ml-specialty video for a machine learning engineer is building a neural network to identify different class labels within an image. The source data comes from
A machine learning engineer is building a neural network to identify different class labels within an image. The source data comes from images taken from a car while it is driving. The camera takes separate frames, and these frames are saved to train them in a model later. The neural network consists of 100 hidden layers. The output consists of 10 labels and some of these labels are: “Traffic sign”, “Pedestrian” and “Cross walk”. The input image could contain more than one label and all labels present should be predicted. Concerning the output layer, which activation function should the engineer use?