Spaces:
Sleeping
Sleeping
Download vibia_ai/embed_utils.py from vibiatest/diagnostic: direct link, hf CLI and curl.
- Browser
- Download file 649 Bytes
-
https://huggingface.co/spaces/vibiatest/diagnostic/resolve/main/vibia_ai/embed_utils.py
- Command line
-
hf download hf://spaces/vibiatest/diagnostic/vibia_ai/embed_utils.py
-
curl -L -o embed_utils.py https://huggingface.co/spaces/vibiatest/diagnostic/resolve/main/vibia_ai/embed_utils.py
649 Bytes
| from openai import OpenAI | |
| client = OpenAI() | |
| def embed_text(text: str): | |
| """ | |
| Genera un vector embedding de 1024 dimensiones compatible con Supabase. | |
| Args: | |
| text (str): Texto que se convertirá a embedding. | |
| Returns: | |
| list[float]: Vector de 1024 dimensiones. | |
| """ | |
| if not text or not isinstance(text, str): | |
| raise ValueError("El texto de entrada debe ser una cadena válida.") | |
| response = client.embeddings.create( | |
| model="text-embedding-3-small", # rápido, económico, 1024 dims | |
| input=text, | |
| dimensions=1024 | |
| ) | |
| return response.data[0].embedding | |