phoenix/litellm
1
0
Fork 0
forked from phoenix/litellm-mirror
Code Pull requests Activity

fix(router.py): add more debug logs

Browse source
This commit is contained in:
Krrish Dholakia 2024-03-11 12:34:22 -07:00
parent e23c68b15a
commit 2f1899284c
2 changed files with 14 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

7
litellm/router.py
View file

@ -294,11 +294,17 @@ class Router:
""" """
returns a copy of the deployment with the api key masked returns a copy of the deployment with the api key masked
""" """
try:
_deployment_copy = copy.deepcopy(deployment) _deployment_copy = copy.deepcopy(deployment)
litellm_params: dict = _deployment_copy["litellm_params"] litellm_params: dict = _deployment_copy["litellm_params"]
if "api_key" in litellm_params: if "api_key" in litellm_params:
litellm_params["api_key"] = litellm_params["api_key"][:2] + "*" * 10 litellm_params["api_key"] = litellm_params["api_key"][:2] + "*" * 10
return _deployment_copy return _deployment_copy
except Exception as e:
verbose_router_logger.debug(
f"Error occurred while printing deployment - {str(e)}"
)
raise e
### COMPLETION, EMBEDDING, IMG GENERATION FUNCTIONS ### COMPLETION, EMBEDDING, IMG GENERATION FUNCTIONS
@ -310,6 +316,7 @@ class Router:
response = router.completion(model="gpt-3.5-turbo", messages=[{"role": "user", "content": "Hey, how's it going?"}] response = router.completion(model="gpt-3.5-turbo", messages=[{"role": "user", "content": "Hey, how's it going?"}]
""" """
try: try:
verbose_router_logger.debug(f"router.completion(model={model},..)")
kwargs["model"] = model kwargs["model"] = model
kwargs["messages"] = messages kwargs["messages"] = messages
kwargs["original_function"] = self._completion kwargs["original_function"] = self._completion

2
litellm/router_strategy/lowest_tpm_rpm.py
View file

@ -148,6 +148,7 @@ class LowestTPMLoggingHandler(CustomLogger):
input_tokens = token_counter(messages=messages, text=input) input_tokens = token_counter(messages=messages, text=input)
except: except:
input_tokens = 0 input_tokens = 0
verbose_router_logger.debug(f"input_tokens={input_tokens}")
# ----------------------- # -----------------------
# Find lowest used model # Find lowest used model
# ---------------------- # ----------------------
@ -209,4 +210,5 @@ class LowestTPMLoggingHandler(CustomLogger):
elif item_tpm < lowest_tpm: elif item_tpm < lowest_tpm:
lowest_tpm = item_tpm lowest_tpm = item_tpm
deployment = _deployment deployment = _deployment
verbose_router_logger.info(f"returning picked lowest tpm/rpm deployment.")
return deployment return deployment