Merge branch 'master' into DOCS-1290

tests/integrations/pandas-test/test_calls_to_pandas.py

@@ -0,0 +1,79 @@
+import pandas as pd
+import pytest
+
+import weave
+
+
+@weave.op
+def func(name: str, age: int) -> str:
+    return f"Hello, {name}! You are {age} years old."
+
+
+@weave.op
+def raising_func(name: str, age: int) -> str:
+    raise ValueError("This is a test error")
+
+
+@pytest.fixture
+def logging_example(client):
+    func("Alice", 30)
+
+    with weave.attributes({"tag": "test", "version": "1.0"}):
+        func("Bob", 25)
+
+    try:
+        raising_func("Claire", 35)
+    except:
+        pass
+
+
+def test_calls_to_pandas_basic(logging_example, client):
+    calls = client.get_calls()
+    df = calls.to_pandas()
+
+    assert isinstance(df, pd.DataFrame)
+    assert len(df) == 3  # The three calls we made
+
+    dictified = df.to_dict(orient="records")
+    calls_as_dicts = [c.to_dict() for c in calls]
+
+    for d1, d2 in zip(dictified, calls_as_dicts):
+        assert d1 == d2
+
+
+def test_calls_to_pandas_with_limit(logging_example, client):
+    calls = client.get_calls(limit=1)
+    df = calls.to_pandas()
+
+    assert isinstance(df, pd.DataFrame)
+    assert len(df) == 1
+
+    dictified = df.to_dict(orient="records")
+
+    # Maintains insertion order
+    d = dictified[0]
+    assert d["inputs"]["name"] == "Alice"
+    assert d["inputs"]["age"] == 30
+
+
+@pytest.mark.asyncio
+async def test_calls_to_pandas_with_evaluations(client):
+    @weave.op
+    def model(x: int, y: int) -> int:
+        return x + y
+
+    ev = weave.Evaluation(
+        dataset=[
+            {"x": 1, "y": 2},
+            {"x": 3, "y": 4},
+            {"x": 5, "y": 6},
+        ]
+    )
+    res = await ev.evaluate(model)
+
+    calls_df = client.get_calls().to_pandas()
+    assert len(calls_df) == (
+        1  # evaluate
+        + 3 * 2  # predict and score + model
+        + 1  # summarize
+    )

weave/trace/weave_client.py

@@ -116,6 +116,8 @@
 from weave.trace_server_bindings.remote_http_trace_server import RemoteHTTPTraceServer
 
 if TYPE_CHECKING:
+    import pandas as pd
+
     from weave.flow.scorer import ApplyScorerResult, Scorer
 
 
@@ -253,6 +255,39 @@ def __len__(self) -> int:
             raise TypeError("This iterator does not support len()")
         return self.size_func()
 
+    def to_pandas(self) -> pd.DataFrame:
+        """Convert the iterator's contents to a pandas DataFrame.
+
+        Returns:
+            A pandas DataFrame containing all the data from the iterator.
+
+        Example:
+            ```python
+            calls = client.get_calls()
+            df = calls.to_pandas()
+            ```
+
+        Note:
+            This method will fetch all data from the iterator, which may involve
+            multiple network calls. For large datasets, consider using limits
+            or filters to reduce the amount of data fetched.
+        """
+        try:
+            import pandas as pd
+        except ImportError:
+            raise ImportError("pandas is required to use this method")
+
+        records = []
+        for item in self:
+            if isinstance(item, dict):
+                records.append(item)
+            elif hasattr(item, "to_dict"):
+                records.append(item.to_dict())
+            else:
+                raise ValueError(f"Unable to convert item to dict: {item}")
+
+        return pd.DataFrame(records)
+
 
 # TODO: should be Call, not WeaveObject
 CallsIter = PaginatedIterator[CallSchema, WeaveObject]