👉 A "strides" in machine learning refers to the number of elements that a convolutional or pooling layer takes as input and produces an output. For example, if we have two images A and B, and we want to extract features from them using a 3x3 convolutional layer, the strides of the first convolutional layer could be set to 1 (meaning it will take in one image at a time) or 2 (meaning it will take in two images at a time
