Actions13
Overview
This node integrates with the FastGPT AI customer service platform to manage datasets, specifically enabling the creation of a "link collection" dataset. It allows users to organize and train AI models on collections of web links grouped under a parent directory in FastGPT.
Typical use cases include:
- Creating structured collections of URLs for AI training or knowledge base purposes.
- Organizing web resources into hierarchical folders for easier management.
- Preparing link-based datasets with specific chunking or QA splitting strategies to optimize AI model training.
For example, a user might create a link collection pointing to various documentation pages, specifying how the content should be split or processed during training.
Properties
| Name | Meaning |
|---|---|
| 父级目录ID (datasetParentId) | The ID of the parent folder in FastGPT where the new link collection will be created. Users can find this by opening the folder in FastGPT and copying the browser's folder ID. |
| 选择知识库 (datasetKnowledgeId) | Selects the knowledge base (dataset) under the specified parent directory. This dropdown is dynamically loaded based on the parent directory ID and lists only datasets of type "dataset". |
| 网页链接 (datasetLink) | The URL link to be included in the link collection dataset. |
| 训练类型 (datasetTrainingType) | The training type for processing the dataset. Options are: - 按文本长度进行分割 ("chunk"): Split by text length. - QA拆分 ("qa"): Split using a question-answer approach. |
| Chunk长度 (datasetChunkSize) | When training type is "chunk", this number defines the maximum length of each text chunk. Default is 3000 characters. |
| 自定义最高优先分割符号 (datasetChunkSplitter) | When training type is "chunk", this optional string defines a custom highest-priority delimiter used to split the text chunks. |
| QA拆分自定义提示词 (datasetQAPrompt) | When training type is "qa", this optional string provides a custom prompt for the QA splitting method. |
Output
The node outputs JSON data representing the result of the dataset creation operation. This typically includes metadata about the newly created link collection dataset such as its ID, name, and other relevant details returned from the FastGPT API.
No binary data output is involved.
Dependencies
- Requires an API key credential for authenticating with the FastGPT platform.
- The node sends HTTP POST requests to FastGPT endpoints, relying on the base URL and authentication configured in the credentials.
- Dynamic loading of knowledge base options depends on the parent directory ID input.
Troubleshooting
- Invalid Parent Directory ID: If the provided parent directory ID is incorrect or inaccessible, the node may fail to load knowledge bases or create the dataset. Verify the ID by checking it directly in the FastGPT UI.
- Missing or Invalid API Credentials: Authentication errors occur if the API key is missing or invalid. Ensure the credential is correctly set up in n8n.
- Incorrect Training Type or Parameters: Providing incompatible parameters for the selected training type (e.g., chunk size without "chunk" training type) may cause errors. Double-check property dependencies.
- Network or API Errors: Since HTTP status errors are ignored by default, unexpected failures might not throw explicit errors. Check API response logs for troubleshooting.
Links and References
- FastGPT Official Documentation (replace with actual URL)
- n8n Documentation on Creating Custom Nodes
- General info on Dataset Management in AI Platforms (replace with relevant resource)