PyTorch is a machine learning package based on Torch. Torch is a ML library based on the Lua programming language, first released in Oct-2002. PyTorch is primarily and actively developed by Facebook's AI Research (FAIR) Lab. Mainly used for NLP and Computer Vision applications, many programs such as Tesla's Autopilot, Uber's Pyro are built on PyTorch.
PyTorch package provides two high-level features:
Use Tensor computation (like Numpy) accelerated using GPU. Deep Neural Networks built on tape-based automatic differentiation system. Convolutional Neural Networks Deep Neural Networks: Traditional Neural Networks are string of neurons connected together.
It has the: Input layer, Hidden Layers, Output layer.
The input flows from the Input layer to the Output layer through the hidden layers.
Kernel Convolution: A process where we take a small grid of numbers & pass them over the whole image transforming it. By using different numbers we can form blurs, edge detection, sharpen etc.,
Convolutional Neural Networks CNN's combine traditional Neural Networks and Convolutions.
About the Dataset Fashion-MNIST is a dataset of Zalando's article images—consisting of a training set of 60,000 examples and a test set of 10,000 examples. Each example is a 28x28 grayscale image, associated with a label from 10 classes.