Machine Learning terms

Weights

learned numbers that control how strongly each input influences the models decision

torch.nn

defines structure of the model (layers, activation functions and loss functions)

torch.optim

changes weights (where the REAL ML happens)

backpropogation

traces error back through layers, calculates how weights should be changed

import torch

imports Pytorch library (tensors: vectors, GPU: acceleration of training)

forward pass

putting data into a model and seeing what comes out