Overview
This node generates embeddings for given text using the DeepSeek API. It is useful for transforming text data into vector representations for tasks like semantic search, recommendation systems, or natural language processing. For example, it can convert product descriptions into embeddings to improve search relevance or analyze customer feedback by embedding text data.
Use Case Examples
- Generating embeddings for product descriptions to enhance search functionality.
- Creating vector representations of customer reviews for sentiment analysis.
Properties
| Name | Meaning |
|---|---|
| Text | The input text to generate embeddings for. |
| Model | The embedding model to use for generating embeddings. |
| Split Into Vectors | Whether to return each vector as a separate item in the output. |
| Vector Dimensions | The dimension size of the embedding vectors (hidden property). |
Output
JSON
embedding- The generated embedding vector for the input text or each vector if split.metadatatext- The original input text used to generate the embedding.model- The model used to generate the embedding.index- The index of the vector if split into multiple vectors.
object- The type of object returned by the API for the embedding.usage- API usage information (present when vectors are not split).model- The model information returned by the API (present when vectors are not split).
Dependencies
- DeepSeek API with API key credential
Troubleshooting
- Common issues include API request failures due to invalid API keys or network problems. Ensure the DeepSeek API credentials are correctly configured.
- If the node throws errors during execution, check if the input text is valid and not empty.
- When 'Split Into Vectors' is enabled, the output will contain multiple items; ensure downstream nodes can handle multiple items.