arize-phoenix 3.0.2__py3-none-any.whl → 3.1.0__py3-none-any.whl

{arize_phoenix-3.0.2.dist-info → arize_phoenix-3.1.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: arize-phoenix
-Version: 3.0.2
+Version: 3.1.0
 Summary: ML Observability in your notebook
 Project-URL: Documentation, https://docs.arize.com/phoenix/
 Project-URL: Issues, https://github.com/Arize-ai/phoenix/issues
@@ -23,6 +23,7 @@ Requires-Dist: numpy
 Requires-Dist: openinference-instrumentation-langchain
 Requires-Dist: openinference-instrumentation-llama-index
 Requires-Dist: openinference-instrumentation-openai
+Requires-Dist: openinference-semantic-conventions
 Requires-Dist: opentelemetry-exporter-otlp
 Requires-Dist: opentelemetry-proto
 Requires-Dist: opentelemetry-sdk
@@ -50,7 +51,7 @@ Requires-Dist: hatch; extra == 'dev'
 Requires-Dist: jupyter; extra == 'dev'
 Requires-Dist: langchain>=0.0.334; extra == 'dev'
 Requires-Dist: litellm>=1.0.3; extra == 'dev'
-Requires-Dist: llama-index<0.10.0; extra == 'dev'
+Requires-Dist: llama-index>=0.10.3; extra == 'dev'
 Requires-Dist: nbqa; extra == 'dev'
 Requires-Dist: pandas-stubs<=2.0.2.230605; extra == 'dev'
 Requires-Dist: pre-commit; extra == 'dev'
@@ -63,7 +64,8 @@ Requires-Dist: strawberry-graphql[debug-server]==0.208.2; extra == 'dev'
 Provides-Extra: experimental
 Requires-Dist: tenacity; extra == 'experimental'
 Provides-Extra: llama-index
-Requires-Dist: llama-index==0.9.45; extra == 'llama-index'
+Requires-Dist: llama-index==0.10.3; extra == 'llama-index'
+Requires-Dist: openinference-instrumentation-llama-index>=1.0.0; extra == 'llama-index'
 Description-Content-Type: text/markdown
 
 <p align="center">

{arize_phoenix-3.0.2.dist-info → arize_phoenix-3.1.0.dist-info}/RECORD

@@ -4,14 +4,14 @@ phoenix/datetime_utils.py,sha256=D955QLrkgrrSdUM6NyqbCeAu2SMsjhR5rHVQEsVUdng,277
 phoenix/exceptions.py,sha256=X5k9ipUDfwSCwZB-H5zFJLas86Gf9tAx0W4l5TZxp5k,108
 phoenix/py.typed,sha256=AbpHGcgLb-kRsJGnwFEktk7uzpZOCcBY74-YBdrKVGs,1
 phoenix/services.py,sha256=f6AeyKTuOpy9RCcTCjVH3gx5nYZhbTMFOuv1WSUOB5o,4992
-phoenix/version.py,sha256=YKXrr5J7dV2n7ZhaRv0tigylRDtfOuvJC9Y4ouFZpzo,22
+phoenix/version.py,sha256=YVoF76lT0p3dIsqphNnDWuqSia3gZP1S1eQYXZ9FbSE,22
 phoenix/core/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 phoenix/core/embedding_dimension.py,sha256=zKGbcvwOXgLf-yrJBpQyKtd-LEOPRKHnUToyAU8Owis,87
 phoenix/core/evals.py,sha256=gJyqQzpud5YjtoY8h4pgXvHDsdubGfqmEewLuZHPPmQ,10224
 phoenix/core/model.py,sha256=vQ6RxpUPlncezJvur5u6xBN0Lkrk2gW0cTyb-qqaSqA,4713
 phoenix/core/model_schema.py,sha256=rR9VdhL_oXxbprDTPQJBXs5hw5sMPQmzx__m6Kwsxug,50394
 phoenix/core/model_schema_adapter.py,sha256=3GkyzqUST4fYi-Bgs8qAam5hwMCdQRZTDLjZ9Bnzdm4,8268
-phoenix/core/traces.py,sha256=0-CApYIFoppwKGss70_-xX_s8FPI2ABfTyT7pyeI_eE,14007
+phoenix/core/traces.py,sha256=I9xsQvEd4s7fZa1lquzBfnY7ZQoLcWwnT51wuuglnyM,14065
 phoenix/datasets/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 phoenix/datasets/dataset.py,sha256=scKVZ7zc6Dpc_ntt-pWhzY-KWqOJEwKePuyNnKSVTGE,30515
 phoenix/datasets/errors.py,sha256=cGp9vxnw4SewFoWBV3ZGMkhE0Kh73lPIv3Ppz_H_RoA,8261
@@ -23,7 +23,7 @@ phoenix/experimental/evals/__init__.py,sha256=q96YKLMt2GJD9zL8sjugvWx1INfw40Wa7E
 phoenix/experimental/evals/evaluators.py,sha256=r7fXrS-l4gn58SUhLAZSfY3P8lxysouSVJwHddrZJ_Q,15956
 phoenix/experimental/evals/retrievals.py,sha256=o3fqrsYbYZjyGj_jWkN_9VQVyXjLkDKDw5Ws7l8bwdI,3828
 phoenix/experimental/evals/functions/__init__.py,sha256=NNd0-_cmIopdV7vm3rspjfgM726qoQJ4DPq_vqbnaxQ,180
-phoenix/experimental/evals/functions/classify.py,sha256=wqsLtDVPckt4OYuyr1YzaIokqnhzfYf_bMhR5ythWww,18961
+phoenix/experimental/evals/functions/classify.py,sha256=rIQLf0GRXPNT2d5Oni0lrpFrubBorAscjN0ievjWx9A,19092
 phoenix/experimental/evals/functions/executor.py,sha256=bM7PI2rcPukQQzZ2rWqN_-Kfo_a935YJj0bh1Red8Ps,13406
 phoenix/experimental/evals/functions/generate.py,sha256=8LnnPAjBM9yxitdkaGZ67OabuDTOWBF3fvinJ_uCFRg,5584
 phoenix/experimental/evals/functions/processing.py,sha256=F4xtLsulLV4a8CkuLldRddsCim75dSTIShEJUYN6I6w,1823
@@ -31,15 +31,15 @@ phoenix/experimental/evals/models/__init__.py,sha256=j1N7DhiOPbcaemtVBONcQ0miNnG
 phoenix/experimental/evals/models/anthropic.py,sha256=BZmLvepkSMj_opCWsZoL34a3yAwRdl7qbJB86DFR84E,6688
 phoenix/experimental/evals/models/base.py,sha256=RWz_Jzj3Z1fENl2WUXIz-4eMsk6HfYXc0K8IZ-BJss4,6306
 phoenix/experimental/evals/models/bedrock.py,sha256=nVOXRZr-iDwHEINozpO2bqZR2KEeDHNyj6jgQPONQYs,8565
-phoenix/experimental/evals/models/litellm.py,sha256=0c-eJFsx41W0MsqeUd4UPquLBKSZp3BRNhKhX2uFCAs,4123
-phoenix/experimental/evals/models/openai.py,sha256=R5DXvIVcASLsoOGQBOevOfpUx4j0dOIVPS3voE8CuuY,17367
+phoenix/experimental/evals/models/litellm.py,sha256=Xo415fJehxIj32zpzVtvzQXADJURPUvNZJHmc_FAKvE,4759
+phoenix/experimental/evals/models/openai.py,sha256=v2qkdFZc-u0ZAfxnV4v5UMWkMc9P3k4Gx1XaTnVFTz4,17922
 phoenix/experimental/evals/models/rate_limiters.py,sha256=5GVN0RQKt36Przg3-9jLgocRmyg-tbeO-cdbuLIx89w,10160
 phoenix/experimental/evals/models/vertex.py,sha256=1VAGJNoiUm56pP8G9Qvnf-4_Rl9u9NI7ToOKbWFNtpk,6226
-phoenix/experimental/evals/models/vertexai.py,sha256=_txsOP2RHyR3AnugeJRFUNvYm3xXvfMbWpULxTko4OA,4821
+phoenix/experimental/evals/models/vertexai.py,sha256=nFN5isv1GPNsvUA4OxSnQd1hTMi-3BcxfDz1y7QcoA0,6189
 phoenix/experimental/evals/templates/__init__.py,sha256=GSJSoWJ4jwyoUANniidmWMUtXQhNQYbTJbfFqCvuYuo,1470
 phoenix/experimental/evals/templates/default_templates.py,sha256=dVKmoLwqgAyGcRuezz9WKnXSHhw7-qk1R8j6wSmqh0s,20722
 phoenix/experimental/evals/templates/template.py,sha256=ImFSaTPo9oalPNwq7cNdOCndrvuwLuIyIFKsgDVcoJE,6715
-phoenix/experimental/evals/utils/__init__.py,sha256=608EX7sG0f5oDG__II16J8xnFJiNpY9dI9AC8vXwR00,5601
+phoenix/experimental/evals/utils/__init__.py,sha256=FE1tbb022t3pwn0CrprR7QXcIsgpsdsotV7-iB3khsI,5601
 phoenix/experimental/evals/utils/threads.py,sha256=ksI-egarPnlxit0qKKjtjZ2L82qGLxqxZ6s92O0eBA4,1005
 phoenix/metrics/README.md,sha256=5gekqTU-5gGdMwvcfNp2Wlu8p1ul9kGY_jq0XXQusoI,1964
 phoenix/metrics/__init__.py,sha256=sLp7td1GIt_0Z8dPUyP4L0-_4x9c871yAaGX30oMsvg,2433
@@ -70,7 +70,7 @@ phoenix/server/api/input_types/DimensionFilter.py,sha256=vcXgglSnZcB5pGh-6oEtRmG
 phoenix/server/api/input_types/DimensionInput.py,sha256=Vfx5FmiMKey4-EHDQsQRPzSAMRJMN5oVMLDUl4NKAa8,164
 phoenix/server/api/input_types/Granularity.py,sha256=6SVfZ5yTZYq1PI6vdpjfkBUc4YilLSkF-k6okuSNbbQ,2301
 phoenix/server/api/input_types/PerformanceMetricInput.py,sha256=fElsLTSEYYgGFGMYTEGcYid39tXUKFdV_JkdHavMcbA,591
-phoenix/server/api/input_types/SpanSort.py,sha256=9OxxFseY-9pwZ9rUJ71P-h_vpwYg5UAiy76x_pzHvPo,3537
+phoenix/server/api/input_types/SpanSort.py,sha256=72oWhYLLIZLy7K5_tFhMg7RcmDgatqMB-ZoRJhHGIR0,3527
 phoenix/server/api/input_types/TimeRange.py,sha256=yzx-gxj8mDeGLft1FzU_x1MVEgIG5Pt6-f8PUVDgipQ,522
 phoenix/server/api/input_types/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 phoenix/server/api/routers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -109,7 +109,7 @@ phoenix/server/api/types/Retrieval.py,sha256=OhMK2ncjoyp5h1yjKhjlKpoTbQrMHuxmgSF
 phoenix/server/api/types/ScalarDriftMetricEnum.py,sha256=IUAcRPpgL41WdoIgK6cNk2Te38SspXGyEs-S1fY23_A,232
 phoenix/server/api/types/Segments.py,sha256=zogJI9MdmctBL7J-fDSR_8tUJLvuISlVYgCLnTaigKE,2937
 phoenix/server/api/types/SortDir.py,sha256=OUpXhlCzCxPoXSDkJJygEs9Rw9pMymfaZUG5zPTrw4Y,152
-phoenix/server/api/types/Span.py,sha256=QWWW2MmWD0bwiR-jAUtJJZ-GvZJhTFYq7qhus7tsz3A,11304
+phoenix/server/api/types/Span.py,sha256=OvY4ffZa3KMnT3J9s0oLB0ghurYSJCWlK0YfoKkg-6s,11633
 phoenix/server/api/types/TimeSeries.py,sha256=QbLfxHnwYsMsirpq4tx9us6ha7YtAVzK4m8mAL3fMt0,5200
 phoenix/server/api/types/UMAPPoints.py,sha256=8l9RJXi308qty4MdHb2pBbiU6ZuLbrRRxXNbPhXoxKI,1639
 phoenix/server/api/types/ValidationResult.py,sha256=pHwdYk4J7SJ5xhlWWHg_6qWkfk4rjOx-bSkGHvkDE3Q,142
@@ -135,32 +135,30 @@ phoenix/session/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 phoenix/session/client.py,sha256=lTV9Q4ljpV9wqasya4u3x7BCijrfAWKDm5akzrg0424,4885
 phoenix/session/data_extractor.py,sha256=V4ntT2FcmbV_-zobcbPR51HKWaDyMnXHj4Pmu7MZ0OQ,1664
 phoenix/session/evaluation.py,sha256=DaAtA0XYJbXRJO_StGywa-9APlz2ORSmCXzxrtn3rvI,4997
-phoenix/session/session.py,sha256=mrTAQkUFMGjN65ApYqRjhGXExbfxu2ODJcJ14MSDOLE,21719
+phoenix/session/session.py,sha256=R83kPeiqLxEJTD6cTjJqqMjMV-iQzQFKw6rMXnbIxnc,21800
 phoenix/trace/__init__.py,sha256=4d_MqzUIFmlY9WWcFeTONJ4xL5mPGoWZaPM2TJ0ZDBQ,266
-phoenix/trace/errors.py,sha256=DbXSJnNErV7305tKv7pUWLD6jcVHJ6EBdSu4mZJ6IM4,112
+phoenix/trace/errors.py,sha256=wB1z8qdPckngdfU-TORToekvg3344oNFAA83_hC2yFY,180
 phoenix/trace/evaluation_conventions.py,sha256=t8jydM3U0-T5YpiQKRJ3tWdWGlHtzKyttYdw-ddvPOk,1048
 phoenix/trace/exporter.py,sha256=O-9Arn-S_B9Me-jy4Qa84y6lvxKNoa8pczrsamPl3_Q,4871
-phoenix/trace/fixtures.py,sha256=LokNedhbGYxpzXznteO4m5QehvNYjzvoh231-CMJQeY,7113
-phoenix/trace/otel.py,sha256=lJoTvkCSFiPT4dFRLMZZgZfsOoTSATctW9lQk9J30ZI,14821
-phoenix/trace/schemas.py,sha256=QDBlM94faceTnloY_ZVo2BHWWnE5UGymefC0jmfj4ew,6011
-phoenix/trace/semantic_conventions.py,sha256=u6NG85ZhbreriZr8cqJaddldM_jUcew7JilszY7JUk8,4652
+phoenix/trace/fixtures.py,sha256=HOjuYOB_xtR7JUeLz7WpEroiGj4E5_SxVLSjBYUy8RQ,7055
+phoenix/trace/otel.py,sha256=naFWTwrqyIoeGJ5YMT9Fyhd44CEtQvpscVluDOXD6Fo,16744
+phoenix/trace/schemas.py,sha256=zlAY8v-PLgEDqTgbmDxY3NJlCAmzj_3ZCJFebQOBz3M,6028
 phoenix/trace/span_evaluations.py,sha256=FvY9YRnKuYIzCa-H9P5SuDaI2DeqGnVCjNgclC2v3HA,12909
-phoenix/trace/span_json_decoder.py,sha256=nrIPkcgbCcNML-0OSjWC6fxIfBEMiP0n67yM_m-vegg,3068
-phoenix/trace/span_json_encoder.py,sha256=C5y7rkyOcV08oJC5t8TZqVxsKCZMJKad7bBQzAgLoDs,1763
-phoenix/trace/trace_dataset.py,sha256=KW0TzmhlKuX8PUPLV172iTK08myYE0QXUC75KiIqJ7k,13204
+phoenix/trace/span_json_decoder.py,sha256=IAFakPRqSMYxTPKYFMiXYxm7U-FipdN8_xbvapDS0Qc,3131
+phoenix/trace/span_json_encoder.py,sha256=hIDd1I6xm01kaNmeKjHOHyxUGI3uTg5J_Os1kXtAb6g,1755
+phoenix/trace/trace_dataset.py,sha256=RpHIfZLbMmULOIb-fKXJkQLhIdC0sJlAOTjlyJppMYA,13776
 phoenix/trace/tracer.py,sha256=JDKlyvjy6AsQmaA60ycJ1hKXoUQU61jqPx3nvYr8xUc,3647
 phoenix/trace/utils.py,sha256=7LurVGXn245cjj4MJsc7v6jq4DSJkpK6YGBfIaSywuw,1307
 phoenix/trace/dsl/__init__.py,sha256=WIQIjJg362XD3s50OsPJJ0xbDsGp41bSv7vDllLrPuA,144
-phoenix/trace/dsl/filter.py,sha256=VwlzL4KsIoyYjjCSKXlOSyVm49WyYRlkK5zedXqZlcc,13136
-phoenix/trace/dsl/helpers.py,sha256=qxRS65NFUFVpLDd8IZVrZUH891MamLeXyfIIN1BJJ_c,1552
+phoenix/trace/dsl/filter.py,sha256=fRPUIsucxXGJcbtei86ApSIjEP5PmdsONmYvT7EBu6w,14240
+phoenix/trace/dsl/helpers.py,sha256=U71HmCecJQ_zHq0g2mFKbxrRuiJDkopskWMpFoBNP-Y,1722
 phoenix/trace/dsl/missing.py,sha256=BWPOHr2_tBkPDgVeq8GVXXVbNbJiBelu4NtwHBg6mTE,1435
-phoenix/trace/dsl/query.py,sha256=W6_e1o_JbucGWWv6pSUQFyIHcTuj7QuCBSuKXodRfCE,14773
+phoenix/trace/dsl/query.py,sha256=XoFwKEALzGqUERy7B5fgD-n0s87zN6jRVrZgW6-jqRo,14819
 phoenix/trace/langchain/__init__.py,sha256=aTKMFmEOgjx_6dnyplalgYi7PQnetablwrwpcUZqcGE,764
 phoenix/trace/langchain/instrumentor.py,sha256=rmqdaEh2rwyZnaddns4RC2WQnj7TbCnJjETgoiwWvzs,1163
 phoenix/trace/langchain/tracer.py,sha256=fSjtPc5VxaZK63x3ob3DAW7eM2kfOZWgbSQ9dmhhtsw,1424
-phoenix/trace/llama_index/__init__.py,sha256=wCcQgD9CG5TA8i-1XsSed4ZzwHTUmqZwegQAV_FqEng,178
-phoenix/trace/llama_index/callback.py,sha256=MHgUNJSG6ecxZ8SZ_9LdcKnAAWmNI-U5pySFyXqubjM,1694
-phoenix/trace/llama_index/debug_callback.py,sha256=SKToD9q_QADSGTJ5lhilqRVKaUnUSRXUvURCzN4by2U,1367
+phoenix/trace/llama_index/__init__.py,sha256=4fpR5702Qh2t5TaXIx584EkA-BveCPftXPOKvI0Oi3I,105
+phoenix/trace/llama_index/callback.py,sha256=YKCZZtFwAwasZt9qNSJO6dj97Jt2eN_PBakUYWrCleE,3574
 phoenix/trace/openai/__init__.py,sha256=J3G0uqCxGdksUpaQVHds_Egv2drvh8UEqoLjiQAOveg,79
 phoenix/trace/openai/instrumentor.py,sha256=Lre96YfPxs5iM3frZfwdAh94LUpyYw_RZpSF0XmPtKw,1061
 phoenix/trace/v1/__init__.py,sha256=-IbAD0ruESMjvQLvGAg9CTfjBUATFDx1OXseDPis6-0,88
@@ -169,8 +167,8 @@ phoenix/trace/v1/evaluation_pb2.pyi,sha256=cCbbx06gwQmaH14s3J1X25TtaARh-k1abbxQd
 phoenix/utilities/__init__.py,sha256=3TVirVnjIGyaCFuJCqeZO4tjlzQ_chZgYM0itIwsEpE,656
 phoenix/utilities/error_handling.py,sha256=7b5rpGFj9EWZ8yrZK1IHvxB89suWk3lggDayUQcvZds,1946
 phoenix/utilities/logging.py,sha256=lDXd6EGaamBNcQxL4vP1au9-i_SXe0OraUDiJOcszSw,222
-arize_phoenix-3.0.2.dist-info/METADATA,sha256=80fEGOl8VUcS4B1STEVmDITkrkYytJtE6f9d7VmkyhI,28555
-arize_phoenix-3.0.2.dist-info/WHEEL,sha256=TJPnKdtrSue7xZ_AVGkp9YXcvDrobsjBds1du3Nx6dc,87
-arize_phoenix-3.0.2.dist-info/licenses/IP_NOTICE,sha256=JBqyyCYYxGDfzQ0TtsQgjts41IJoa-hiwDrBjCb9gHM,469
-arize_phoenix-3.0.2.dist-info/licenses/LICENSE,sha256=HFkW9REuMOkvKRACuwLPT0hRydHb3zNg-fdFt94td18,3794
-arize_phoenix-3.0.2.dist-info/RECORD,,
+arize_phoenix-3.1.0.dist-info/METADATA,sha256=REf61_QdSCtku8jKwT3K_Pdaseow4atJjtb7htsYWno,28694
+arize_phoenix-3.1.0.dist-info/WHEEL,sha256=TJPnKdtrSue7xZ_AVGkp9YXcvDrobsjBds1du3Nx6dc,87
+arize_phoenix-3.1.0.dist-info/licenses/IP_NOTICE,sha256=JBqyyCYYxGDfzQ0TtsQgjts41IJoa-hiwDrBjCb9gHM,469
+arize_phoenix-3.1.0.dist-info/licenses/LICENSE,sha256=HFkW9REuMOkvKRACuwLPT0hRydHb3zNg-fdFt94td18,3794
+arize_phoenix-3.1.0.dist-info/RECORD,,

phoenix/core/traces.py

@@ -20,12 +20,13 @@ from typing import (
 
 import opentelemetry.proto.trace.v1.trace_pb2 as otlp
 from ddsketch import DDSketch
+from openinference.semconv.trace import SpanAttributes
 from sortedcontainers import SortedKeyList
 from typing_extensions import TypeAlias
 from wrapt import ObjectProxy
 
+import phoenix.trace.schemas
 from phoenix.datetime_utils import right_open_time_range
-from phoenix.trace import semantic_conventions
 from phoenix.trace.otel import decode
 from phoenix.trace.schemas import (
     ATTRIBUTE_PREFIX,
@@ -33,12 +34,10 @@ from phoenix.trace.schemas import (
     CONTEXT_PREFIX,
     ComputedAttributes,
     Span,
-    SpanAttributes,
     SpanID,
     SpanStatusCode,
     TraceID,
 )
-from phoenix.trace.semantic_conventions import RETRIEVAL_DOCUMENTS
 
 END_OF_QUEUE = None  # sentinel value for queue termination
 
@@ -50,9 +49,9 @@ SPAN_ID = CONTEXT_PREFIX + "span_id"
 PARENT_ID = "parent_id"
 START_TIME = "start_time"
 END_TIME = "end_time"
-LLM_TOKEN_COUNT_TOTAL = ATTRIBUTE_PREFIX + semantic_conventions.LLM_TOKEN_COUNT_TOTAL
-LLM_TOKEN_COUNT_PROMPT = ATTRIBUTE_PREFIX + semantic_conventions.LLM_TOKEN_COUNT_PROMPT
-LLM_TOKEN_COUNT_COMPLETION = ATTRIBUTE_PREFIX + semantic_conventions.LLM_TOKEN_COUNT_COMPLETION
+LLM_TOKEN_COUNT_TOTAL = ATTRIBUTE_PREFIX + SpanAttributes.LLM_TOKEN_COUNT_TOTAL
+LLM_TOKEN_COUNT_PROMPT = ATTRIBUTE_PREFIX + SpanAttributes.LLM_TOKEN_COUNT_PROMPT
+LLM_TOKEN_COUNT_COMPLETION = ATTRIBUTE_PREFIX + SpanAttributes.LLM_TOKEN_COUNT_COMPLETION
 
 
 class ReadableSpan(ObjectProxy):  # type: ignore
@@ -73,7 +72,9 @@ class ReadableSpan(ObjectProxy):  # type: ignore
     @property
     def span(self) -> Span:
         span = decode(self._self_otlp_span)
-        span.attributes.update(cast(SpanAttributes, self._self_computed_values))
+        span.attributes.update(
+            cast(phoenix.trace.schemas.SpanAttributes, self._self_computed_values)
+        )
         # TODO: compute latency rank percent (which can change depending on how
         # many spans already ingested).
         return span
@@ -333,9 +334,13 @@ class Traces:
             self._token_count_total -= existing_span[LLM_TOKEN_COUNT_TOTAL] or 0
         self._token_count_total += new_span[LLM_TOKEN_COUNT_TOTAL] or 0
         # Update number of documents
-        num_documents_update = len(new_span.attributes.get(RETRIEVAL_DOCUMENTS) or ())
+        num_documents_update = len(
+            new_span.attributes.get(SpanAttributes.RETRIEVAL_DOCUMENTS) or ()
+        )
         if existing_span:
-            num_documents_update -= len(existing_span.attributes.get(RETRIEVAL_DOCUMENTS) or ())
+            num_documents_update -= len(
+                existing_span.attributes.get(SpanAttributes.RETRIEVAL_DOCUMENTS) or ()
+            )
         if num_documents_update:
             self._num_documents[span_id] += num_documents_update
         # Process previously orphaned spans, if any.

phoenix/experimental/evals/functions/classify.py

@@ -19,6 +19,7 @@ from typing import (
 )
 
 import pandas as pd
+from openinference.semconv.trace import DocumentAttributes, SpanAttributes
 from pandas import DataFrame
 from typing_extensions import TypeAlias
 
@@ -41,9 +42,12 @@ from phoenix.experimental.evals.utils import (
     parse_openai_function_call,
     snap_to_rail,
 )
-from phoenix.trace.semantic_conventions import DOCUMENT_CONTENT, INPUT_VALUE, RETRIEVAL_DOCUMENTS
 from phoenix.utilities.logging import printif
 
+DOCUMENT_CONTENT = DocumentAttributes.DOCUMENT_CONTENT
+INPUT_VALUE = SpanAttributes.INPUT_VALUE
+RETRIEVAL_DOCUMENTS = SpanAttributes.RETRIEVAL_DOCUMENTS
+
 logger = logging.getLogger(__name__)
 
 

phoenix/experimental/evals/models/litellm.py

@@ -1,4 +1,5 @@
 import logging
+import warnings
 from dataclasses import dataclass, field
 from typing import TYPE_CHECKING, Any, Dict, List, Optional
 
@@ -12,7 +13,7 @@ logger = logging.getLogger(__name__)
 
 @dataclass
 class LiteLLMModel(BaseEvalModel):
-    model_name: str = "gpt-3.5-turbo"
+    model: str = "gpt-3.5-turbo"
     """The model name to use."""
     temperature: float = 0.0
     """What sampling temperature to use."""
@@ -34,22 +35,42 @@ class LiteLLMModel(BaseEvalModel):
     max_content_size: Optional[int] = None
     """If you're using a fine-tuned model, set this to the maximum content size"""
 
+    # Deprecated fields
+    model_name: Optional[str] = None
+    """
+    .. deprecated:: 3.0.0
+       use `model` instead. This will be removed in a future release.
+    """
+
     def __post_init__(self) -> None:
+        self._migrate_model_name()
         self._init_environment()
         self._init_model_encoding()
 
+    def _migrate_model_name(self) -> None:
+        if self.model_name is not None:
+            warning_message = "The `model_name` field is deprecated. Use `model` instead. \
+                This will be removed in a future release."
+            warnings.warn(
+                warning_message,
+                DeprecationWarning,
+            )
+            print(warning_message)
+            self.model = self.model_name
+            self.model_name = None
+
     def _init_environment(self) -> None:
         try:
             import litellm
             from litellm import validate_environment
 
             self._litellm = litellm
-            env_info = validate_environment(self._litellm.utils.get_llm_provider(self.model_name))
+            env_info = validate_environment(self._litellm.utils.get_llm_provider(self.model))
 
             if not env_info["keys_in_environment"]:
                 raise RuntimeError(
                     f"Missing environment variable(s): '{str(env_info['missing_keys'])}', for "
-                    f"model: {self.model_name}. \nFor additional information about the right "
+                    f"model: {self.model}. \nFor additional information about the right "
                     "environment variables for specific model providers:\n"
                     "https://docs.litellm.ai/docs/completion/input#provider-specific-params."
                 )
@@ -67,14 +88,14 @@ class LiteLLMModel(BaseEvalModel):
 
     @property
     def max_context_size(self) -> int:
-        context_size = self.max_content_size or self._litellm.get_max_tokens(self.model_name).get(
+        context_size = self.max_content_size or self._litellm.get_max_tokens(self.model).get(
             "max_tokens", None
         )
 
         if context_size is None:
             raise ValueError(
-                "Can't determine maximum context size. An unknown model name was "
-                + f"used: {self.model_name}."
+                "Can't determine maximum context size. An unknown model was "
+                + f"used: {self.model}."
             )
 
         return context_size
@@ -84,11 +105,11 @@ class LiteLLMModel(BaseEvalModel):
         raise NotImplementedError
 
     def get_tokens_from_text(self, text: str) -> List[int]:
-        result: List[int] = self._encoding(model=self.model_name, text=text)
+        result: List[int] = self._encoding(model=self.model, text=text)
         return result
 
     def get_text_from_tokens(self, tokens: List[int]) -> str:
-        return str(self._decoding(model=self.model_name, tokens=tokens))
+        return str(self._decoding(model=self.model, tokens=tokens))
 
     async def _async_generate(self, prompt: str, **kwargs: Dict[str, Any]) -> str:
         return self._generate(prompt, **kwargs)
@@ -96,7 +117,7 @@ class LiteLLMModel(BaseEvalModel):
     def _generate(self, prompt: str, **kwargs: Dict[str, Any]) -> str:
         messages = self._get_messages_from_prompt(prompt)
         response = self._litellm.completion(
-            model=self.model_name,
+            model=self.model,
             messages=messages,
             temperature=self.temperature,
             max_tokens=self.max_tokens,

phoenix/experimental/evals/models/openai.py

@@ -1,5 +1,6 @@
 import logging
 import os
+import warnings
 from dataclasses import dataclass, field, fields
 from typing import (
     TYPE_CHECKING,
@@ -64,8 +65,10 @@ class OpenAIModel(BaseEvalModel):
     An optional base URL to use for the OpenAI API. If not provided, will default
     to what's configured in OpenAI
     """
-    model_name: str = "gpt-4"
-    """Model name to use. In of azure, this is the deployment name such as gpt-35-instant"""
+    model: str = "gpt-4"
+    """
+    Model name to use. In of azure, this is the deployment name such as gpt-35-instant
+    """
     temperature: float = 0.0
     """What sampling temperature to use."""
     max_tokens: int = 256
@@ -106,7 +109,15 @@ class OpenAIModel(BaseEvalModel):
     azure_ad_token: Optional[str] = field(default=None)
     azure_ad_token_provider: Optional[Callable[[], str]] = field(default=None)
 
+    # Deprecated fields
+    model_name: Optional[str] = field(default=None)
+    """
+    .. deprecated:: 3.0.0
+       use `model` instead. This will be removed
+    """
+
     def __post_init__(self) -> None:
+        self._migrate_model_name()
         self._init_environment()
         self._init_open_ai()
         self._init_tiktoken()
@@ -115,6 +126,17 @@ class OpenAIModel(BaseEvalModel):
     def reload_client(self) -> None:
         self._init_open_ai()
 
+    def _migrate_model_name(self) -> None:
+        if self.model_name:
+            warning_message = "The `model_name` field is deprecated. Use `model` instead. \
+                This will be removed in a future release."
+            print(
+                warning_message,
+            )
+            warnings.warn(warning_message, DeprecationWarning)
+            self.model = self.model_name
+            self.model_name = None
+
     def _init_environment(self) -> None:
         try:
             import openai
@@ -141,9 +163,7 @@ class OpenAIModel(BaseEvalModel):
         # For Azure, you need to provide the endpoint and the endpoint
         self._is_azure = bool(self.azure_endpoint)
 
-        self._model_uses_legacy_completion_api = self.model_name.startswith(
-            LEGACY_COMPLETION_API_MODELS
-        )
+        self._model_uses_legacy_completion_api = self.model.startswith(LEGACY_COMPLETION_API_MODELS)
         if self.api_key is None:
             api_key = os.getenv(OPENAI_API_KEY_ENVVAR_NAME)
             if api_key is None:
@@ -203,7 +223,7 @@ class OpenAIModel(BaseEvalModel):
 
     def _init_tiktoken(self) -> None:
         try:
-            encoding = self._tiktoken.encoding_for_model(self.model_name)
+            encoding = self._tiktoken.encoding_for_model(self.model)
         except KeyError:
             encoding = self._tiktoken.get_encoding("cl100k_base")
         self._tiktoken_encoding = encoding
@@ -333,20 +353,20 @@ class OpenAIModel(BaseEvalModel):
 
     @property
     def max_context_size(self) -> int:
-        model_name = self.model_name
+        model = self.model
         # handling finetuned models
-        if "ft-" in model_name:
-            model_name = self.model_name.split(":")[0]
-        if model_name == "gpt-4":
+        if "ft-" in model:
+            model = self.model.split(":")[0]
+        if model == "gpt-4":
             # Map gpt-4 to the current default
-            model_name = "gpt-4-0613"
+            model = "gpt-4-0613"
 
-        context_size = MODEL_TOKEN_LIMIT_MAPPING.get(model_name, None)
+        context_size = MODEL_TOKEN_LIMIT_MAPPING.get(model, None)
 
         if context_size is None:
             raise ValueError(
                 "Can't determine maximum context size. An unknown model name was "
-                f"used: {model_name}. Please provide a valid OpenAI model name. "
+                f"used: {model}. Please provide a valid OpenAI model name. "
                 "Known models are: " + ", ".join(MODEL_TOKEN_LIMIT_MAPPING.keys())
             )
 
@@ -355,7 +375,7 @@ class OpenAIModel(BaseEvalModel):
     @property
     def public_invocation_params(self) -> Dict[str, Any]:
         return {
-            **({"model": self.model_name}),
+            **({"model": self.model}),
             **self._default_params,
             **self.model_kwargs,
         }
@@ -388,8 +408,8 @@ class OpenAIModel(BaseEvalModel):
 
         Official documentation: https://github.com/openai/openai-cookbook/blob/main/examples/How_to_format_inputs_to_ChatGPT_models.ipynb
         """  # noqa
-        model_name = self.model_name
-        if model_name == "gpt-3.5-turbo-0301":
+        model = self.model
+        if model == "gpt-3.5-turbo-0301":
             tokens_per_message = 4  # every message follows <|start|>{role/name}\n{content}<|end|>\n
             tokens_per_name = -1  # if there's a name, the role is omitted
         else:

phoenix/experimental/evals/models/vertexai.py

@@ -1,3 +1,5 @@
+import logging
+import warnings
 from dataclasses import dataclass
 from typing import TYPE_CHECKING, Any, Dict, List, Optional
 
@@ -6,6 +8,7 @@ from phoenix.experimental.evals.models.base import BaseEvalModel
 if TYPE_CHECKING:
     from google.auth.credentials import Credentials  # type:ignore
 
+logger = logging.getLogger(__name__)
 
 MINIMUM_VERTEX_AI_VERSION = "1.33.0"
 
@@ -18,9 +21,9 @@ class VertexAIModel(BaseEvalModel):
     "location (str): The default location to use when making API calls. If not "
     "set defaults to us-central-1."
     credentials: Optional["Credentials"] = None
-    model_name: str = "text-bison"
-    tuned_model_name: Optional[str] = None
-    "The name of a tuned model. If provided, model_name is ignored."
+    model: str = "text-bison"
+    tuned_model: Optional[str] = None
+    "The name of a tuned model. If provided, model is ignored."
     max_retries: int = 6
     """Maximum number of retries to make when generating."""
     retry_min_seconds: int = 10
@@ -40,11 +43,50 @@ class VertexAIModel(BaseEvalModel):
     "How the model selects tokens for output, the next token is selected from "
     "among the top-k most probable tokens. Top-k is ignored for Codey models."
 
+    # Deprecated fields
+    model_name: Optional[str] = None
+    """
+    .. deprecated:: 3.0.0
+       use `model` instead. This will be removed in a future release.
+    """
+    tuned_model_name: Optional[str] = None
+    """
+    .. deprecated:: 3.0.0
+       use `tuned_model` instead. This will be removed in a future release.
+    """
+
     def __post_init__(self) -> None:
+        self._migrate_model_name()
         self._init_environment()
         self._init_vertex_ai()
         self._instantiate_model()
 
+    def _migrate_model_name(self) -> None:
+        if self.model_name is not None:
+            warning_message = (
+                "The `model_name` field is deprecated. Use `model` instead. "
+                + "This will be removed in a future release."
+            )
+            warnings.warn(
+                warning_message,
+                DeprecationWarning,
+            )
+            print(warning_message)
+            self.model = self.model_name
+            self.model_name = None
+        if self.tuned_model_name is not None:
+            warning_message = (
+                "`tuned_model_name` field is deprecated. Use `tuned_model` instead. "
+                + "This will be removed in a future release."
+            )
+            warnings.warn(
+                warning_message,
+                DeprecationWarning,
+            )
+            print(warning_message)
+            self.tuned_model = self.tuned_model_name
+            self.tuned_model_name = None
+
     def _init_environment(self) -> None:
         try:
             import google.api_core.exceptions as google_exceptions  # type:ignore
@@ -72,10 +114,10 @@ class VertexAIModel(BaseEvalModel):
 
             model = TextGenerationModel
 
-        if self.tuned_model_name:
-            self._model = model.get_tuned_model(self.tuned_model_name)
+        if self.tuned_model:
+            self._model = model.get_tuned_model(self.tuned_model)
         else:
-            self._model = model.from_pretrained(self.model_name)
+            self._model = model.from_pretrained(self.model)
 
     def verbose_generation_info(self) -> str:
         return f"VertexAI invocation parameters: {self.invocation_params}"
@@ -93,7 +135,7 @@ class VertexAIModel(BaseEvalModel):
 
     @property
     def is_codey_model(self) -> bool:
-        return is_codey_model(self.tuned_model_name or self.model_name)
+        return is_codey_model(self.tuned_model or self.model)
 
     @property
     def _init_params(self) -> Dict[str, Any]:

phoenix/experimental/evals/utils/__init__.py

@@ -32,7 +32,7 @@ def download_benchmark_dataset(task: str, dataset_name: str) -> pd.DataFrame:
         pandas.DataFrame: A pandas dataframe containing the data.
     """
     jsonl_file_name = f"{dataset_name}.jsonl"
-    url = f"http://storage.googleapis.com/arize-assets/phoenix/evals/{task}/{jsonl_file_name}.zip"
+    url = f"http://storage.googleapis.com/arize-phoenix-assets/evals/{task}/{jsonl_file_name}.zip"
     try:
         with urlopen(url) as response:
             zip_byte_stream = BytesIO(response.read())

phoenix/server/api/input_types/SpanSort.py

@@ -4,6 +4,7 @@ from typing import Any, Iterable, Iterator, Optional, Protocol
 
 import pandas as pd
 import strawberry
+from openinference.semconv.trace import SpanAttributes
 from strawberry import UNSET
 from typing_extensions import assert_never
 
@@ -13,7 +14,6 @@ from phoenix.core.traces import (
     START_TIME,
 )
 from phoenix.server.api.types.SortDir import SortDir
-from phoenix.trace import semantic_conventions
 from phoenix.trace.schemas import ComputedAttributes, Span, SpanID
 
 
@@ -22,9 +22,9 @@ class SpanColumn(Enum):
     startTime = START_TIME
     endTime = END_TIME
     latencyMs = ComputedAttributes.LATENCY_MS.value
-    tokenCountTotal = semantic_conventions.LLM_TOKEN_COUNT_TOTAL
-    tokenCountPrompt = semantic_conventions.LLM_TOKEN_COUNT_PROMPT
-    tokenCountCompletion = semantic_conventions.LLM_TOKEN_COUNT_COMPLETION
+    tokenCountTotal = SpanAttributes.LLM_TOKEN_COUNT_TOTAL
+    tokenCountPrompt = SpanAttributes.LLM_TOKEN_COUNT_PROMPT
+    tokenCountCompletion = SpanAttributes.LLM_TOKEN_COUNT_COMPLETION
     cumulativeTokenCountTotal = ComputedAttributes.CUMULATIVE_LLM_TOKEN_COUNT_TOTAL.value
     cumulativeTokenCountPrompt = ComputedAttributes.CUMULATIVE_LLM_TOKEN_COUNT_PROMPT.value
     cumulativeTokenCountCompletion = ComputedAttributes.CUMULATIVE_LLM_TOKEN_COUNT_COMPLETION.value

phoenix/server/api/types/Span.py

@@ -5,6 +5,7 @@ from enum import Enum
 from typing import Any, DefaultDict, Dict, List, Mapping, Optional, Sized, cast
 
 import strawberry
+from openinference.semconv.trace import EmbeddingAttributes, SpanAttributes
 from strawberry import ID, UNSET
 from strawberry.types import Info
 
@@ -15,19 +16,17 @@ from phoenix.server.api.types.DocumentRetrievalMetrics import DocumentRetrievalM
 from phoenix.server.api.types.Evaluation import DocumentEvaluation, SpanEvaluation
 from phoenix.server.api.types.MimeType import MimeType
 from phoenix.trace.schemas import ComputedAttributes, SpanID
-from phoenix.trace.semantic_conventions import (
-    EMBEDDING_EMBEDDINGS,
-    EMBEDDING_VECTOR,
-    EXCEPTION_MESSAGE,
-    INPUT_MIME_TYPE,
-    INPUT_VALUE,
-    LLM_TOKEN_COUNT_COMPLETION,
-    LLM_TOKEN_COUNT_PROMPT,
-    LLM_TOKEN_COUNT_TOTAL,
-    OUTPUT_MIME_TYPE,
-    OUTPUT_VALUE,
-    RETRIEVAL_DOCUMENTS,
-)
+
+EMBEDDING_EMBEDDINGS = SpanAttributes.EMBEDDING_EMBEDDINGS
+EMBEDDING_VECTOR = EmbeddingAttributes.EMBEDDING_VECTOR
+INPUT_MIME_TYPE = SpanAttributes.INPUT_MIME_TYPE
+INPUT_VALUE = SpanAttributes.INPUT_VALUE
+LLM_TOKEN_COUNT_COMPLETION = SpanAttributes.LLM_TOKEN_COUNT_COMPLETION
+LLM_TOKEN_COUNT_PROMPT = SpanAttributes.LLM_TOKEN_COUNT_PROMPT
+LLM_TOKEN_COUNT_TOTAL = SpanAttributes.LLM_TOKEN_COUNT_TOTAL
+OUTPUT_MIME_TYPE = SpanAttributes.OUTPUT_MIME_TYPE
+OUTPUT_VALUE = SpanAttributes.OUTPUT_VALUE
+RETRIEVAL_DOCUMENTS = SpanAttributes.RETRIEVAL_DOCUMENTS
 
 
 @strawberry.enum
@@ -87,7 +86,7 @@ class SpanEvent:
     ) -> "SpanEvent":
         return SpanEvent(
             name=event.name,
-            message=cast(str, event.attributes.get(EXCEPTION_MESSAGE) or ""),
+            message=cast(str, event.attributes.get(trace_schema.EXCEPTION_MESSAGE) or ""),
             timestamp=event.timestamp,
         )