Edit on GitHub

ChatAPICall

Documentation for ChatAPICall.

ChatAPICall.ErrResp — Type
ErrResp

Error response from OpenAI API.

Fields

  • message::String: The error message.
  • type::String: The error type.
  • param::Union{Nothing, String}: The parameter that caused the error.
  • code::String: The error code.
source
ChatAPICall.Resp — Type
Resp

Response from OpenAI API.

Fields

  • total_tokens::Int: Total number of tokens in the response.
  • prompt_tokens::Int: Number of tokens in the prompt.
  • completion_tokens::Int: Number of tokens in the completion.
  • content::String: The completion.
  • id::String: The ID of the completion.
  • model::String: The model used to generate the completion.
source
Base.length — Method
length(chat::Chat)

Get the number of messages in the chat log.

source
Base.pop! — Method
pop!(chat::Chat)

Pop the last message from the chat log.

source
Base.show — Method
show(io::IO, chat::Chat)

Show information of the chat log.

source
ChatAPICall.add! — Method
add!(chat::Chat, role::AbstractString, content::AbstractString)

Add a message to the chat log.

source
ChatAPICall.add! — Method
add!(chat::Chat, resp::Resp)

Add a response from OpenAI API to the chat log.

Arguments

  • chat::Chat: The chat log.
  • resp::Resp: The response from OpenAI API.
source
ChatAPICall.getresponse — Function
getresponse( chat::Chat
           , maxrequests::Int=1
           , compact::Bool=true
           , model::AbstractString="gpt-3.5-turbo"
           , options...)::Resp

Get a response from OpenAI API.

Arguments

  • chat::Chat: The chat log.
  • maxrequests::Int: The maximum number of requests to OpenAI API.
  • compact::Bool: Whether to compact the response.
  • model::AbstractString: The model to use.
  • **options: Other options to pass to ChatAPICall.request.
source
ChatAPICall.proxy_on — Method
proxy_on(http::Union{Nothing, AbstractString}, https::Union{Nothing, AbstractString})

Turn on proxy.

source