Skip to content

solarspaceclouds/building_AI_Applications_with_ChatGPT_API

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

20 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Overview

This is the updated code for the projects from Building AI Applications with ChatGPT APIs book by Martin Yanev after the openai library revamp in Oct/Nov 2023. This code works as of 23 February 2024.

(https://github.com/PacktPublishing/Building-AI-Applications-with-ChatGPT-APIs)

Within each folder,

Step 1: create config.py file to store your openai API_KEY

Step 2: Install required dependencies with

pip install requirements.txt

Step 3:Run the app with

python3 app.py

Sample results for the relevant projects can be found within the respective folders

2 ChatGPT CLone

chatgpt_clone

5 Quiz Generator App with Django

python3 manage.py runserver

chatgpt_quiz_generator chatgpt_quiz_generator_download

6 Language Translator App

chatgpt_language_translator

8 Powerpoint Generator with ChatGPT and DALLE API

  • 1 slide generated per paragraph;
  • paragraphs to be separated by '\n\n'

11 Models Cost Estimation and Performance Analysis

  • jupyter notebook

12 Fine Tuning

  • jupyter notebooks

Chapter12_helper.ipynb has helper functions to convert train_data.json into train_data_prepared.jsonl; training file required to be in the format as specified in train_data_prepared.jsonl for fine-tuning

Chapter12_fine_tune.ipynb uses train_data_prepared.jsonl for fine-tuning.

Helpful resource for openai API migrations

openai/openai-python#742

Other information

Currently no intention to do Chapter 3/4 Code Bug Fixer, Chapter 7 Outlook Email Automatic Reply Generator, Chapter 10 Whisper please refer to the original repository for these: https://github.com/PacktPublishing/Building-AI-Applications-with-ChatGPT-APIs

About

using updated API calls as after openai library Nov 2023 revamp

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published