Open sda030 opened 1 year ago
Hi @sda030, thanks for this. In principle, I can do it. However, I tried to be as consistent with the official Python package as possible. If I will return other objects, then it won't be inline.
I suggest keeping messages in a stand-alone variable:
library(openai)
messages <- list(
list(
"role" = "system",
"content" = "You are a helpful assistant."
),
list(
"role" = "user",
"content" = "Who won the world series in 2020?"
),
list(
"role" = "assistant",
"content" = "The Los Angeles Dodgers won the World Series in 2020."
),
list(
"role" = "user",
"content" = "Where was it played?"
)
)
response <- create_chat_completion(
model = "gpt-3.5-turbo",
messages = messages
)
Or even keeping it in one list:
library(openai)
chat <- list()
chat[["input_messages"]] <- list(
list(
"role" = "system",
"content" = "You are a helpful assistant."
),
list(
"role" = "user",
"content" = "Who won the world series in 2020?"
),
list(
"role" = "assistant",
"content" = "The Los Angeles Dodgers won the World Series in 2020."
),
list(
"role" = "user",
"content" = "Where was it played?"
)
)
chat[["reply"]] <- create_chat_completion(
model = "gpt-3.5-turbo",
messages = chat[["input_messages"]]
)
Please let me know if this helps.
Hi, great package. For the sake of reproducibility and for ease of continuing a chat, could the
create_chat_completion()
returned result object include the inputted settings (in particular the messages argument)?