Sha256: 15c2897ef3074451eca5b193cc7e14951468b3af82d52ac6f4fe096399be4b55
Contents?: true
Size: 1.81 KB
Versions: 3
Compression:
Stored size: 1.81 KB
Contents
# frozen_string_literal: true
module RubyAmazonBedrock
module PayloadBuilders
module Cohere
# Builds and returns a payload hash suitable for the Cohere command model processing.
# This method constructs a payload with specific parameters like `model_id`,
# `content_type`, `accept`, and a `body` that includes various AI-related settings.
#
# @return [Hash] The constructed payload containing AI model parameters and settings.
class EmbedBase < RubyAmazonBedrock::PayloadBuilders::Base
# Builds and returns a payload suitable for text embedding or search tasks.
# The method configures the necessary payload structure for processing text input,
# specifically for embedding or indexing in an AI-driven search system.
#
# @return [Hash] A hash representing the structured payload for text embedding or search:
# - :model_id [String] Identifier of the AI model that will process the embedding or search task.
# - :content_type [String] Specifies the content type of the payload, typically 'application/json'.
# - :accept [String] Indicates the MIME type the response should conform to.
# - :body [String] A JSON string that includes the following:
# - :texts [Array<String>] The text(s) to be processed, either a single string or an array of strings.
# - :input_type [String] Specifies the nature of the input, set to 'search_document' for search tasks.
def build
{
model_id: model_id,
content_type: 'application/json',
accept: '*/*',
body: {
texts: [@input],
input_type: 'search_document'
}.to_json
}
end
def model_id
# noop
end
end
end
end
end
Version data entries
3 entries across 3 versions & 1 rubygems