abetlen / abetlen/llama-cpp-python

Different embedding results between LlamaCppEmbeddings vs server?

Abierto
#975 1 comentario 0 reacciones 0 asignados Ver en GitHub
bug
Lenguaje dominante
Python
Estrellas
10.6k
Forks
1.4k
Métricas de merge de PR
Métricas de PR pendientes

Descripción

# Prerequisites

Please answer the following questions for yourself before submitting an issue.

- [x] I am running the latest code. Development is very rapid so there are no tagged versions as of now.
- [x] I carefully followed the [README.md](https://github.com/abetlen/llama-cpp-python/blob/main/README.md).
- [x] I [searched using keywords relevant to my issue](https://docs.github.com/en/issues/tracking-your-work-with-issues/filtering-and-searching-issues-and-pull-requests) to make sure that I am creating a new issue that is not already open (or closed).
- [x] I reviewed the [Discussions](https://github.com/abetlen/llama-cpp-python/discussions), and have a new bug or useful enhancement to share.

# Expected Behavior

These two Python scripts _should_ return the same vector.

```python3
from langchain.embeddings import LlamaCppEmbeddings
input_size = 5251
llama = LlamaCppEmbeddings(model_path="/home/david_manouchehri/CodeLlama-7b_ggml-model-f16.gguf", n_gpu_layers=1500, n_ctx=input_size, n_batch=input_size)
text = "I am a tomato."
results = llama.embed_documents([text])
print(results)
```

vs.

```sh
python3 -m llama_cpp.server --model ~/CodeLlama-7b_ggml-model-f16.gguf --n_ctx 5251 --n_batch 5251 --host 127.0.0.1 --port 52193 --n_gpu_layers 1500
```

```python
import http.client

conn = http.client.HTTPConnection("127.0.0.1", "52193")
text = "I am a tomato."
payload = {
"input": [text]
}

headers = {
'Content-Type': "application/json",
'Accept': "application/json"
}
import json
conn.request("POST", "/v1/embeddings", json.dumps(payload), headers)

res = conn.getresponse()
data = res.read()

decoded = data.decode("utf-8")

results_decoded = json.loads(decoded)

results = []
for result in results_decoded["data"]:
results.append(result["embedding"])

print(results)
```

# Current Behavior

Right now, these two scripts return two _different_ vectors. I'm not sure why, as they should be identical.

```sh
python3 server_version.py >> server_version.json
python3 langchain_version.py >> langchain_version.json
```

```sh
md5sum server_version.json langchain_version.json
# 74fd5e1d73f95d7d54d6f237a1002661 server_version.json
# 7d40fa1ceb8676ad16c61f69eedcca64 langchain_version.json
```

I'm unsure if this is the same issue as https://github.com/ggerganov/llama.cpp/issues/3287. Still looking into it, just opening this ticket so I and others have something to reference.

Guía de contribución

Abrir la guía de contribución

Evaluación

Este issue todavía no se ha evaluado.

Recibe los nuevos issues en tu correo

Un resumen breve de issues de GitHub para principiantes.