[Last updated 8/2024] A deep understanding of deep learning (with Python intro) (Udemy – Vietsub and Engsub)

About Course

Views

Gain
A Deep Understanding of Deep Learning
– Master Neural Networks, AI, and Machine Learning (Includes
Python Introduction
)
What you’ll learn:
The theory and math underlying deep learning
How to build artificial neural networks
Architectures of feedforward and convolutional networks
Building models in PyTorch
The calculus and code of gradient descent
Fine-tuning deep network models
Learn Python from scratch (no prior coding experience necessary)
How and why autoencoders work
How to use transfer learning
Improving model performance using regularization
Optimizing weight initializations
Understand image convolution using predefined and learned kernels
Whether deep learning models are understandable or mysterious black-boxes!
Using GPUs for deep learning (much faster than CPUs!)
Description:
Unlock the power of
deep learning
in this comprehensive course designed for beginners and professionals alike. With a solid introduction to
Python
, you’ll delve into neural networks, AI, and cutting-edge machine learning techniques. Gain hands-on experience with real-world projects, understand the intricacies of
deep learning algorithms
, and discover how to apply these skills in various industries. This course ensures you build a strong foundation while preparing you to excel in the dynamic world of AI and data science. Join now and transform your career with expertise in
deep learning
!

Link gốc:

https://www.udemy.com/course/deeplearning_x/

Time Course:
57.5 hours (265 Lectures + Documents)

Instructor
: Mike X Cohen
Total Weight:
23.82 GB
** Note
:

Chú ý:

Course Content

16 – Autoencoders

001 What are autoencoders and what do they do.mp4

11:42
002 Denoising MNIST.mp4

15:48
003 CodeChallenge How many units.mp4

19:52
004 AEs for occlusion.mp4

17:55
005 The latent code of MNIST.mp4

21:57
006 Autoencoder with tied weights.mp4

24:14

31 – Python intro Text and plots

30 – Python intro Flow control

001 If-else statements.mp4

15:03
002 If-else statements, part 2.mp4

16:58
003 For loops.mp4

17:37
004 Enumerate and zip.mp4

12:11
005 Continue.mp4

07:23
006 Initializing variables.mp4

18:00
007 Single-line loops (list comprehension).mp4

15:25
008 while loops.mp4

19:30
009 Broadcasting in numpy.mp4

15:40
010 Function error checking and handling.mp4

17:42

29 – Python intro Functions

28 – Python intro Indexing, slicing

27 – Python intro Data types

26 – Where to go from here

25 – Ethics of deep learning

24 – RNNs (Recurrent Neural Networks) (and GRULSTM)

001 Leveraging sequences in deep learning.mp4

12:53
002 How RNNs work.mp4

15:14
003 The RNN class in PyTorch.mp4

17:44
004 Predicting alternating sequences.mp4

19:30
005 CodeChallenge sine wave extrapolation.mp4

24:49
006 More on RNNs Hidden states, embeddings.mp4

15:51
007 GRU and LSTM.mp4

23:08
008 The LSTM and GRU classes.mp4

13:26
009 Lorem ipsum.mp4

25:10

23 – Generative adversarial networks

22 – Style transfer

21 – Transfer learning

20 – CNN milestone projects

19 – Understand and design CNNs

001 The canonical CNN architecture.mp4

10:47
002 CNN to classify MNIST digits.mp4

26:06
003 CNN on shifted MNIST.mp4

08:36
004 Classify Gaussian blurs.mp4

24:10
005 Examine feature map activations.mp4

27:50
006 CodeChallenge Softcode internal parameters.mp4

16:48
007 CodeChallenge How wide the FC.mp4

11:25
008 Do autoencoders clean Gaussians.mp4

17:10
009 CodeChallenge AEs and occluded Gaussians.mp4

09:37
010 CodeChallenge Custom loss functions.mp4

20:14
011 Discover the Gaussian parameters.mp4

16:58
012 The EMNIST dataset (letter recognition).mp4

24:59
013 Dropout in CNNs.mp4

10:14
014 CodeChallenge How low can you go.mp4

06:45
015 CodeChallenge Varying number of channels.mp4

13:39
016 So many possibilities! How to create a CNN.mp4

04:42

18 – Convolution and transformations

001 Convolution concepts.mp4

21:33
002 Feature maps and convolution kernels.mp4

09:32
003 Convolution in code.mp4

21:04
004 Convolution parameters (stride, padding).mp4

12:14
005 The Conv2 class in PyTorch.mp4

13:23
006 CodeChallenge Choose the parameters.mp4

07:10
007 Transpose convolution.mp4

13:41
008 Maxmean pooling.mp4

18:35
009 Pooling in PyTorch.mp4

13:43
010 To pool or to stride.mp4

09:47
011 Image transforms.mp4

16:57
012 Creating and using custom DataLoaders.mp4

19:05

17 – Running models on a GPU

01 – Introduction

15 – Weight inits and investigations

001 Explanation of weight matrix sizes.mp4

11:53
002 A surprising demo of weight initializations.mp4

15:52
003 Theory Why and how to initialize weights.mp4

12:46
004 CodeChallenge Weight variance inits.mp4

13:14
005 Xavier and Kaiming initializations.mp4

15:42
006 CodeChallenge Xavier vs. Kaiming.mp4

16:54
007 CodeChallenge Identically random weights.mp4

12:40
008 Freezing weights during learning.mp4

12:58
009 Learning-related changes in weights.mp4

21:54
010 Use default inits or apply your own.mp4

04:36

14 – FFN milestone projects

13 – Measuring model performance

12 – More on data

001 Anatomy of a torch dataset and dataloader.mp4

17:57
002 Data size and network size.mp4

16:35
003 CodeChallenge unbalanced data.mp4

20:05
004 What to do about unbalanced designs.mp4

07:45
005 Data oversampling in MNIST.mp4

16:30
006 Data noise augmentation (with devset+test).mp4

13:16
007 Data feature augmentation.mp4

19:40
008 Getting data into colab.mp4

06:05
009 Save and load trained models.mp4

06:14
010 Save the best-performing model.mp4

15:18
011 Where to find online datasets.mp4

05:32

11 – FFNs (Feed-Forward Networks)

001 What are fully-connected and feedforward networks.mp4

04:58
002 The MNIST dataset.mp4

12:33
003 FFN to classify digits.mp4

22:20
004 CodeChallenge Binarized MNIST images.mp4

05:24
005 CodeChallenge Data normalization.mp4

16:16
006 Distributions of weights pre- and post-learning.mp4

14:48
007 CodeChallenge MNIST and breadth vs. depth.mp4

12:35
008 CodeChallenge Optimizers and MNIST.mp4

07:06
009 Scrambled MNIST.mp4

08:00
010 Shifted MNIST.mp4

11:24
011 CodeChallenge The mystery of the missing 7.mp4

10:47
012 Universal approximation theorem.mp4

08:31

10 – Metaparameters (activations, optimizers)

013 CodeChallenge Predict sugar.mp4

17:06
024 How to pick the right metaparameters.mp4

11:47
023 Learning rate decay.mp4

12:15
022 CodeChallenge Adam with L2 regularization.mp4

07:42
021 CodeChallenge Optimizers and… something.mp4

06:57
020 Optimizers comparison.mp4

10:17
019 Optimizers (RMSprop, Adam).mp4

15:40
018 SGD with momentum.mp4

07:46
017 Optimizers (minibatch, momentum).mp4

18:41
016 More practice with multioutput ANNs.mp4

14:04
015 Loss functions in PyTorch.mp4

18:41
014 Loss functions.mp4

16:47
001 What are metaparameters.mp4

05:02
012 CodeChallenge Compare relu variants.mp4

07:48
011 Activation functions comparison.mp4

09:27
010 Activation functions in PyTorch.mp4

12:12
009 Activation functions.mp4

17:59
008 CodeChallenge Batch-normalize the qwerties.mp4

05:06
007 Batch normalization in practice.mp4

07:38
006 Batch normalization.mp4

13:15
005 The importance of data normalization.mp4

09:33
004 Data normalization.mp4

13:12
003 CodeChallenge Minibatch size in the wine dataset.mp4

15:38
002 The wine quality dataset.mp4

17:29

09 – Regularization

001 Regularization Concept and methods.mp4

13:38
002 train() and eval() modes.mp4

07:14
003 Dropout regularization.mp4

21:56
004 Dropout regularization in practice.mp4

23:13
005 Dropout example 2.mp4

06:33
006 Weight regularization (L1L2) math.mp4

18:25
007 L2 regularization in practice.mp4

13:24
008 L1 regularization in practice.mp4

12:22
009 Training in mini-batches.mp4

11:32
010 Batch training in action.mp4

10:47
011 The importance of equal batch sizes.mp4

06:59
012 CodeChallenge Effects of mini-batch size.mp4

11:57

08 – Overfitting and cross-validation

07 – ANNs (Artificial Neural Networks)

011 Linear solutions to linear problems.mp4

08:14
021 Reflection Are DL models understandable yet.mp4

08:26
019 CodeChallenge convert sequential to class.mp4

06:37
018 Model depth vs. breadth.mp4

20:31
017 Defining models using sequential vs. class.mp4

13:17
016 Depth vs. breadth number of parameters.mp4

17:25
015 Comparing the number of hidden units.mp4

09:59
014 CodeChallenge more qwerties!.mp4

11:56
013 Multi-output ANN (iris dataset).mp4

26:59
012 Why multilayer linear models don’t exist.mp4

06:20
001 The perceptron and ANN architecture.mp4

19:50
010 Multilayer ANN.mp4

19:51
009 Learning rates comparison.mp4

23:46
008 ANN for classifying qwerties.mp4

22:22
007 CodeChallenge manipulate regression slopes.mp4

18:58
006 ANN for regression.mp4

24:09
005 ANN math part 3 (backprop).mp4

12:10
004 ANN math part 2 (errors, loss, cost).mp4

10:54
003 ANN math part 1 (forward prop).mp4

16:22
002 A geometric view of ANNs.mp4

13:38

06 – Gradient descent

001 Overview of gradient descent.mp4

14:15
002 What about local minima.mp4

11:56
003 Gradient descent in 1D.mp4

17:11
004 CodeChallenge unfortunate starting value.mp4

11:30
005 Gradient descent in 2D.mp4

14:48
006 CodeChallenge 2D gradient ascent.mp4

05:16
007 Parametric experiments on g.d.mp4

18:56
008 CodeChallenge fixed vs. dynamic learning rate.mp4

15:33
009 Vanishing and exploding gradients.mp4

06:04
010 Tangent Notebook revision history.mp4

01:52

05 – Math, numpy, PyTorch

011 Entropy and cross-entropy.mp4

18:18
019 Derivatives product and chain rules.mp4

10:00
018 Derivatives find minima.mp4

08:32
017 Derivatives intuition and polynomials.mp4

16:38
016 The t-test.mp4

13:57
015 Reproducible randomness via seeding.mp4

08:37
014 Random sampling and sampling variability.mp4

11:18
013 Mean and variance.mp4

15:34
012 Minmax and argminargmax.mp4

12:47
002 Introduction to this section.mp4

02:06
010 Logarithms.mp4

08:26
009 Softmax.mp4

19:26
008 Matrix multiplication.mp4

15:27
007 OMG it’s the dot product!.mp4

09:45
006 Vector and matrix transpose.mp4

06:58
005 Converting reality to numbers.mp4

06:33
004 Terms and datatypes in math and computers.mp4

07:05
003 Spectral theories in mathematics.mp4

09:15

04 – About the Python tutorial

03 – Concepts in deep learning

02 – Download all course materials

Student Ratings & Reviews

No Review Yet

About Course

Course Content

16 – Autoencoders

001 What are autoencoders and what do they do.mp4

002 Denoising MNIST.mp4

003 CodeChallenge How many units.mp4

004 AEs for occlusion.mp4

005 The latent code of MNIST.mp4

006 Autoencoder with tied weights.mp4

31 – Python intro Text and plots

001 Printing and string interpolation.mp4

002 Plotting dots and lines.mp4

003 Subplot geometry.mp4

004 Making the graphs look nicer.mp4

005 Seaborn.mp4

006 Images.mp4

007 Export plots in low and high resolution.mp4

30 – Python intro Flow control

001 If-else statements.mp4

002 If-else statements, part 2.mp4

003 For loops.mp4

004 Enumerate and zip.mp4

005 Continue.mp4

006 Initializing variables.mp4

007 Single-line loops (list comprehension).mp4

008 while loops.mp4

009 Broadcasting in numpy.mp4

010 Function error checking and handling.mp4

29 – Python intro Functions

001 Inputs and outputs.mp4

002 Python libraries (numpy).mp4

003 Python libraries (pandas).mp4

004 Getting help on functions.mp4

005 Creating functions.mp4

006 Global and local variable scopes.mp4

007 Copies and referents of variables.mp4

008 Classes and object-oriented programming.mp4

28 – Python intro Indexing, slicing

001 Indexing.mp4

002 Slicing.mp4

27 – Python intro Data types

001 How to learn from the Python tutorial.mp4

002 Variables.mp4

003 Math and printing.mp4

004 Lists (1 of 2).mp4

005 Lists (2 of 2).mp4

006 Tuples.mp4

007 Booleans.mp4

008 Dictionaries.mp4

26 – Where to go from here

001 How to learn topic _X_ in deep learning.mp4

002 How to read academic DL papers.mp4

25 – Ethics of deep learning

001 Will AI save us or destroy us.mp4

002 Example case studies.mp4

003 Some other possible ethical scenarios.mp4

004 Will deep learning take our jobs.mp4

005 Accountability and making ethical AI.mp4

24 – RNNs (Recurrent Neural Networks) (and GRULSTM)

001 Leveraging sequences in deep learning.mp4

002 How RNNs work.mp4

003 The RNN class in PyTorch.mp4

004 Predicting alternating sequences.mp4

005 CodeChallenge sine wave extrapolation.mp4

006 More on RNNs Hidden states, embeddings.mp4

007 GRU and LSTM.mp4

008 The LSTM and GRU classes.mp4

009 Lorem ipsum.mp4

23 – Generative adversarial networks

001 GAN What, why, and how.mp4

002 Linear GAN with MNIST.mp4

003 CodeChallenge Linear GAN with FMNIST.mp4

004 CNN GAN with Gaussians.mp4

005 CodeChallenge Gaussians with fewer layers.mp4

006 CNN GAN with FMNIST.mp4

007 CodeChallenge CNN GAN with CIFAR.mp4

22 – Style transfer

001 What is style transfer and how does it work.mp4

002 The Gram matrix (feature activation covariance).mp4

003 The style transfer algorithm.mp4