Important Machine Learning Concepts Part – 1
Input data/variables used by the ML model.
Transforming input features to be more useful for the models. e.g., mapping categories to buckets, normalizing between -1 and 1, removing null.
Training is data used to optimize the model, evaluation is used to asses the model on new data during training, test is used to provide the final result.
Regression helps predict a continuous quantity (e.g., housing price). Classification predicts discrete class labels (e.g., predicting red/blue/green).
Predicts an output by multiplying and summing input features with weights and biases.
Similar to linear regression but predicts a probability.
Model performs great on the input data but poorly on the test data (combat by dropout, early stopping, or reduce # of nodes or layers).
Model neither perform well on training data nor on testing data, and generates a high error rate on both the training set and unseen data.
How much output is determined by the features. More variance often can mean overfitting, more bias can mean a bad model.
Variety of approaches to reduce overfitting, including adding the weights to the loss function, randomly dropping layers (dropout).