chatgptmax
A Python package for sending long input text to OpenAI's GPT models using message chunking.
Read the blog post that inspired this project.
Installation
You can install chatgptmax
using pip:
pip install chatgptmax
Usage
Here's are basic usage examples of the chatgptmax
module. This assumes you have an OpenAI API key set up as an environment variable, OPENAI_API_KEY
.
You can also try out chatgptmax
using the Jupyter notebook provided here.
Clean text with preprocessing
Suppose you have a file named sample.txt
with the following content:
This is a sample text. It contains some stop words that should be removed. We will use the chatgptmax module to clean and process this text.
You can use the clean.text_from_file
method to read and clean the text from this file:
from chatgptmax import clean
file_path = "sample.txt"
cleaned_text = clean.text_from_file(file_path)
print(cleaned_text)
This code will read the content of sample.txt
, remove common stopwords, and print the cleaned text:
This sample text contains stop words removed. We chatgptmax module clean process text.
Send lots of text to ChatGPT
You can use the send
function to send a prompt along with a large amount of text data from a file to ChatGPT:
from chatgptmax import send, read_data
file_path = "path_to_your_file.txt"
text_data = read_data(file_path)
prompt_text = "Summarize the following text for me:"
responses = send(prompt=prompt_text, text_data=text_data)
for response in responses:
print(response)
This code will send the cleaned text as text data to ChatGPT along with the prompt. ChatGPT will provide responses based on your prompt and text data.
Please make sure you have your OpenAI API key properly set up as an environment variable or using a secret management service for this example to work.
For more examples and instructions, read the documentation
Documentation
For more information on how to use chatgptmax, please refer to the official documentation.
Contributing
If you'd like to contribute to this project, please read our Contribution Guide for details on how to get started.
License
This project is licensed under the MIT License - see the LICENSE file for details.