April 2020
Intermediate to advanced
294 pages
7h 53m
English
Machine learning can only get so far without having a dataset to train the model. You may recall from earlier in the chapter that we discussed how a neural network is a collection of neurons and connections between those neurons. The connection weights and the activation bias can't be set properly without using some form of training to set their values. Setting those weights and values is really the learning aspect for the model.
If you were to go out on the internet, you would discover that there are tons of different datasets that are available online that can be used to train a machine learning model. If you are interested in training a model to detect whether someone is smiling, you can use the Smiles ...
Read now
Unlock full access