Method embed_text

skills/embedding/scripts/embed.py:472–500

(self, text: str)

        return vector

    def embed_text(self, text: str) -> List[float]:
        normalized_text = self._normalize_text(text)
        if not normalized_text:
            return [0.0] * self.dimensions

        cached = self._load_cached(normalized_text)
        if cached is not None:
            return cached

        try:
            if self.provider == "openai" and self.client is not None:
                embedding = self._get_openai_embedding(normalized_text)
            elif self.provider == "nscale_api" and self.client is not None:
                embedding = self._get_nscale_api_embedding(normalized_text)
            elif self.provider == "hf_api" and self.client is not None:
                embedding = self._get_hf_api_embedding(normalized_text)
            elif self.provider == "local" and self.local_model is not None:
                embedding = self._get_local_embedding(normalized_text)
            else:
                embedding = self._get_hash_embedding(normalized_text)
        except Exception as exc:
            print(f"Embedding error ({self.provider}:{self.model}): {exc}")
            self.provider = "hash"
            self.model = "hash"
            embedding = self._get_hash_embedding(normalized_text)

        embedding = self._resize_vector(list(embedding))
        self._save_cached(normalized_text, embedding)
        return embedding

    def embed_batch(self, texts: List[str], batch_size: int = 32) -> List[List[float]]:
        normalized_texts = [self._normalize_text(text) for text in texts]