lmnr 0.3.3__tar.gz → 0.3.5__tar.gz

{lmnr-0.3.3 → lmnr-0.3.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: lmnr
-Version: 0.3.3
+Version: 0.3.5
 Summary: Python SDK for Laminar AI
 License: Apache-2.0
 Author: lmnr.ai
@@ -41,13 +41,12 @@ You can send events in two ways:
 - `.evaluate_event(name, data)` – for an event that our agent checks for and assigns a value from possible values.
 
 There are 3 types of events:
-- SCORE - this is an integer score where you specify inclusive minimum and maximum.
-- CLASS - this is a classifier with one of the possible values.
-- TAG - this event has no value and can be assigned to a span.
+- Number - Numeric value.
+- String - Arbitrary string.
+- Boolean - Convenient to classify if something has took place or not.
 
 Important notes:
 - If event name does not match anything pre-defined in the UI, the event won't be saved.
-- If event value (when sent with `.event()`) is not in the domain, the event won't be saved.
 
 ## Instrumentation
 

{lmnr-0.3.3 → lmnr-0.3.5}/README.md

@@ -21,13 +21,12 @@ You can send events in two ways:
 - `.evaluate_event(name, data)` – for an event that our agent checks for and assigns a value from possible values.
 
 There are 3 types of events:
-- SCORE - this is an integer score where you specify inclusive minimum and maximum.
-- CLASS - this is a classifier with one of the possible values.
-- TAG - this event has no value and can be assigned to a span.
+- Number - Numeric value.
+- String - Arbitrary string.
+- Boolean - Convenient to classify if something has took place or not.
 
 Important notes:
 - If event name does not match anything pre-defined in the UI, the event won't be saved.
-- If event value (when sent with `.event()`) is not in the domain, the event won't be saved.
 
 ## Instrumentation
 

{lmnr-0.3.3 → lmnr-0.3.5}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "lmnr"
-version = "0.3.3"
+version = "0.3.5"
 description = "Python SDK for Laminar AI"
 authors = [
   { name = "lmnr.ai", email = "founders@lmnr.ai" }
@@ -11,7 +11,7 @@ license = "Apache-2.0"
 
 [tool.poetry]
 name = "lmnr"
-version = "0.3.3"
+version = "0.3.5"
 description = "Python SDK for Laminar AI"
 authors = ["lmnr.ai"]
 readme = "README.md"

{lmnr-0.3.3 → lmnr-0.3.5}/src/lmnr/sdk/context.py

@@ -269,9 +269,9 @@ class LaminarContextManager:
     ) -> Span:
         """Internal method to update a span object. Use `SpanContext.update()` instead."""
         span.update(
-            input=input,
+            input=input or span.input,
+            output=output or span.output,
             end_time=end_time,
-            output=output,
             metadata=metadata,
             attributes=attributes,
             evaluate_events=evaluate_events,

{lmnr-0.3.3 → lmnr-0.3.5}/src/lmnr/sdk/decorators.py

@@ -168,12 +168,12 @@ class LaminarDecorator:
         laminar = LaminarSingleton().get()
         laminar.evaluate_event(name, data)
 
-    def run_pipeline(
+    def run(
         self,
         pipeline: str,
         inputs: dict[str, NodeInput],
-        env: dict[str, str] = None,
-        metadata: dict[str, str] = None,
+        env: dict[str, str] = {},
+        metadata: dict[str, str] = {},
     ) -> PipelineRunResponse:
         """Run the laminar pipeline with the given inputs. Pipeline must be defined in the Laminar UI and have a target version.
 

{lmnr-0.3.3 → lmnr-0.3.5}/src/lmnr/sdk/types.py

@@ -3,6 +3,8 @@ import pydantic
 import uuid
 from typing import Optional, Union
 
+from .utils import to_dict
+
 
 class ChatMessage(pydantic.BaseModel):
     role: str
@@ -30,7 +32,7 @@ class PipelineRunRequest(pydantic.BaseModel):
     def to_dict(self):
         return {
             "inputs": {
-                k: v.model_dump() if isinstance(v, pydantic.BaseModel) else v
+                k: v.model_dump() if isinstance(v, pydantic.BaseModel) else to_dict(v)
                 for k, v in self.inputs.items()
             },
             "pipeline": self.pipeline,

{lmnr-0.3.3 → lmnr-0.3.5}/src/lmnr/sdk/utils.py

@@ -76,10 +76,7 @@ def to_dict(obj: typing.Any) -> dict[str, typing.Any]:
 
         return str(o)
 
-    for key in obj.keys():
-        obj[key] = to_dict_inner(obj[key])
-
-    return obj
+    return to_dict_inner(obj)
 
 
 def get_input_from_func_args(