Skip to content

Latest commit

 

History

History
40 lines (31 loc) · 3.52 KB

File metadata and controls

40 lines (31 loc) · 3.52 KB

PublicRetrieverTemplateResponse

Response containing public retriever configuration as a reusable template. This returns the retriever's configuration in a format that can be directly used in a CreateRetrieverRequest. Users can copy this config, modify it for their needs (e.g., change collection_identifiers), and create their own retriever. Use Case: 1. Browse public retrievers to find patterns you like 2. GET /public/retrievers/{public_name}/template to get the config 3. Modify collection_identifiers and other fields as needed 4. POST /retrievers to create your own retriever with this config 5. Optionally POST /retrievers/{id}/publish to publish it similarly

Properties

Name Type Description Notes
retriever_name str Original retriever name (you'll change this when creating your own). Provided as reference.
description str Original retriever description (you can use or modify this). Provides context about what this retriever does. [optional]
collection_identifiers List[str] IMPORTANT: These are the original collections. You MUST replace these with your own collection identifiers when creating a retriever from this template.
stages List[Dict[str, object]] Pipeline stages configuration. You can use as-is or modify for your needs. This is the core retrieval logic.
input_schema Dict[str, RetrieverInputSchemaFieldOutput] Input schema defining expected inputs. If you change the input field names, make sure to update references in stages (e.g., {{inputs.query}}).
budget_limits Dict[str, object] Budget limits for execution. You can adjust these based on your needs.
tags List[str] Original tags (optional, for reference) [optional]
display_config DisplayConfigOutput OPTIONAL: Display configuration used for the public interface. Include this if you plan to publish your retriever and want to use a similar UI design. Otherwise, you can omit it. [optional]
source_public_name str Public name of the source retriever (for reference)
source_public_url str Public URL of the source retriever (to view it in action)
feature_extractors List[Dict[str, object]] Feature extractors from all collections used by this retriever. Each extractor includes: feature_extractor_name, version, params, input_mappings, collection_id, and collection_name for reference. Shows how each collection processes data into searchable features. [optional]

Example

from mixpeek.models.public_retriever_template_response import PublicRetrieverTemplateResponse

# TODO update the JSON string below
json = "{}"
# create an instance of PublicRetrieverTemplateResponse from a JSON string
public_retriever_template_response_instance = PublicRetrieverTemplateResponse.from_json(json)
# print the JSON string representation of the object
print(PublicRetrieverTemplateResponse.to_json())

# convert the object into a dict
public_retriever_template_response_dict = public_retriever_template_response_instance.to_dict()
# create an instance of PublicRetrieverTemplateResponse from a dict
public_retriever_template_response_from_dict = PublicRetrieverTemplateResponse.from_dict(public_retriever_template_response_dict)

[Back to Model list] [Back to API list] [Back to README]