mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2025-10-30 08:42:00 +00:00
a19b5cef16
* ggml : FA supports F32 V * graph : cast KV to F16 when the KV cache is not used ggml-ci * server : add test that exercises embeddings with FA enabled ggml-ci
36 lines
969 B
Python
36 lines
969 B
Python
import asyncio
|
|
import asyncio.threads
|
|
import requests
|
|
import numpy as np
|
|
|
|
|
|
n = 8
|
|
|
|
result = []
|
|
|
|
async def requests_post_async(*args, **kwargs):
|
|
return await asyncio.threads.to_thread(requests.post, *args, **kwargs)
|
|
|
|
async def main():
|
|
model_url = "http://127.0.0.1:6900"
|
|
responses: list[requests.Response] = await asyncio.gather(*[requests_post_async(
|
|
url= f"{model_url}/embedding",
|
|
json= {"content": "a "*1022}
|
|
) for i in range(n)])
|
|
|
|
for response in responses:
|
|
embedding = response.json()["embedding"]
|
|
print(embedding[-8:])
|
|
result.append(embedding)
|
|
|
|
asyncio.run(main())
|
|
|
|
# compute cosine similarity
|
|
|
|
for i in range(n-1):
|
|
for j in range(i+1, n):
|
|
embedding1 = np.array(result[i])
|
|
embedding2 = np.array(result[j])
|
|
similarity = np.dot(embedding1, embedding2) / (np.linalg.norm(embedding1) * np.linalg.norm(embedding2))
|
|
print(f"Similarity between {i} and {j}: {similarity:.2f}")
|