Machine learning metrics, loss functions, and statistical learning formulae.
Average of squared differences between predicted and actual values.
Square root of MSE, in the same units as the target variable.
Average of absolute differences between predictions and actual values.
Proportion of variance in the target explained by the model.
Fraction of predictions that are correct.
Of all positive predictions, the fraction that were correct.
Of all actual positives, the fraction correctly identified.
Harmonic mean of precision and recall.
Of all actual negatives, the fraction correctly identified.
Maps any real number to the range (0, 1); used in logistic regression.
Converts a vector of scores into a probability distribution.
Rectified Linear Unit: outputs the input if positive, else zero.
Loss function for binary classification.
Iterative optimization step to minimize a cost function.
Measures similarity between two vectors by the angle between them.
Straight-line distance between two points in n-dimensional space.
Distance measured along axes at right angles (taxicab geometry).
Measures the uncertainty or information content of a distribution.
Probability of misclassifying a random element; used in decision trees.
Standardizes features to zero mean and unit variance.
Rescales features to the range [0, 1].
Total expected error decomposes into bias, variance, and irreducible noise.
Loss for multi-class classification with one-hot labels.
Term importance: frequency in a document scaled by rarity across documents.