forked from phoenix/litellm-mirror
(test) proxy: test_embedding
This commit is contained in:
parent
486037ff8b
commit
c929c274f6
2 changed files with 19 additions and 76 deletions
|
@ -1,52 +0,0 @@
|
||||||
# COMMENT: This is a new test added today Nov 16th, that is flaky - will need to look into this and update what's going wrong here
|
|
||||||
# import subprocess
|
|
||||||
# import time
|
|
||||||
# import openai
|
|
||||||
# import pytest
|
|
||||||
# from dotenv import load_dotenv
|
|
||||||
# import os
|
|
||||||
|
|
||||||
# load_dotenv()
|
|
||||||
|
|
||||||
# ## This tests the litellm proxy cli, it creates a proxy server and makes a basic chat completion request to gpt-3.5-turbo
|
|
||||||
# ## Do not comment this test out
|
|
||||||
|
|
||||||
# def test_basic_proxy_cli_command():
|
|
||||||
|
|
||||||
# # Command to run
|
|
||||||
# print("current working dir", os.getcwd())
|
|
||||||
|
|
||||||
# command = "python3 litellm/proxy/proxy_cli.py --model gpt-3.5-turbo --port 51670 --debug"
|
|
||||||
# print("Running command to start proxy")
|
|
||||||
|
|
||||||
# # Start the subprocess asynchronously
|
|
||||||
# process = subprocess.Popen(command, shell=True)
|
|
||||||
|
|
||||||
# # Allow some time for the proxy server to start (adjust as needed)
|
|
||||||
# time.sleep(1)
|
|
||||||
|
|
||||||
# # Make a request using the openai package
|
|
||||||
# client = openai.OpenAI(
|
|
||||||
# api_key="Your API Key", # Replace with your actual API key
|
|
||||||
# base_url="http://0.0.0.0:51670"
|
|
||||||
# )
|
|
||||||
|
|
||||||
# try:
|
|
||||||
# response = client.chat.completions.create(model="gpt-3.5-turbo", messages=[
|
|
||||||
# {
|
|
||||||
# "role": "user",
|
|
||||||
# "content": "this is a test request, write a short poem"
|
|
||||||
# }
|
|
||||||
# ])
|
|
||||||
# print(response)
|
|
||||||
# response_str = response.choices[0].message.content
|
|
||||||
# assert len(response_str) > 10
|
|
||||||
# except Exception as e:
|
|
||||||
# print("Got exception")
|
|
||||||
# print(e)
|
|
||||||
# process.terminate() # Terminate the subprocess to close down the server
|
|
||||||
# pytest.fail("Basic test, proxy cli failed", e)
|
|
||||||
|
|
||||||
# # Terminate the subprocess to close down the server
|
|
||||||
# process.terminate()
|
|
||||||
# test_basic_proxy_cli_command()
|
|
|
@ -45,7 +45,7 @@ def test_chat_completion():
|
||||||
pytest.fail("LiteLLM Proxy test failed. Exception", e)
|
pytest.fail("LiteLLM Proxy test failed. Exception", e)
|
||||||
|
|
||||||
# Run the test
|
# Run the test
|
||||||
test_chat_completion()
|
# test_chat_completion()
|
||||||
|
|
||||||
|
|
||||||
def test_chat_completion_azure():
|
def test_chat_completion_azure():
|
||||||
|
@ -56,7 +56,7 @@ def test_chat_completion_azure():
|
||||||
"messages": [
|
"messages": [
|
||||||
{
|
{
|
||||||
"role": "user",
|
"role": "user",
|
||||||
"content": "hi"
|
"content": "write 1 sentence poem"
|
||||||
},
|
},
|
||||||
],
|
],
|
||||||
"max_tokens": 10,
|
"max_tokens": 10,
|
||||||
|
@ -67,34 +67,29 @@ def test_chat_completion_azure():
|
||||||
assert response.status_code == 200
|
assert response.status_code == 200
|
||||||
result = response.json()
|
result = response.json()
|
||||||
print(f"Received response: {result}")
|
print(f"Received response: {result}")
|
||||||
|
assert len(result["choices"][0]["message"]["content"]) > 0
|
||||||
except Exception as e:
|
except Exception as e:
|
||||||
pytest.fail("LiteLLM Proxy test failed. Exception", e)
|
pytest.fail("LiteLLM Proxy test failed. Exception", e)
|
||||||
|
|
||||||
# Run the test
|
# Run the test
|
||||||
test_chat_completion()
|
# test_chat_completion_azure()
|
||||||
|
|
||||||
|
|
||||||
# def test_embedding():
|
def test_embedding():
|
||||||
# try:
|
try:
|
||||||
# # Your test data
|
test_data = {
|
||||||
# test_data = {
|
"model": "azure/azure-embedding-model",
|
||||||
# "model": "",
|
"input": ["good morning from litellm"],
|
||||||
# "messages": [
|
}
|
||||||
# {
|
print("testing proxy server with OpenAI embedding")
|
||||||
# "role": "user",
|
response = client.post("/v1/embeddings", json=test_data)
|
||||||
# "content": "hi"
|
|
||||||
# },
|
|
||||||
# ],
|
|
||||||
# "max_tokens": 10,
|
|
||||||
# }
|
|
||||||
# print("testing proxy server with OpenAI embedding")
|
|
||||||
# response = client.post("/v1/embeddings", json=test_data)
|
|
||||||
|
|
||||||
# assert response.status_code == 200
|
assert response.status_code == 200
|
||||||
# result = response.json()
|
result = response.json()
|
||||||
# print(f"Received response: {result}")
|
print(len(result["data"][0]["embedding"]))
|
||||||
# except Exception as e:
|
assert len(result["data"][0]["embedding"]) > 10 # this usually has len==1536 so
|
||||||
# pytest.fail("LiteLLM Proxy test failed. Exception", e)
|
except Exception as e:
|
||||||
|
pytest.fail("LiteLLM Proxy test failed. Exception", e)
|
||||||
|
|
||||||
# # Run the test
|
# Run the test
|
||||||
# test_embedding()
|
# test_embedding()
|
||||||
|
|
Loading…
Add table
Add a link
Reference in a new issue