No-Code Deep Learning
Getting started with
No-Code Deep Learning
Getting started with
- Simulink
- Network Designer App
Deep Learning???
weight
height
Deep Learning???
weight
height
??
Deep Learning???
weight
height
??
y_i = mx_i + c
(x_i,y_i)
Deep Learning???
weight
height
??
y_i = mx_i + c
(x_i,y_i)
y=f(x)
x
y
Deep Learning???
y=f(x)
x
y
NN
Deep Learning???
Primer: NN's
Deep Learning???
Primer: CNN's
Deep Learning???
CNN's
Deep Learning???
Deep Learning???
CNN's
Deep Learning???
Deep Learning???
Why Convolutions?
Deep Learning???
\begin{bmatrix}
0 & -1 & 0 \\
-1 & 0 & 1 \\
0 & 1 & 0 \\
\end{bmatrix}
y_{i+1} - y_{i-1}
x_{i+1} - x_{i-1}
Deep Learning???
Why Convolutions?
Deep Learning???
\begin{bmatrix}
0 & -1 & 0 \\
-1 & 0 & 1 \\
0 & 1 & 0 \\
\end{bmatrix}
y_{i+1} - y_{i-1}
x_{i+1} - x_{i-1}
Differencing
Deep Learning???
Why Convolutions?
Deep Learning???
CNNS: Learn Filters ( Learn to extract features)
No-Code Deep Learning
Agenda
- Inference In Simulink
- Working with Speech Data
- Working with ECG data
- Working with Image Data
- How to use Deep Network Designer App
- Hands-on Demo using Deep Network Designer App
5 Minute Introduction to......
No-Code Deep Learning
Pre-reqs
Matlab 2020b+
Deep Learning Tool box
Check if u have this
No-Code Deep Learning
Getting Started
y=f(x)
No-Code Deep Learning
Getting Started
y=f(x)
x
y
No-Code Deep Learning
Getting Started
y=f(x)
x
y
NN
No-Code Deep Learning
Getting Started
y=f(x)
x
y
NN
They are universal approximators!!!!!!!!!!!
No-Code Deep Learning
They are universal approximators!!!!!!!!!!!
Why Deep Learning??
No-Code Deep Learning
Why Deep Learning??
No-Code Deep Learning
Why Deep Learning??
Deep Learning Systems
No-Code Deep Learning
Getting Started
y=f(x)
x
y
NN
Data
labels
No-Code Deep Learning
Getting Started
x
y
NN
Data
labels
class NaturalSceneClassification(ImageClassificationBase):
def __init__(self):
super().__init__()
self.network = nn.Sequential(
nn.Conv2d(3, 32, kernel_size = 3, padding = 1),
nn.ReLU(),
nn.Conv2d(32,64, kernel_size = 3, stride = 1, padding = 1),
nn.ReLU(),
nn.MaxPool2d(2,2),
nn.Conv2d(64, 128, kernel_size = 3, stride = 1, padding = 1),
nn.ReLU(),
nn.Conv2d(128 ,128, kernel_size = 3, stride = 1, padding = 1),
nn.ReLU(),
nn.MaxPool2d(2,2),
nn.Conv2d(128, 256, kernel_size = 3, stride = 1, padding = 1),
nn.ReLU(),
nn.Conv2d(256,256, kernel_size = 3, stride = 1, padding = 1),
nn.ReLU(),
nn.MaxPool2d(2,2),
nn.Flatten(),
nn.Linear(82944,1024),
nn.ReLU(),
nn.Linear(1024, 512),
nn.ReLU(),
nn.Linear(512,6)
)
def forward(self, xb):
return self.network(xb)No-Code Deep Learning
Getting Started
x
y
NN
Data
labels
class NaturalSceneClassification(ImageClassificationBase):
def __init__(self):
super().__init__()
self.network = nn.Sequential(
nn.Conv2d(3, 32, kernel_size = 3, padding = 1),
nn.ReLU(),
nn.Conv2d(32,64, kernel_size = 3, stride = 1, padding = 1),
nn.ReLU(),
nn.MaxPool2d(2,2),
nn.Conv2d(64, 128, kernel_size = 3, stride = 1, padding = 1),
nn.ReLU(),
nn.Conv2d(128 ,128, kernel_size = 3, stride = 1, padding = 1),
nn.ReLU(),
nn.MaxPool2d(2,2),
nn.Conv2d(128, 256, kernel_size = 3, stride = 1, padding = 1),
nn.ReLU(),
nn.Conv2d(256,256, kernel_size = 3, stride = 1, padding = 1),
nn.ReLU(),
nn.MaxPool2d(2,2),
nn.Flatten(),
nn.Linear(82944,1024),
nn.ReLU(),
nn.Linear(1024, 512),
nn.ReLU(),
nn.Linear(512,6)
)
def forward(self, xb):
return self.network(xb)
No-Code Deep Learning
Workflow in DeepLearning Network Designer App
No-Code Deep Learning
Workflow in DeepLearning Network Designer App
Design Network
Import Data
digitDatasetPath = fullfile(matlabroot,'toolbox','nnet','nndemos', ...
'nndatasets','DigitDataset');
imds = imageDatastore(digitDatasetPath, ...
'IncludeSubfolders',true, ...
'LabelSource','foldernames');Import Data
Import Data
Training
Training
Training
Training
Training
Training
Training
Training
Training
Training
Training
Exporting
Transfer Learning with Network Designer
Easy Projects and papers
