This is a gentle introduction to the architecture of the convolutional neural net.Here I have mentioned about operations in the convolutional layer in depth
21. These nets are huge!
Millions of layers
Millions of nodes
Billions of parameters
BUT DON’T KNOW HOW
MANY HIDDEN LAYERS /
NODES TO USE?
WHEN YOU WANT TO
BUILD A NEURAL NET
46. More generally...
Filter search for things in the image
When filter sees something similar to
it’s orientation in an image it will fire
up!
47. Image get transformed into
something new !
We call this new image as a feature map
Because it mapps some features from the original
image w.r.t to the filter!
Right - Feature map obtained by applying canny edge filter on an image
48. It’s not only a single filter!
Many filters searching for diffrent things!
How amazing!
So we will have different feature maps!
49. Simply each filter has it’s own task
Gabor filters are similar to those of the human visual system, and
they have been found to be particularly appropriate for texture
representation and discrimination.
54. All these filters are trainable!
Means ?
These are not man made filters!
Look at the values of following filter kernels
55.
56.
57. What about the filter?
We can design size , channels of the filter
But not its values!
Each filter values is trainable weight
58. All in one!
So every filter values is trainable
It’s like a set of weights !
But how ?
With backprop
Convolution is a perfectly differentiable function!
So we can learn it’s weight parameters
62. Then we apply some nonlinearity
We apply rely on each point in the feature maps
Remember : These points in the feature maps are like neurons
This is like just forgetting the negative parts of the
65. Then the pooling layer
This is like a filter without
parameters!
This is for subsample or to
reduce the dimensions of
the feature maps
Simply this will extract the
most important features
from the feature map
67. So finally….
We can train these filter weights using backprop and some
kind of optimization algorithms(Adam)
Here’s a such view of a trained filter