Merge pull request #144 from yujonglee/fix-modelresponse-typing

Fix ModelResponse typing

litellm/llms/anthropic.py

@@ -4,6 +4,7 @@ import requests
 from litellm import logging
 import time 
 from typing import Callable
+from litellm.utils import ModelResponse
 
 class AnthropicConstants(Enum):
     HUMAN_PROMPT = "\n\nHuman:"
@@ -36,7 +37,7 @@ class AnthropicLLM:
             "x-api-key": self.api_key 
         }
 
-    def completion(self, model: str, messages: list, model_response: dict, print_verbose: Callable, optional_params=None, litellm_params=None, logger_fn=None): # logic for parsing in - calling - parsing out model completion calls
+    def completion(self, model: str, messages: list, model_response: ModelResponse, print_verbose: Callable, optional_params=None, litellm_params=None, logger_fn=None): # logic for parsing in - calling - parsing out model completion calls
         model = model
         prompt = f"{AnthropicConstants.HUMAN_PROMPT.value}"
         for message in messages:

litellm/llms/huggingface_restapi.py

@@ -5,6 +5,7 @@ import requests
 from litellm import logging
 import time 
 from typing import Callable
+from litellm.utils import ModelResponse
 
 class HuggingfaceError(Exception):
     def __init__(self, status_code, message):
@@ -26,7 +27,7 @@ class HuggingfaceRestAPILLM():
         if self.api_key != None:
             self.headers["Authorization"] = f"Bearer {self.api_key}" 
 
-    def completion(self, model: str, messages: list, custom_api_base: str, model_response: dict, print_verbose: Callable, optional_params=None, litellm_params=None, logger_fn=None): # logic for parsing in - calling - parsing out model completion calls
+    def completion(self, model: str, messages: list, custom_api_base: str, model_response: ModelResponse, print_verbose: Callable, optional_params=None, litellm_params=None, logger_fn=None): # logic for parsing in - calling - parsing out model completion calls
         if custom_api_base:
             completion_url = custom_api_base
         elif "HF_API_BASE" in os.environ:

litellm/main.py

@@ -40,7 +40,7 @@ def completion(
     # model specific optional params
     # used by text-bison only
     top_k=40, request_timeout=0, # unused var for old version of OpenAI API
-  ):
+  ) -> ModelResponse:
   try:
     model_response = ModelResponse()
     if azure: # this flag is deprecated, remove once notebooks are also updated.

litellm/tests/test_completion.py

@@ -94,14 +94,9 @@ def test_completion_openai():
 
         response_str = response['choices'][0]['message']['content']
         response_str_2 = response.choices[0].message.content
-        print(response_str)
+        assert response_str == response_str_2
-        print(response_str_2)
+        assert type(response_str) == str
-        if type(response_str) != str:
+        assert len(response_str) > 1
-            pytest.fail(f"Error occurred: {e}")
-        if type(response_str_2) != str:
-            pytest.fail(f"Error occurred: {e}")
-        # Add any assertions here to check the response
-        print(response)
     except Exception as e:
         pytest.fail(f"Error occurred: {e}")