Cette formation a pour objectif de fournir aux participants une base solide pour comprendre et utiliser les techniques fondamentales du deep learning. Elle s’adresse à ceux qui souhaitent découvrir les concepts clés de ce domaine en plein essor.Cette formation est idéale pour les développeurs, data scientists, ingénieurs, ou toute personne souhaitant se familiariser avec le deep learning.
- Comprendre les principes de base du deep learning et des réseaux de neurones.
- Apprendre à entraîner et évaluer un modèle de deep learning.
- Découvrir les différentes architectures adaptées à des cas d’usage spécifiques.
.
Il est recommandé d'avoir des connaissances de base sur la programmation et les mathématiques.
AI Fundamentals with IBM SkillsBuild
Discover the history of AI and how it impacts various industries.
Learn how AI interprets language and images.
Understand core concepts and neural networks.
Build and run models using IBM Watson Studio.
Explore responsible AI development.
Learn about AI-related jobs and market demand.
Deep Learning
- Overview of AI, Machine Learning, and Deep Learning.
- Real-world applications (healthcare, self-driving cars).
- Setting up Python and Jupyter Notebooks on Google Colab or Kaggle.
- Real-world applications (healthcare, self-driving cars).
- Setting up Python and Jupyter Notebooks on Google Colab or Kaggle.
- Core Python: Loops, conditions, functions, lists.
- Data manipulation with NumPy and Pandas.
- Data visualization with Matplotlib.
- Data manipulation with NumPy and Pandas.
- Data visualization with Matplotlib.
- TensorFlow vs PyTorch: Choosing a framework.
- Installing deep learning tools (Ubuntu, Windows, Miniconda, CUDA).
- Installing deep learning tools (Ubuntu, Windows, Miniconda, CUDA).
- Introduction to supervised, unsupervised, and reinforcement learning.
- Basic algorithms: Linear Regression, K-Nearest Neighbors.
- Basic algorithms: Linear Regression, K-Nearest Neighbors.
- Structure of neural networks: Neurons, weights, biases.
- Activation functions and training with gradient descent.
- Activation functions and training with gradient descent.
- Computer Vision: MNIST and CIFAR-10 with CNNs.
- Model performance comparison.
- Model performance comparison.
- Autoencoders and Variational Autoencoders (VAEs).
- Generative Adversarial Networks (GANs).
- Generative Adversarial Networks (GANs).
- RNNs and LSTMs: Sentiment analysis.
- Applications of pre-trained models (BERT, GPT, YOLO).
- Applications of pre-trained models (BERT, GPT, YOLO).
- Real-world project on image classification, text generation, or GANs.
Prix: 20 000,00 DA
Durée: 70 h
Travaux pratiques
Dirigé par un instructeur
