Currently docGPT get a JSON string as response from ChatGPT which contains many useless information. For example, when asking "Can you write a PySide6 example?", the response message looks as follows:
{"choices":[{"finish_reason":"stop","index":0,"message":{"content":"\n\nAs an AI language model, writing code is not my strong suit, but I can give you an example of how you can create a PySide6 program that creates a simple user interface. Here is the code snippet for creating a simple PySide6 window:\n\n```\nfrom PySide6.QtWidgets import QApplication, QMainWindow, QLabel\n\napp = QApplication([])\nwindow = QMainWindow()\nwindow.setWindowTitle(\"My App\")\nlabel = QLabel(\"Hello, World!\")\nwindow.setCentralWidget(label)\nwindow.show()\napp.exec()\n```\n\nIn this example, we import the necessary PySide6 elements and create an instance of the QApplication. Then, we create a QMainWindow instance and set its title to \"My App\". We also create a QLabel and set its text to \"Hello, World!\". We set the label as the central widget of the window and show the window. Finally, we call the `exec()` method of the QApplication instance to start the event loop that handles user input and updates the window.","role":"assistant"}}],"created":1677848730,"id":"chatcmpl-6pzOEBYo5OsArqc1WqfyX1zY6Feke","model":"gpt-3.5-turbo-0301","object":"chat.completion","usage":{"completion_tokens":199,"prompt_tokens":16,"total_tokens":215}}
I believe that for the users, only the "content" field is required to extract and output in human-friendly formats:
As an AI language model, writing code is not my strong suit, but I can give you an example of how you can create a PySide6 program that creates a simple user interface. Here is the code snippet for creating a simple PySide6 window:
'''
from PySide6.QtWidgets import QApplication, QMainWindow, QLabel
app = QApplication([])
window = QMainWindow()
window.setWindowTitle("My App")
label = QLabel("Hello, World!")
window.setCentralWidget(label)
window.show()
app.exec()
'''
In this example, we import the necessary PySide6 elements and create an instance of the QApplication. Then, we create a QMainWindow instance and set its title to "My App". We also create a QLabel and set its text to "Hello, World!". We set the label as the central widget of the window and show the window. Finally, we call the `exec()` method of the QApplication instance to start the event loop that handles user input and updates the window.Hello there,
I would like to use my openAI account instead of this one:
function getGPTResponse(question) {
var options = {
'method' : 'post',
'contentType': 'application/json',
// Convert the JavaScript object to a JSON string.
'payload' : JSON.stringify({
'message': question
})
};
const response = UrlFetchApp.fetch('https://chatgpt-api.kesarx.repl.co/chat', options);
const json = JSON.parse(response.getContentText());
return json.choices[0].message.content;
}
Does anyone know how to change it for the openAI chat API? (https://platform.openai.com/docs/api-reference/chat)
Thanks!
Hi!
I got this error trying to run docGPT in Docs:
Exception: Request failed for https://chatgpt-api.kesarx.repl.co returned code 502. Truncated server response: <!DOCTYPE html> <html lang="en"> <head> <title>Unable To Wake Up</title> <link rel="stylesheet" href="https://fonts.googleapis.... (use muteHttpExceptions option to examine full response)
Looks like a bad gateway, but I’m by no means an expert.
Thanks for any help and please let me know if you need more information.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent