arize-phoenix 0.0.31rc1__tar.gz → 0.0.31rc2__tar.gz

{arize_phoenix-0.0.31rc1 → arize_phoenix-0.0.31rc2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: arize-phoenix
-Version: 0.0.31rc1
+Version: 0.0.31rc2
 Summary: ML Observability in your notebook
 Project-URL: Documentation, https://docs.arize.com/phoenix/
 Project-URL: Issues, https://github.com/Arize-ai/phoenix/issues
@@ -43,6 +43,7 @@ Requires-Dist: pytest-lazy-fixture; extra == 'dev'
 Requires-Dist: strawberry-graphql[debug-server]==0.178.0; extra == 'dev'
 Provides-Extra: experimental
 Requires-Dist: openai; extra == 'experimental'
+Requires-Dist: tenacity; extra == 'experimental'
 Description-Content-Type: text/markdown
 
 <p align="center">

{arize_phoenix-0.0.31rc1 → arize_phoenix-0.0.31rc2}/pyproject.toml

@@ -53,6 +53,7 @@ dev = [
 ]
 experimental = [
   "openai",
+  "tenacity",
 ]
 
 [project.urls]
@@ -93,6 +94,8 @@ dependencies = [
   "pytest-cov",
   "pytest-lazy-fixture",
   "arize",
+  "openai",
+  "tenacity",
 ]
 
 [tool.hatch.envs.type]
@@ -101,6 +104,7 @@ dependencies = [
   "pandas-stubs",
   "pytest",
   "types-psutil",
+  "tenacity",
 ]
 
 [tool.hatch.envs.style]
@@ -244,6 +248,7 @@ module = [
   "arize.*",
   "portpicker",
   "wrapt",
+  "openai",
 ]
 ignore_missing_imports = true
 

{arize_phoenix-0.0.31rc1 → arize_phoenix-0.0.31rc2}/src/phoenix/__init__.py

@@ -2,8 +2,9 @@ from .datasets.dataset import Dataset
 from .datasets.fixtures import ExampleDatasets, load_example
 from .datasets.schema import EmbeddingColumnNames, RetrievalEmbeddingColumnNames, Schema
 from .session.session import Session, active_session, close_app, launch_app
+from .trace.fixtures import load_example_traces
 
-__version__ = "0.0.31rc1"
+__version__ = "0.0.31rc2"
 
 # module level doc-string
 __doc__ = """
@@ -32,4 +33,5 @@ __all__ = [
     "close_app",
     "launch_app",
     "Session",
+    "load_example_traces",
 ]

{arize_phoenix-0.0.31rc1 → arize_phoenix-0.0.31rc2}/src/phoenix/experimental/evals/retrievals.py

@@ -2,7 +2,7 @@
 Helper functions for evaluating the retrieval step of retrieval-augmented generation.
 """
 
-from typing import List, Optional, Tuple
+from typing import List, Optional
 
 import openai
 from tenacity import (
@@ -26,29 +26,26 @@ _QUERY_CONTEXT_PROMPT_TEMPLATE = """# Query: {query}
 # Answer ("relevant" or "irrelevant"): """
 
 
-def compute_precisions_at_k_and_relevance_classifications(
-    query_texts: List[str],
-    list_of_retrieved_document_texts: List[List[str]],
-    model_name: str = "gpt-4",
-) -> Tuple[List[List[Optional[float]]], List[List[Optional[bool]]]]:
-    if len(query_texts) != len(list_of_retrieved_document_texts):
-        raise ValueError()
-    list_of_precisions_at_k_lists: List[List[Optional[float]]] = []
-    list_of_relevance_classifications_lists: List[List[Optional[bool]]] = []
-    for query_text, retrieved_document_texts in zip(query_texts, list_of_retrieved_document_texts):
-        relevance_classifications = [
-            classify_relevance(query_text, document, model_name=model_name)
-            for document in retrieved_document_texts
-        ]
-        precisions_at_k = compute_precisions_at_k(relevance_classifications)
-        list_of_relevance_classifications_lists.append(relevance_classifications)
-        list_of_precisions_at_k_lists.append(precisions_at_k)
-    return list_of_precisions_at_k_lists, list_of_relevance_classifications_lists
-
-
 def compute_precisions_at_k(
     relevance_classifications: List[Optional[bool]],
 ) -> List[Optional[float]]:
+    """Given a list of relevance classifications, computes precision@k for k = 1, 2, ..., n, where
+    n is the length of the input list.
+
+    Args:
+        relevance_classifications (List[Optional[bool]]): A list of relevance classifications for a
+            set of retrieved documents, sorted by order of retrieval (i.e., the first element is the
+            classification for the first retrieved document, the second element is the
+            classification for the second retrieved document, etc.). The list may contain None
+            values, which indicate that the relevance classification for the corresponding document
+            is unknown.
+
+    Returns:
+        List[Optional[float]]: A list of precision@k values for k = 1, 2, ..., n, where n is the
+            length of the input list. The first element is the precision@1 value, the second element
+            is the precision@2 value, etc. If the input list contains any None values, those values
+            are omitted when computing the precision@k values.
+    """
     precisions_at_k = []
     num_relevant_classifications = 0
     num_non_none_classifications = 0
@@ -66,6 +63,18 @@ def compute_precisions_at_k(
 
 @retry(wait=wait_random_exponential(min=1, max=60), stop=stop_after_attempt(6))
 def classify_relevance(query: str, document: str, model_name: str) -> Optional[bool]:
+    """Given a query and a document, determines whether the document contains an answer to the
+    query.
+
+    Args:
+        query (str): The query text. document (str): The document text. model_name (str): The name
+        of the OpenAI API model to use for the classification.
+
+    Returns:
+        Optional[bool]: A boolean indicating whether the document contains an answer to the query
+            (True meaning relevant, False meaning irrelevant), or None if the LLM produces an
+            unparseable output.
+    """
     prompt = _QUERY_CONTEXT_PROMPT_TEMPLATE.format(
         query=query,
         reference=document,