Introduction to Deep Learning Training

Participants must have experience in coding in Python language. If they don't, Python Programming course is recommended before this training. 

The course is suitable for all participants who are familiar with the Python coding language, in addition to having basic knowledge of mathematics and algorithms.

Fundamentals of Deep Learning,

General information about Artificial Neural Networks and how they work, practical applied studies on how they are used in real life,

Designing deep learning models with the Python programming language,

Pytorch Tensors and more advanced Pytorch functions,

Things to consider in Data Science Projects,

Baseline Model, Gradient, Sigmoid, Non-linearity, Regularization, Binary ve Multi-Class Classification, Loss Function...

Introduction to PyTorch

• Pytorch Tensors
• Broadcastin
• Reshaping
• Concatenation-Stacking
• Automatic Differentiation
• (Other more advanced Pytorch functionalities will be shown in the deep learning section as we create our own neural network)

Foundation of Deep Learning

• What should we focus when we first start the project ?
• Baseline Model
• Why is Gradient the direction of the greatest increase ?
• Foundation of Neural Network -Everything can be thought of as functions
• Why to use sigmoid, really ?
• Adding Non-linearity
• How does normalization model helps while training ? - Same space
• Understanding Regularization
• Defining Loss function
• Loss function vs Metric - Loss is for machine Metric is for you
• What is batch and why its size is important ?
• Binary Classification from scratch
• Multi-class classification from scratch