It is not necessary for you to have participated in the 3rd year Bachelor Neural Network optional class.Â
The final grade is computed as follows: 0.3*(Laboratory Grade) + 0.4*(Project Grade) + 0.3 *(Exam Grade).
Points for the laboratory part are obtained by presenting homework during the laboratory classes.
Points for the project part are obtained at the end of the semester from the Benchmark Performance, Project Presentation, and Technical Report.
Points for the course part are obtained at the end of the semester from the exam.
Obtaining at least 50% points for each and all components is required to pass the subject.
GANs
Space & memory considerations
Momentum & Nesterov Accelerated Gradient
Resilient Backpropagation
Adaptive Gradient Algorithm
Root Mean Square Propagation
Adaptive Moment Estimation & Nesterov-accelerated Adaptive Moment Estimation
AdaDelta
Dropout
Basic Statistical Notions
Optimizations (overview)
Weight initialization
Loss Functions
Computation Graph
Backpropagation in Pytorch
Stochastic Gradient Descent
What is PyTorch
Data types
Tensors
Layers
Module base class
Datasets & DataLoaders
Checkpoints
PyTorch Dataset and Dataloader + MLP example
torchvision transforms
Custom dataset and transforms/augmentations
Numpy library
DataTypes
Shapes
Slicing and indexes
Operations with arrays
Serialization/Deserialization