openAICreateEmbeddings

Given an input text generates the corresponding embedding.

Input

•input: Text from which the embedding is generated.

•model: Model identifier to use. It can use text-embedding-ada-002 or text-search-ada-doc-001. By default, text-embedding-ada-002 is used.

Output

•response: Embedding generated.

openAIIntegrationRule

Given a text and additional parameters, a query is sent to ChatGPT and its answer is awaited.

Input

•model: Model identifier to use. It can use any provided by OpenAI.

•messages: List of messages indicating the context of the conversation. Each message should keep to the format required by OpenAI where the role and content must be specified.

•name: Name of the author of the message.

•temperature: Value indicating randomness in the response by OpenAI. Values correspond to those indicated in its API Reference.

•topp: Alternative to the temperature parameter. Either one should be used.

•n:  Indicates the number of possible answers for the chat indicated in the message list.

•maxtokens: Maximum number of tokens to generate in the chat completion. It is limited by the model used.

•presencepenalty: Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the likelihood that the model will talk about new topics.

•frequencypenalty: Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, reducing the likelihood that the model will repeat the same line word for word.

•user: Unique identifier representing the end user, which can help OpenAI monitor and detect abuse.

Output

•completeResponse: Full answer by OpenAI.

•listOfMessages:  List of messages generated by OpenAI.

•listOfMessagesWithRoles: List of messages generated by OpenAI with roles included.

openAIModerationRule

Given an input text, it is checked for inappropriate content according to the OpenAI content policies.

Input

•input: Input text to classify.

Output

•hate: Represents whether or not the given text has hate content.

•hatethreatening: Represents whether or not the given text has threatening or hate content.

•selfharm: Represents whether or not the given text has self-harm content.

•sexual: Represents whether or not the given text has sexual content.

•sexualminors: Represents whether or not the given text has sexual content on minors.

•violence: Represents whether or not the given text has violence content.

•violencegraphic:  Represents whether or not the given text has violence graphic content.