All language subtitles for 003 Convolutional Neural Networks summary

After we have a detailed overview of what happens inside convolutional neural networks.

Now, we will have an insight for simple case illustration of how classification works using CNN.

This is the overview of the case and input images processed by CNN architecture and classified as a

cross or a circle.

Now we will go trough on how exactly a classification happens on this simple convolutional architecture.

The input image is a simple black and white image, whereas white pixels represented by pixel one and

black pixel represented by pixel zero.

Following the conversion was carried out using a number of filters in the convolution layers, producing

some filter maps.

After completing the convolution, the activation function is used.

We to maintain the positive value while changing the negative value to zero.

As was previously explained.

The fetal map then comes into the pooling layer.

In this example, we use max palling with radicals, one from the pooling process.

We will have done simple feature maps.

The feature maps is then transformed from a multi-dimensional array to a flat vector so that it can

be used as input for the fully connected layer.

In this example, the flattened result is connected to two output neurons.

Each connection on a neuron has an associated way.

As an additional note, the weights value is obtained during the training process.

The soft makes activation function are applied to the two neurons in order to determine the class probability.

CNN predicts that the input image has a cross shape because the probability of a cross class is the

highest based on the image that was actually seen.

After understanding how deep learning works, now is the time for you to understand how all of these

seven works.

See you then.