sleepyask 3.1.0

A small tool for automating collecting data from ChatGPT

A small tool for automating collecting data from ChatGPT over long periods of time.

What does it do?

ChatGPT currently limits the number of questions that users may ask per hour. The goal of this project is to allow users to just leave their computers on for extended periods of time to collect large amounts of responses from ChatGPT. There might not be a lot of practical use for this. Its main use is in research or data analysis.

Install as a Python Library

pip install sleepyask

Documentation

Using the Official ChatGPT API

Authentication

You are required to provide an organization as well as an API Key
organization - Your organization ID. Get it here: https://platform.openai.com/account/org-settings api_key - You create an API Key on OpenAI by. Get it here: https://platform.openai.com/account/api-keys

Clicking on your profile picture on the top-right > View API Keys > Create new secret key.  

count - This specifies the number of workers to create for asking questions. You can have multiple workers asking questions in parallel.

Sample config

config = {
	"organization": "Your OpenAI organization",
	"api_key": "Your OpenAI api key",
	"count": 1 
}

Sample code

from sleepyask.openai import chat

# Your ChatGPT login information
config_1 = {
	"organization": "Your ChatGPT organization",
	"api_key": "Your ChatGPT api key",
	"count": 1
}

config_2 = {
	"organization": "Your ChatGPT organization",
	"api_key": "Your ChatGPT api key",
	"count": 1
}

configs = [config_1, config_2]

## List of questions you would like to ask ChatGPT
question_list = [
  'What is 1 + 1?',
  'What is 1 + 2?',
  'What is 1 + 3?'
]

# The filename in which you would like your responses to be stored.
output_file_path = 'draw.json'  

# Run sleepy_ask
chat.ask(configs=configs,
           questions=question_list,
           output_file_path=output_file_path,
           verbose=True)

Using the Unofficial ChatGPT API

Authentication

There are multiple ways to configure authentication with ChatGPT:
Email + Password

config = {
  "email": "Your ChatGPT email",
  "password": "Your ChatGPT password"
}

Access Token
Access token can be found here: https://chat.openai.com/api/auth/session

config = {
  "access_token": "Your access token"
}

Sleepyask supports the use of multiple accounts

You can use multiple accounts to collect responses at a significantly quicker rate.
Example usage:

from sleepyask.free import chat

# Your ChatGPT login information
config_1 = {
  "email": "Your ChatGPT email",
  "password": "Your ChatGPT password"
}

config_2 = {
  "email": "Your ChatGPT email",
  "password": "Your ChatGPT password"
}

configs = [config_1, config_2]

# List of questions you would like to ask ChatGPT
question_list = [
  'What is 1 + 1?',
  'What is 1 + 2?',
  'What is 1 + 3?'
]

# The filename in which you would like your responses to be stored.
output_file_path = 'draw.json'  

# Run sleepy_ask
chat.ask(configs=configs,
           questions=question_list,
           output_file_path=output_file_path,
           verbose=True)

Get involved

• 🐛 Found a bug or interested in adding a feature? - Create an issue
• 🤗 Want to help? - Create a pull-request!

Credits

• I copied/stole Catppuccin's beautiful colors and README.