Encoding removes the redundancies from the data which decreases the size of the file. Since encoded data is smaller in size, we should be able to save space on your storage devicesMachine learning models require all input and output variables to be numeric. This means that the data contains categorical data, must encode it into numbers before it fit and evaluates a model. Encoding is a required pre-processing step when working with categorical data for machine learning. Encoding removes redundancies from data, the size of files will be a lot smaller. This results in a faster input speed when data is saved. Since encoded data is smaller in size it saves space on storage devices. so encode is important.