classify the units in Deep learning for image classification -


suppose have database 10 classes, , classification test on deep belief network or convolutional neural network. question how can understand neurons in last layer related object? in 1 of post, person wrote " understand neurons object shoes , ones not put units in last layer supervised classifier(this can multi-class-svm or soft-max-layer). not know how should done? need more expansion.

if have 10 classes, make last layer have 10 neurons , use softmax activation function. make sure lie between 0 , 1 , add 1. then, use index of neuron largest value output class.


Comments

Popular posts from this blog

java - Spring Data JPA: Why findOne(id) executing delete query internally? -

python - Mongodb How to add addtional information when aggregating? -

java - Incorrect order of records in M-M relationship in hibernate -