Aim of the project: The project involves implementation of ESRGAN using PyTorch to enhance low resolution images having a lot of loss in details to images having great spatial resolution.
My learnings:
- The ML terminology: training set, features, targets, parameters, model, cost, loss, learning algorithm, learning curve, regression, classification
- Regression: linear regression model, gradient descent learning algorithm, mean squared cost funciton
- Binary Classification: logistic regression model, gradient descent learning algorithm, binary cross entropy cost function
- Neural Network: Neuron, Neural Net, Deep neural nets, hidden layers, activations
- Multiclass Classification: Softmax activation function, sparse categorical cross entropy cost function
- Some useful functions of Python libraries like NumPy, TenserFlow and PyTorch