# Binary classification Logistic regression is an algorithm for binary classification. Used for supervised learning (i.e., with training/labeled data available) Example: input is an image. Using logistic regression, you want to output a label (y) to recognize whether the image is (1) cat or (0) non cat. Images are stored using three matrices. One for each color channel: Red, Green, Blue. If the image is 64x64 pixels, then there are three 64x64 matrices corresponding to the pixel intensity values of the image. To turn the pixel intensity values into a feature vector, we need to unroll the pixel values into an input feature vector x. To do that, we take all the values from all the matrices and add them all to the vector (i.e., one column of values). We end up with a 64x64x3 vector = 12288. * `nx` or `n` will be used to represent the dimension of the input feature x. Binary classification is going from x (image input) to y (output label).