency_ms"
+            ),
+        ).group_by(average_repetition_latency_ms.c.experiment_id)
         async with self._db() as session:
-            avg_latencies = {
-                experiment_id: avg_latency
-                async for experiment_id, avg_latency in await session.stream(query)
+            average_run_latencies_ms = {
+                experiment_id: average_run_latency_ms
+                async for experiment_id, average_run_latency_ms in await session.stream(query)
             }
-        return [
-            avg_latencies.get(experiment_id, ValueError(f"Unknown experiment: {experiment_id}"))
-            for experiment_id in keys
-        ]
+        return [average_run_latencies_ms.get(experiment_id) for experiment_id in keys]

phoenix/server/api/dataloaders/experiment_annotation_summaries.py

@@ -2,7 +2,7 @@ from collections import defaultdict
 from dataclasses import dataclass
 from typing import Optional
 
-from sqlalchemy import func, select
+from sqlalchemy import and_, func, select
 from strawberry.dataloader import AbstractCache, DataLoader
 from typing_extensions import TypeAlias
 
@@ -37,43 +37,97 @@ class ExperimentAnnotationSummaryDataLoader(DataLoader[Key, Result]):
     async def _load_fn(self, keys: list[Key]) -> list[Result]:
         experiment_ids = keys
         summaries: defaultdict[ExperimentID, Result] = defaultdict(list)
+        repetition_mean_scores_by_example_subquery = (
+            select(
+                models.ExperimentRun.experiment_id.label("experiment_id"),
+                models.ExperimentRunAnnotation.name.label("annotation_name"),
+                func.avg(models.ExperimentRunAnnotation.score).label("mean_repetition_score"),
+            )
+            .select_from(models.ExperimentRunAnnotation)
+            .join(
+                models.ExperimentRun,
+                models.ExperimentRunAnnotation.experiment_run_id == models.ExperimentRun.id,
+            )
+            .where(models.ExperimentRun.experiment_id.in_(experiment_ids))
+            .group_by(
+                models.ExperimentRun.dataset_example_id,
+                models.ExperimentRun.experiment_id,
+                models.ExperimentRunAnnotation.name,
+            )
+            .subquery()
+            .alias("repetition_mean_scores_by_example")
+        )
+        repetition_mean_scores_subquery = (
+            select(
+                repetition_mean_scores_by_example_subquery.c.experiment_id.label("experiment_id"),
+                repetition_mean_scores_by_example_subquery.c.annotation_name.label(
+                    "annotation_name"
+                ),
+                func.avg(repetition_mean_scores_by_example_subquery.c.mean_repetition_score).label(
+                    "mean_score"
+                ),
+            )
+            .select_from(repetition_mean_scores_by_example_subquery)
+            .group_by(
+                repetition_mean_scores_by_example_subquery.c.experiment_id,
+                repetition_mean_scores_by_example_subquery.c.annotation_name,
+            )
+            .subquery()
+            .alias("repetition_mean_scores")
+        )
+        repetitions_subquery = (
+            select(
+                models.ExperimentRun.experiment_id.label("experiment_id"),
+                models.ExperimentRunAnnotation.name.label("annotation_name"),
+                func.min(models.ExperimentRunAnnotation.score).label("min_score"),
+                func.max(models.ExperimentRunAnnotation.score).label("max_score"),
+                func.count().label("count"),
+                func.count(models.ExperimentRunAnnotation.error).label("error_count"),
+            )
+            .select_from(models.ExperimentRunAnnotation)
+            .join(
+                models.ExperimentRun,
+                models.ExperimentRunAnnotation.experiment_run_id == models.ExperimentRun.id,
+            )
+            .where(models.ExperimentRun.experiment_id.in_(experiment_ids))
+            .group_by(models.ExperimentRun.experiment_id, models.ExperimentRunAnnotation.name)
+            .subquery()
+        )
+        run_scores_query = (
+            select(
+                repetition_mean_scores_subquery.c.experiment_id.label("experiment_id"),
+                repetition_mean_scores_subquery.c.annotation_name.label("annotation_name"),
+                repetition_mean_scores_subquery.c.mean_score.label("mean_score"),
+                repetitions_subquery.c.min_score.label("min_score"),
+                repetitions_subquery.c.max_score.label("max_score"),
+                repetitions_subquery.c.count.label("count_"),
+                repetitions_subquery.c.error_count.label("error_count"),
+            )
+            .select_from(repetition_mean_scores_subquery)
+            .join(
+                repetitions_subquery,
+                and_(
+                    repetitions_subquery.c.experiment_id
+                    == repetition_mean_scores_subquery.c.experiment_id,
+                    repetitions_subquery.c.annotation_name
+                    == repetition_mean_scores_subquery.c.annotation_name,
+                ),
+            )
+            .order_by(repetition_mean_scores_subquery.c.annotation_name)
+        )
         async with self._db() as session:
-            async for (
-                experiment_id,
-                annotation_name,
-                min_score,
-                max_score,
-                mean_score,
-                count,
-                error_count,
-            ) in await session.stream(
-                select(
-                    models.ExperimentRun.experiment_id,
-                    models.ExperimentRunAnnotation.name,
-                    func.min(models.ExperimentRunAnnotation.score),
-                    func.max(models.ExperimentRunAnnotation.score),
-                    func.avg(models.ExperimentRunAnnotation.score),
-                    func.count(),
-                    func.count(models.ExperimentRunAnnotation.error),
-                )
-                .join(
-                    models.ExperimentRun,
-                    models.ExperimentRunAnnotation.experiment_run_id == models.ExperimentRun.id,
-                )
-                .where(models.ExperimentRun.experiment_id.in_(experiment_ids))
-                .group_by(models.ExperimentRun.experiment_id, models.ExperimentRunAnnotation.name)
-            ):
-                summaries[experiment_id].append(
+            async for scores_tuple in await session.stream(run_scores_query):
+                summaries[scores_tuple.experiment_id].append(
                     ExperimentAnnotationSummary(
-                        annotation_name=annotation_name,
-                        min_score=min_score,
-                        max_score=max_score,
-                        mean_score=mean_score,
-                        count=count,
-                        error_count=error_count,
+                        annotation_name=scores_tuple.annotation_name,
+                        min_score=scores_tuple.min_score,
+                        max_score=scores_tuple.max_score,
+                        mean_score=scores_tuple.mean_score,
+                        count=scores_tuple.count_,
+                        error_count=scores_tuple.error_count,
                     )
                 )
         return [
             sorted(summaries[experiment_id], key=lambda summary: summary.annotation_name)
-            for experiment_id in keys
+            for experiment_id in experiment_ids
         ]

phoenix/server/api/dataloaders/experiment_error_rates.py

@@ -1,6 +1,6 @@
 from typing import Optional
 
-from sqlalchemy import case, func, select
+from sqlalchemy import func, select
 from strawberry.dataloader import DataLoader
 from typing_extensions import TypeAlias
 
@@ -23,36 +23,29 @@ class ExperimentErrorRatesDataLoader(DataLoader[Key, Result]):
 
     async def _load_fn(self, keys: list[Key]) -> list[Result]:
         experiment_ids = keys
-        resolved_experiment_ids = (
-            select(models.Experiment.id)
-            .where(models.Experiment.id.in_(set(experiment_ids)))
-            .subquery()
-        )
-        query = (
+        average_repetition_error_rates_subquery = (
             select(
-                resolved_experiment_ids.c.id,
-                case(
-                    (
-                        func.count(models.ExperimentRun.id) != 0,
-                        func.count(models.ExperimentRun.error)
-                        / func.count(models.ExperimentRun.id),
-                    ),
-                    else_=None,
-                ),
+                models.ExperimentRun.experiment_id.label("experiment_id"),
+                (
+                    func.count(models.ExperimentRun.error) / func.count(models.ExperimentRun.id)
+                ).label("average_repetition_error_rate"),
             )
-            .outerjoin_from(
-                from_=resolved_experiment_ids,
-                target=models.ExperimentRun,
-                onclause=resolved_experiment_ids.c.id == models.ExperimentRun.experiment_id,
-            )
-            .group_by(resolved_experiment_ids.c.id)
+            .where(models.ExperimentRun.experiment_id.in_(experiment_ids))
+            .group_by(models.ExperimentRun.dataset_example_id, models.ExperimentRun.experiment_id)
+            .subquery()
+            .alias("average_repetition_error_rates")
         )
+        average_run_error_rates_query = select(
+            average_repetition_error_rates_subquery.c.experiment_id,
+            func.avg(average_repetition_error_rates_subquery.c.average_repetition_error_rate).label(
+                "average_run_error_rates"
+            ),
+        ).group_by(average_repetition_error_rates_subquery.c.experiment_id)
         async with self._db() as session:
-            error_rates = {
+            average_run_error_rates = {
                 experiment_id: error_rate
-                async for experiment_id, error_rate in await session.stream(query)
+                async for experiment_id, error_rate in await session.stream(
+                    average_run_error_rates_query
+                )
             }
-        return [
-            error_rates.get(experiment_id, ValueError(f"Unknown experiment ID: {experiment_id}"))
-            for experiment_id in keys
-        ]
+        return [average_run_error_rates.get(experiment_id) for experiment_id in experiment_ids]

phoenix/server/api/routers/v1/evaluations.py

@@ -1,5 +1,6 @@
 import gzip
 from collections.abc import Callable
+from datetime import datetime, timezone
 from itertools import chain
 from typing import Any, Iterator, Optional, Union, cast
 
@@ -269,6 +270,7 @@ def _document_annotation_factory(
     Callable[..., Precursors.DocumentAnnotation],
 ]:
     return lambda index: lambda **kwargs: Precursors.DocumentAnnotation(
+        datetime.now(timezone.utc),
         span_id=str(index[span_id_idx]),
         document_position=int(index[document_position_idx]),
         obj=models.DocumentAnnotation(
@@ -280,6 +282,7 @@ def _document_annotation_factory(
 
 def _span_annotation_factory(span_id: str) -> Callable[..., Precursors.SpanAnnotation]:
     return lambda **kwargs: Precursors.SpanAnnotation(
+        datetime.now(timezone.utc),
         span_id=str(span_id),
         obj=models.SpanAnnotation(**kwargs),
     )
@@ -287,6 +290,7 @@ def _span_annotation_factory(span_id: str) -> Callable[..., Precursors.SpanAnnot
 
 def _trace_annotation_factory(trace_id: str) -> Callable[..., Precursors.TraceAnnotation]:
     return lambda **kwargs: Precursors.TraceAnnotation(
+        datetime.now(timezone.utc),
         trace_id=str(trace_id),
         obj=models.TraceAnnotation(**kwargs),
     )

phoenix/server/api/routers/v1/spans.py

@@ -880,6 +880,7 @@ class SpanAnnotationData(V1RoutesBaseModel):
 
     def as_precursor(self, *, user_id: Optional[int] = None) -> Precursors.SpanAnnotation:
         return Precursors.SpanAnnotation(
+            datetime.now(timezone.utc),
             self.span_id,
             models.SpanAnnotation(
                 name=self.name,

phoenix/server/api/routers/v1/traces.py

@@ -1,5 +1,6 @@
 import gzip
 import zlib
+from datetime import datetime, timezone
 from typing import Any, Literal, Optional
 
 from fastapi import APIRouter, BackgroundTasks, Depends, Header, HTTPException, Path, Query
@@ -134,6 +135,7 @@ class TraceAnnotation(V1RoutesBaseModel):
 
     def as_precursor(self, *, user_id: Optional[int] = None) -> Precursors.TraceAnnotation:
         return Precursors.TraceAnnotation(
+            datetime.now(timezone.utc),
             self.trace_id,
             models.TraceAnnotation(
                 name=self.name,