forked from phoenix/litellm-mirror
refactor: add black formatting
This commit is contained in:
Krrish Dholakia
parent
b87d630b0a
commit
4905929de3
156 changed files with 19723 additions and 10869 deletions
|
|
@ -9,11 +9,12 @@ import os, io
|
|||
|
||||
sys.path.insert(
|
||||
0, os.path.abspath("../..")
|
||||
) # Adds the parent directory to the system path
|
||||
) # Adds the parent directory to the system path
|
||||
import pytest, logging
|
||||
import litellm
|
||||
from litellm import embedding, completion, completion_cost, Timeout
|
||||
from litellm import RateLimitError
|
||||
|
||||
# Configure logging
|
||||
logging.basicConfig(
|
||||
level=logging.DEBUG, # Set the desired logging level
|
||||
|
|
@ -23,19 +24,23 @@ logging.basicConfig(
|
|||
# test /chat/completion request to the proxy
|
||||
from fastapi.testclient import TestClient
|
||||
from fastapi import FastAPI
|
||||
from litellm.proxy.proxy_server import router, save_worker_config, initialize # Replace with the actual module where your FastAPI router is defined
|
||||
from litellm.proxy.proxy_server import (
|
||||
router,
|
||||
save_worker_config,
|
||||
initialize,
|
||||
) # Replace with the actual module where your FastAPI router is defined
|
||||
|
||||
# Your bearer token
|
||||
token = ""
|
||||
|
||||
headers = {
|
||||
"Authorization": f"Bearer {token}"
|
||||
}
|
||||
|
||||
headers = {"Authorization": f"Bearer {token}"}
|
||||
|
||||
|
||||
@pytest.fixture(scope="function")
|
||||
def client_no_auth():
|
||||
# Assuming litellm.proxy.proxy_server is an object
|
||||
from litellm.proxy.proxy_server import cleanup_router_config_variables
|
||||
|
||||
cleanup_router_config_variables()
|
||||
filepath = os.path.dirname(os.path.abspath(__file__))
|
||||
config_fp = f"{filepath}/test_configs/test_config_no_auth.yaml"
|
||||
|
|
@ -46,6 +51,7 @@ def client_no_auth():
|
|||
|
||||
return TestClient(app)
|
||||
|
||||
|
||||
def test_chat_completion(client_no_auth):
|
||||
global headers
|
||||
try:
|
||||
|
|
@ -53,14 +59,11 @@ def test_chat_completion(client_no_auth):
|
|||
test_data = {
|
||||
"model": "gpt-3.5-turbo",
|
||||
"messages": [
|
||||
{
|
||||
"role": "user",
|
||||
"content": "hi"
|
||||
},
|
||||
{"role": "user", "content": "hi"},
|
||||
],
|
||||
"max_tokens": 10,
|
||||
}
|
||||
|
||||
|
||||
print("testing proxy server with chat completions")
|
||||
response = client_no_auth.post("/v1/chat/completions", json=test_data)
|
||||
print(f"response - {response.text}")
|
||||
|
|
@ -70,41 +73,41 @@ def test_chat_completion(client_no_auth):
|
|||
except Exception as e:
|
||||
pytest.fail(f"LiteLLM Proxy test failed. Exception - {str(e)}")
|
||||
|
||||
|
||||
# Run the test
|
||||
|
||||
def test_chat_completion_azure(client_no_auth):
|
||||
|
||||
def test_chat_completion_azure(client_no_auth):
|
||||
global headers
|
||||
try:
|
||||
# Your test data
|
||||
test_data = {
|
||||
"model": "azure/chatgpt-v-2",
|
||||
"messages": [
|
||||
{
|
||||
"role": "user",
|
||||
"content": "write 1 sentence poem"
|
||||
},
|
||||
{"role": "user", "content": "write 1 sentence poem"},
|
||||
],
|
||||
"max_tokens": 10,
|
||||
}
|
||||
|
||||
|
||||
print("testing proxy server with Azure Request /chat/completions")
|
||||
response = client_no_auth.post("/v1/chat/completions", json=test_data)
|
||||
|
||||
assert response.status_code == 200
|
||||
result = response.json()
|
||||
print(f"Received response: {result}")
|
||||
assert len(result["choices"][0]["message"]["content"]) > 0
|
||||
assert len(result["choices"][0]["message"]["content"]) > 0
|
||||
except Exception as e:
|
||||
pytest.fail(f"LiteLLM Proxy test failed. Exception - {str(e)}")
|
||||
|
||||
|
||||
# Run the test
|
||||
# test_chat_completion_azure()
|
||||
|
||||
### EMBEDDING
|
||||
|
||||
### EMBEDDING
|
||||
def test_embedding(client_no_auth):
|
||||
global headers
|
||||
from litellm.proxy.proxy_server import user_custom_auth
|
||||
from litellm.proxy.proxy_server import user_custom_auth
|
||||
|
||||
try:
|
||||
test_data = {
|
||||
|
|
@ -117,13 +120,14 @@ def test_embedding(client_no_auth):
|
|||
assert response.status_code == 200
|
||||
result = response.json()
|
||||
print(len(result["data"][0]["embedding"]))
|
||||
assert len(result["data"][0]["embedding"]) > 10 # this usually has len==1536 so
|
||||
assert len(result["data"][0]["embedding"]) > 10 # this usually has len==1536 so
|
||||
except Exception as e:
|
||||
pytest.fail(f"LiteLLM Proxy test failed. Exception - {str(e)}")
|
||||
|
||||
|
||||
def test_bedrock_embedding(client_no_auth):
|
||||
global headers
|
||||
from litellm.proxy.proxy_server import user_custom_auth
|
||||
from litellm.proxy.proxy_server import user_custom_auth
|
||||
|
||||
try:
|
||||
test_data = {
|
||||
|
|
@ -136,13 +140,14 @@ def test_bedrock_embedding(client_no_auth):
|
|||
assert response.status_code == 200
|
||||
result = response.json()
|
||||
print(len(result["data"][0]["embedding"]))
|
||||
assert len(result["data"][0]["embedding"]) > 10 # this usually has len==1536 so
|
||||
assert len(result["data"][0]["embedding"]) > 10 # this usually has len==1536 so
|
||||
except Exception as e:
|
||||
pytest.fail(f"LiteLLM Proxy test failed. Exception - {str(e)}")
|
||||
|
||||
|
||||
def test_sagemaker_embedding(client_no_auth):
|
||||
global headers
|
||||
from litellm.proxy.proxy_server import user_custom_auth
|
||||
from litellm.proxy.proxy_server import user_custom_auth
|
||||
|
||||
try:
|
||||
test_data = {
|
||||
|
|
@ -155,24 +160,26 @@ def test_sagemaker_embedding(client_no_auth):
|
|||
assert response.status_code == 200
|
||||
result = response.json()
|
||||
print(len(result["data"][0]["embedding"]))
|
||||
assert len(result["data"][0]["embedding"]) > 10 # this usually has len==1536 so
|
||||
assert len(result["data"][0]["embedding"]) > 10 # this usually has len==1536 so
|
||||
except Exception as e:
|
||||
pytest.fail(f"LiteLLM Proxy test failed. Exception - {str(e)}")
|
||||
|
||||
|
||||
# Run the test
|
||||
# test_embedding()
|
||||
#### IMAGE GENERATION
|
||||
|
||||
|
||||
|
||||
def test_img_gen(client_no_auth):
|
||||
global headers
|
||||
from litellm.proxy.proxy_server import user_custom_auth
|
||||
from litellm.proxy.proxy_server import user_custom_auth
|
||||
|
||||
try:
|
||||
test_data = {
|
||||
"model": "dall-e-3",
|
||||
"prompt": "A cute baby sea otter",
|
||||
"n": 1,
|
||||
"size": "1024x1024"
|
||||
"size": "1024x1024",
|
||||
}
|
||||
|
||||
response = client_no_auth.post("/v1/images/generations", json=test_data)
|
||||
|
|
@ -180,41 +187,41 @@ def test_img_gen(client_no_auth):
|
|||
assert response.status_code == 200
|
||||
result = response.json()
|
||||
print(len(result["data"][0]["url"]))
|
||||
assert len(result["data"][0]["url"]) > 10
|
||||
assert len(result["data"][0]["url"]) > 10
|
||||
except Exception as e:
|
||||
pytest.fail(f"LiteLLM Proxy test failed. Exception - {str(e)}")
|
||||
|
||||
#### ADDITIONAL
|
||||
|
||||
#### ADDITIONAL
|
||||
# @pytest.mark.skip(reason="hitting yaml load issues on circle-ci")
|
||||
def test_add_new_model(client_no_auth):
|
||||
global headers
|
||||
try:
|
||||
try:
|
||||
test_data = {
|
||||
"model_name": "test_openai_models",
|
||||
"litellm_params": {
|
||||
"model": "gpt-3.5-turbo",
|
||||
"model": "gpt-3.5-turbo",
|
||||
},
|
||||
"model_info": {
|
||||
"description": "this is a test openai model"
|
||||
}
|
||||
"model_info": {"description": "this is a test openai model"},
|
||||
}
|
||||
client_no_auth.post("/model/new", json=test_data, headers=headers)
|
||||
response = client_no_auth.get("/model/info", headers=headers)
|
||||
assert response.status_code == 200
|
||||
result = response.json()
|
||||
result = response.json()
|
||||
print(f"response: {result}")
|
||||
model_info = None
|
||||
for m in result["data"]:
|
||||
if m["model_name"] == "test_openai_models":
|
||||
model_info = m["model_info"]
|
||||
assert model_info["description"] == "this is a test openai model"
|
||||
except Exception as e:
|
||||
except Exception as e:
|
||||
pytest.fail(f"LiteLLM Proxy test failed. Exception {str(e)}")
|
||||
|
||||
|
||||
def test_health(client_no_auth):
|
||||
global headers
|
||||
import time
|
||||
|
||||
try:
|
||||
response = client_no_auth.get("/health")
|
||||
assert response.status_code == 200
|
||||
|
|
@ -222,19 +229,24 @@ def test_health(client_no_auth):
|
|||
assert result["unhealthy_count"] == 0
|
||||
except Exception as e:
|
||||
pytest.fail(f"LiteLLM Proxy test failed. Exception - {str(e)}")
|
||||
|
||||
|
||||
# test_add_new_model()
|
||||
|
||||
from litellm.integrations.custom_logger import CustomLogger
|
||||
|
||||
|
||||
class MyCustomHandler(CustomLogger):
|
||||
def log_pre_api_call(self, model, messages, kwargs):
|
||||
def log_pre_api_call(self, model, messages, kwargs):
|
||||
print(f"Pre-API Call")
|
||||
|
||||
def log_success_event(self, kwargs, response_obj, start_time, end_time):
|
||||
def log_success_event(self, kwargs, response_obj, start_time, end_time):
|
||||
print(f"On Success")
|
||||
assert kwargs["user"] == "proxy-user"
|
||||
assert kwargs["model"] == "gpt-3.5-turbo"
|
||||
assert kwargs["max_tokens"] == 10
|
||||
|
||||
|
||||
customHandler = MyCustomHandler()
|
||||
|
||||
|
||||
|
|
@ -243,19 +255,16 @@ def test_chat_completion_optional_params(client_no_auth):
|
|||
# This tests if all the /chat/completion params are passed to litellm
|
||||
try:
|
||||
# Your test data
|
||||
litellm.set_verbose=True
|
||||
litellm.set_verbose = True
|
||||
test_data = {
|
||||
"model": "gpt-3.5-turbo",
|
||||
"messages": [
|
||||
{
|
||||
"role": "user",
|
||||
"content": "hi"
|
||||
},
|
||||
{"role": "user", "content": "hi"},
|
||||
],
|
||||
"max_tokens": 10,
|
||||
"user": "proxy-user"
|
||||
"user": "proxy-user",
|
||||
}
|
||||
|
||||
|
||||
litellm.callbacks = [customHandler]
|
||||
print("testing proxy server: optional params")
|
||||
response = client_no_auth.post("/v1/chat/completions", json=test_data)
|
||||
|
|
@ -265,28 +274,39 @@ def test_chat_completion_optional_params(client_no_auth):
|
|||
except Exception as e:
|
||||
pytest.fail("LiteLLM Proxy test failed. Exception", e)
|
||||
|
||||
|
||||
# Run the test
|
||||
# test_chat_completion_optional_params()
|
||||
|
||||
# Test Reading config.yaml file
|
||||
# Test Reading config.yaml file
|
||||
from litellm.proxy.proxy_server import load_router_config
|
||||
|
||||
|
||||
def test_load_router_config():
|
||||
try:
|
||||
print("testing reading config")
|
||||
# this is a basic config.yaml with only a model
|
||||
filepath = os.path.dirname(os.path.abspath(__file__))
|
||||
result = load_router_config(router=None, config_file_path=f"{filepath}/example_config_yaml/simple_config.yaml")
|
||||
result = load_router_config(
|
||||
router=None,
|
||||
config_file_path=f"{filepath}/example_config_yaml/simple_config.yaml",
|
||||
)
|
||||
print(result)
|
||||
assert len(result[1]) == 1
|
||||
|
||||
# this is a load balancing config yaml
|
||||
result = load_router_config(router=None, config_file_path=f"{filepath}/example_config_yaml/azure_config.yaml")
|
||||
result = load_router_config(
|
||||
router=None,
|
||||
config_file_path=f"{filepath}/example_config_yaml/azure_config.yaml",
|
||||
)
|
||||
print(result)
|
||||
assert len(result[1]) == 2
|
||||
|
||||
# config with general settings - custom callbacks
|
||||
result = load_router_config(router=None, config_file_path=f"{filepath}/example_config_yaml/azure_config.yaml")
|
||||
result = load_router_config(
|
||||
router=None,
|
||||
config_file_path=f"{filepath}/example_config_yaml/azure_config.yaml",
|
||||
)
|
||||
print(result)
|
||||
assert len(result[1]) == 2
|
||||
|
||||
|
|
@ -295,24 +315,38 @@ def test_load_router_config():
|
|||
litellm.cache = None
|
||||
load_router_config(
|
||||
router=None,
|
||||
config_file_path=f"{filepath}/example_config_yaml/cache_no_params.yaml"
|
||||
config_file_path=f"{filepath}/example_config_yaml/cache_no_params.yaml",
|
||||
)
|
||||
assert litellm.cache is not None
|
||||
assert "redis_client" in vars(litellm.cache.cache) # it should default to redis on proxy
|
||||
assert litellm.cache.supported_call_types == ['completion', 'acompletion', 'embedding', 'aembedding'] # init with all call types
|
||||
|
||||
assert "redis_client" in vars(
|
||||
litellm.cache.cache
|
||||
) # it should default to redis on proxy
|
||||
assert litellm.cache.supported_call_types == [
|
||||
"completion",
|
||||
"acompletion",
|
||||
"embedding",
|
||||
"aembedding",
|
||||
] # init with all call types
|
||||
|
||||
print("testing reading proxy config for cache with params")
|
||||
load_router_config(
|
||||
router=None,
|
||||
config_file_path=f"{filepath}/example_config_yaml/cache_with_params.yaml"
|
||||
config_file_path=f"{filepath}/example_config_yaml/cache_with_params.yaml",
|
||||
)
|
||||
assert litellm.cache is not None
|
||||
print(litellm.cache)
|
||||
print(litellm.cache.supported_call_types)
|
||||
print(vars(litellm.cache.cache))
|
||||
assert "redis_client" in vars(litellm.cache.cache) # it should default to redis on proxy
|
||||
assert litellm.cache.supported_call_types == ['embedding', 'aembedding'] # init with all call types
|
||||
assert "redis_client" in vars(
|
||||
litellm.cache.cache
|
||||
) # it should default to redis on proxy
|
||||
assert litellm.cache.supported_call_types == [
|
||||
"embedding",
|
||||
"aembedding",
|
||||
] # init with all call types
|
||||
|
||||
except Exception as e:
|
||||
pytest.fail("Proxy: Got exception reading config", e)
|
||||
# test_load_router_config()
|
||||
|
||||
|
||||
# test_load_router_config()
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue