IA générative : API d'OpenAI, DeepSeek et ChatGPT en Python

Aperçu

Empower Your Business With GenAI, Artificial Intelligence, GPT-4o, o1, o3, DeepSeek, and More! What you'll learn:

How to setup and use the OpenAI API with ChatGPTHow to effectively use prompt engineeringRAG (retrieval-augmented generation) with the OpenAI Embeddings APIFAISS (Facebook AI Similarity Search)How to Fine-Tune ChatGPT Welcome to the forefront of artificial intelligence with our groundbreaking course on Generative AI (GenAI), the OpenAI API, DeepSeek, and ChatGPT.

With ChatGPT and DeepSeek, you'll learn how to build with the world's most advanced Large Language Models (LLMs). This course is a must-have if you want to know how to use this cutting-edge technology for your business and work projects.This course contains 5 main sections:

Basic API Usage:

All the fundamentals:

signup for an account, get your API key, set environment variables on Windows / Linux / Mac, using the API in Python, setup billing, understand the pricing model, and OpenAI's usage policies.

Of note is the chatbot tutorial, which goes over how to incorporate chat history into the model so that ChatGPT "remembers" what it said to you previously. A customer service chatbot will serve as a running example throughout this course.Prompt Engineering:

ChatGPT Prompt Engineering for Developers - All about how to make ChatGPT do what you want it to do.

We'll explore various example use-cases, such as getting ChatGPT to output structured data (JSON, tables), sentiment analysis, language translation, creative writing, text summarization, and question-answering. We'll explore techniques like chain-of-thought (CoT) prompting, and we'll even look at how to use ChatGPT to build a stock trading system!Retrieval Augmented Generation (RAG):

Learn how to incorporate external data into LLMs.

This powerful technique helps mitigate a common problem called "hallucination". It's critical if you have proprietary data (like product info for your company) that your LLM doesn't know about.

You'll learn how semantic search / similarity search works, and how to implement it using FAISS (Facebook AI Similarity Search library). Learn how this will allow you to "chat with your data".Fine-Tuning:

Learn how to "train" an LLM on your own dataset so that it behaves the way you want it to.

Sometimes prompt engineering and RAG won't cut it.GPT-4 with Vision:

Everything in this course can be done with GPT-4, but what makes GPT-4 (and GPT-4 Turbo) special is its vision capabilities. That is, it can understand images.

In this section, we'll explore many of the amazing applications of combined text-image understanding, some of which include automated homework grading, explaining memes and humor, handwriting transcription, web development, game development, and writing product descriptions based on images (business owners - you already know how this will skyrocket your productivity).Throughout this course, you'll engage in hands-on exercises, real-world applications, and expert guidance to solidify your understanding and mastery of generative AI concepts. Whether you're a seasoned developer, aspiring AI enthusiast, or industry professional, this course offers a transformative experience that will empower you to harness the true potential of AI.Are you ready to embark on this exhilarating journey into the future of AI?

Join us and unlock the endless possibilities of Generative AI today!Suggested Prerequisites:

Python coding

Programme

• Introduction au Cours
Aperçu et objectifs du cours
Prérequis et exigences
• Utilisation de base de l'API
Configuration d'un compte OpenAI et obtention d'une clé API
Configuration des variables d'environnement (Windows/Linux/Mac)
Utilisation de l'API OpenAI en Python
Configuration de la facturation et compréhension du modèle de tarification
Aperçu des politiques d'utilisation d'OpenAI
Création d'un chatbot de service client avec historique de chat
• Ingénierie de Prompts
Notions de base des prompts et meilleures pratiques
Cas d'utilisation : sortie en JSON, tableaux, analyse de sentiments
Traduction de langues et écriture créative
Résumé de texte et questions-réponses
Chaîne de pensée (CoT) et prompts
Développement d'un système de trading d'actions avec ChatGPT
• Génération Augmentée par Récupération (RAG)
Introduction à RAG et incorporation de données
Atténuation des hallucinations dans les modèles d'IA
Concepts de recherche sémantique et de recherche de similarité
Mise en œuvre de RAG avec FAISS (Facebook AI Similarity Search)
Applications : "Discuter avec vos données"
• Affinage
Vue d'ensemble et avantages de l'affinage
Entraînement d'un LLM sur des ensembles de données personnalisés
Limitations et quand utiliser l'affinage plutôt que l'ingénierie de prompts
• GPT-4 avec Vision
Introduction aux capacités de vision de GPT-4
Applications de compréhension texte-image
Correction automatisée de devoirs et explication de mèmes
Transcription d'écriture manuscrite
Développement web et de jeux
Création de descriptions de produits à partir d'images
• Exercices pratiques et applications réelles
Projets pratiques et exercices
Exemples réels et études de cas
• Conclusion et prochaines étapes
Récapitulatif des apprentissages clés
Ressources pour un apprentissage et une exploration ultérieurs
Retour et clôture du cours

---

Enseigné par

Lazy Programmer Inc. and Lazy Programmer Team

---

Matières

Computer Science