Scribd
Upload
36 views17 pages

Day 10

The document provides an overview of different deep learning algorithms including artificial neural networks, recurrent neural networks, and convolutional neural networks. It describes their basic architectures and applications. Implementation details like softmax, backpropagation, and batch training are also covered.

Uploaded by

Sulakx Kuru
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
36 views17 pages

Day 10

The document provides an overview of different deep learning algorithms including artificial neural networks, recurrent neural networks, and convolutional neural networks. It describes their basic architectures and applications. Implementation details like softmax, backpropagation, and batch training are also covered.

Uploaded by

Sulakx Kuru
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 17

Prerequisites for this

Course
Laptop with Good Internet
No basic knowledge is required.
Note & Pen
Deep Learning Algorithm
ARTIFICIAL NEURAL
01
NETWORK (ANN)

RECURRENT NEURAL
02
NETWORK (RNN)

CONVOLUTIONAL
03
NEURAL NETWORK (CNN)
ANN
• Learns any Non-Linear Function, It is known as
Universal Function Approximators Hidden
Input
• Activation Function introduce non linear
property to network, so it will identify complex
relationship between input & output Output

• Output of each neuron is the activation of


weighted sum of Input, If there is no Activation
function, network can't learn non-linear function

• Feed Forward Neural Network – Input processed


in one direction

• When hidden layer is more than one, that is Deep


Neural Network
RNN
Hidden
• Looping system in hidden layer of ANN is Input
known as RNN

• It captures sequential info of input data, that is Output


dependency between words to make
prediction. Whereas, ANN cannot capture
sequential information

• It is mostly used in NLP (Natural Language


Processing)
CNN
• CNN learns the filter automatically to extract the right features from the data

• It captures spatial features (Arrangement of pixels) whereas ANN can’t.

• It don’t have recurrent connections like RNN, instead it has convolution type of
hidden layers

• Convolution and pooling functions are used as activation functions

• CONVOLUTION: Input image and other as Filter on input image(Kernel)


produces output image.

• POOLING: picking maximum value from selected region is Max pooling and vice
versa.
CNN Architecture FC_4
FC_3 Fully Connected
Fully Connected Neural Network
Max-pooling
0
Conv_2
Convolution (2x2) ReLU Activation
(5x5)
Conv_1
Convolution
(5x5)
Max-pooling
(2x2) 1
.
.
2
.
. 9
Output
Input
28x28x1 n1 channels n1 channels n2 channels Flattened
n3 units
(24 x 24 x n1) (12 x 12 x n1) (8 x 8 x n2) n2 channels
(4 x 4 x n2)
Simple Softmax Classification
Simple Softmax Classification
784 Pixels

….

.
.
.
.
Input
28x28x1

.
.
.
.
0 1 2 3 9
100 image at a time
In TensorFlow
SOFTMAX Function

• Softmax activation function will be applied in the last layer of


Neural network, instead of ReLU, tanh, Sigmoid.

• It is used to map the non-normalized output of a network to a


probability distribution over predicted output class. That is it
converts output of last layer into a essential probability
distribution.
Implementation of NN
Feed Forward
• Set of Input features and random weights
• Weights will be optimized by back propagation

Back Propagation
• Calculating error between predicted output and target output
and use Gradient descent method to update weights

Gradient Descent
• Machine Learning algorithm
• It operates iteratively to find the optimal values for its parameters.
user-defined learning rate, and initial parameter values
Vanishing & Exploding Gradient
• It is very common problem in every Neural Network, which is associated
with Backpropagation.

• Weights of network are updated through backpropagation by finding


gradients.

• When the number of hidden layer is high, then the gradient vanishes or
explodes as it propagates backward. It leads instability in network,
unable to learn from training

• The explosion occurs through exponential growth by repeatedly


multiplying gradients through the network layers that have values larger
than 1.0

• It can be fixed by redesigning the network, using Long Short Term


Memory networks, Gradient clipping, etc.
Keras Basic Syntax

Adding Layers

model.add(Dense(12, input_dim=8, init='uniform', activation='relu'))

model.add(Dense(8, activation='relu'))

Compile Model
model.compile(loss='binary_crossentropy',
optimizer='adam', metrics=['accuracy'])
Batch vs Epoch
Training occurs over epochs and each epoch is split into batches.

Epoch - One pass through all of the rows in the training dataset

Batch - One or more samples considered by the model within an epoch before weights are updated.

Keras Basic Syntax


Save & Load Model
Save Load
model_json = model.to_json() json_file = open('model.json', 'r')
with open("model.json", "w") as json_file: loaded_model_json = json_file.read()
json_file.write(model_json) json_file.close()
loaded_model = model_from_json(loaded_model_json)
model.save_weights("model.h5") loaded_model.load_weights("model.h5")
print("Saved model to disk")
THANK YOU

NOVITECH COIMBATORE
novitechresearchanddevelopment

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy