Autoplay
Autocomplete
Previous Lesson
Complete and Continue
Artificial Intelligence, Deep Learning and Computer Vision Masterclass
Introduction
Introduction (2:31)
Environment Setup
Installing PyCharm and Python on Windows
Installing PyCharm and Python on Mac
Installing TensorFlow and Keras (2:02)
Artificial Intelligence Basics
Why to learn artificial intelligence and machine learning? (5:24)
Types of artificial intelligence learning (8:29)
### MACHINE LEARNING ###
Machine learning algorithms
Linear Regression
What is linear regression? (8:52)
Linear regression theory - optimization (8:00)
Linear regression theory - gradient descent (7:23)
Linear regression implementation I (14:11)
Linear regression implementation II (4:24)
Mathematical formulation of linear regression
Logistic Regession
What is logistic regression? (11:54)
Logistic regression and maximum likelihood estimation (4:58)
Logistic regression example I - sigmoid function (11:29)
Logistic regression example II- credit scoring (10:53)
Logistic regression example III - credit scoring (5:51)
Mathematical formulation of logistic regression
Cross Validation
What is cross validation? (5:57)
Cross validation example (5:17)
K-Nearest Neighbor Classifier
What is the k-nearest neighbor classifier? (5:52)
Concept of lazy learning (3:33)
Distance metrics - Euclidean-distance (6:47)
Bias and variance trade-off (3:51)
K-nearest neighbor implementation I (7:10)
K-nearest neighbor implementation II (8:45)
K-nearest neighbor implementation III (4:12)
Mathematical formulation of k-nearest neighbor classifier
Naive Bayes Classifier
What is the naive Bayes classifier? (10:03)
Naive Bayes classifier illustration (4:24)
Naive Bayes classifier implementation (4:01)
What is text clustering? (8:55)
Text clustering - inverse document frequency (TF-IDF) (4:42)
Naive Bayes example - clustering news (14:23)
Mathematical formulation of naive Bayes classifier
Support Vector Machine (SVM)
What are Support Vector Machines (SVMs)? (5:19)
Linearly separable problems (14:10)
Non-linearly separable problems (6:32)
Kernel functions (9:49)
Support vector machine example I - simple (10:50)
Support vector machine example II - iris dataset (6:31)
Support vector machines example III - parameter tuning (7:17)
Support vector machine example IV - digit recognition (10:04)
Support vector machine example V - digit recognition (5:43)
Advantages and disadvantages (2:32)
Mathematical formulation of Support Vector Machines (SVMs)
Decision Trees
Decision trees introduction - basics (7:43)
Decision trees introduction - entropy (8:56)
Decision trees introduction - information gain (7:51)
The Gini-index approach (10:00)
Decision trees introduction - pros and cons (2:31)
Decision trees implementation I (5:50)
Decision trees implementation II - parameter tuning (4:28)
Decision tree implementation III - identifying cancer (4:49)
Mathematical formulation of decision trees
Random Forest Classifier
Pruning introduction (6:56)
Bagging introduction (7:49)
Random forest classifier introduction (5:36)
Random forests example I - iris dataset (4:14)
Random forests example II - credit scoring (3:29)
Random forests example III - OCR parameter tuning (9:49)
Mathematical formulation of random forest classifiers
Boosting
Boosting introduction - basics (4:18)
Boosting introduction - illustration (5:40)
Boosting introduction - equations (7:13)
Boosting introduction - final formula (8:27)
Boosting implementation I - iris dataset (6:29)
Boosting implementation II -wine classification (11:46)
Boosting vs. bagging (3:08)
Mathematical formulation of boosting
Principal Component Analysis (PCA)
Principal component analysis (PCA) introduction (8:18)
Principal component analysis example (10:38)
Principal component analysis example II (9:25)
Mathematical formulation of principle component analysis (PCA)
Clustering
K-means clustering introduction (9:57)
K-means clustering example (7:49)
K-means clustering - text clustering (9:01)
DBSCAN introduction (6:25)
DBSCAN example (8:43)
Hierarchical clustering introduction (6:33)
Hierarchical clustering example (8:10)
Hierarchical clustering - market segmentation (9:34)
Mathematical formulation of clustering
Computer Vision - Face Detection
Computer vision introduction (3:49)
Viola-Jones algorithm (10:53)
Haar-features (8:31)
Integral images (6:22)
Boosting in computer vision (6:18)
Cascading (4:13)
Face detection implementation I - installing OpenCV (2:50)
Face detection implementation II - CascadeClassifier (9:58)
Face detection implementation III - CascadeClassifier parameters (4:06)
Face detection implementation IV - tuning the parameters (4:51)
Face detection implementation V - detecting faces real-time (5:22)
Machine Learning Project I - Face Recognition
The Olivetti dataset (7:54)
Understanding the dataset (6:10)
Finding optimal number of principal components (eigenvectors) (6:17)
Understanding "eigenfaces" (8:01)
Constructing the machine learning models (4:33)
Using cross-validation (2:53)
### NEURAL NETWORKS AND DEEP LEARNING ###
Neural networks and deep learning algorithms
Feed-Forward Neural Network Theory
Artificial neural networks - inspiration (5:20)
Artificial neural networks - layers (4:42)
Artificial neural networks - the model (5:09)
Why to use activation functions? (6:46)
Neural networks - the big picture (9:07)
Using bias nodes in the neural network (1:44)
How to measure the error of the network? (4:46)
Optimization with gradient descent (8:28)
Gradient descent with backpropagation (6:33)
Backpropagation explained (12:15)
Mathematical formulation of feed-forward neural networks
Single Layer Networks Implementation
Simple neural network implementation - XOR problem (12:56)
Simple neural network implementation - Iris dataset (13:26)
Credit scoring with simple neural networks (4:24)
Deep Learning
Types of neural networks (3:51)
Deep Neural Networks Theory
Deep neural networks (5:15)
Activation functions revisited (9:44)
Loss functions (5:57)
Gradient descent and stochastic gradient descent (7:17)
Hyperparameters (5:14)
Mathematical formulation of deep neural networks
Deep Neural Networks Implementation
Deep neural network implementation I (6:30)
Deep neural network implementation II (6:26)
Deep neural network implementation III (4:45)
Multiclass classification implementation I (7:48)
Multiclass classification implementation II (5:27)
Machine Learning Project II - Smile Detector
Understanding the classification problem (2:14)
Reading the images and constructing the dataset I (6:13)
Reading the images and constructing the dataset II (4:45)
Building the deep neural network model (3:23)
Evaluating and testing the model (3:32)
Convolutional Neural Networks (CNNs) Theory
Convolutional neural networks basics (6:04)
Feature selection (4:11)
Convolutional neural networks - kernel (4:16)
Convolutional neural networks - kernel II (5:42)
Convolutional neural networks - pooling (5:47)
Convolutional neural networks - flattening (4:59)
Convolutional neural networks - illustration (2:38)
Mathematical formulation of convolution neural networks
Convolutional Neural Networks (CNNs) Implementation
Handwritten digit classification I (11:12)
Handwritten digit classification II (12:09)
Handwritten digit classification III (5:32)
Machine Learning Project III - Identifying Objects with CNNs
What is the CIFAR-10 dataset? (6:30)
Preprocessing the data (2:46)
Fitting the model (5:34)
Tuning the parameters - regularization (9:21)
Recurrent Neural Networks (RNNs) Theory
Why do recurrent neural networks are important? (4:30)
Recurrent neural networks basics (8:58)
Vanishing and exploding gradients problem (9:22)
Long-short term memory (LSTM) model (10:55)
Gated recurrent units (GRUs) (3:23)
Mathematical formulation of recurrent neural networks
Recurrent Neural Networks (RNNs) Implementation
Time series analysis example I (4:01)
Time series analysis example II (5:11)
Time series analysis example III (6:06)
Time series analysis example IV (2:33)
Time series analysis example V (3:59)
Time series analysis example VI (3:58)
Reinforcement Learning
What is reinforcement learning?
Applications of reinforcement learning (2:44)
Markov Decision Process (MDP) Theory
Markov decision processes basics I (5:38)
Markov decision processes basics II (6:21)
Markov decision processes - equations (12:00)
Markov decision processes - illustration (7:49)
Bellman-equation (5:41)
How to solve MDP problems? (2:20)
What is value iteration? (6:28)
What is policy iteration? (3:52)
Mathematical formulation of reinforcement learning
Exploration vs. Exploitation Problem
Exploration vs exploitation problem (3:29)
N-armed bandit problem introduction (8:46)
N-armed bandit problem implementation (11:12)
Applications: A/B testing in marketing (4:11)
Q Learning Theory
What is Q learning? (5:44)
Q learning introduction - the algorithm (7:08)
Q learning illustration (11:06)
Mathematical formulation of Q learning
Q Learning Implementation (Tic Tac Toe)
Tic tac toe with Q learning implementation I (3:45)
Tic tac toe with Q learning implementation II (7:38)
Tic tac toe with Q learning implementation III (7:26)
Tic tac toe with Q learning implementation IV (7:27)
Tic tac toe with Q learning implementation V (4:56)
Tic tac toe with Q learning implementation VI (12:06)
Tic tac toe with Q learning implementation VII (6:23)
Tic tac toe with Q learning implementation VIII (6:39)
Deep Q Learning Theory
What is deep Q learning? (4:47)
Deep Q learning and ε-greedy strategy (3:11)
Remember and replay (3:34)
Mathematical formulation of deep Q learning
Deep Q Learning Implementation (Tic Tac Toe)
Tic Tac Toe with deep Q learning implementation I (3:53)
Tic Tac Toe with deep Q learning implementation II (6:27)
Tic Tac Toe with deep Q learning implementation III (10:39)
Tic Tac Toe with deep Q learning implementation IV (5:05)
Tic Tac Toe with deep Q learning implementation V (4:43)
### COMPUTER VISION ###
Computer vision algorithms
History of Computer Vision
Evolution of computer vision related algorithms (3:36)
Handling Images and Pixels
Images and pixel intensities (5:13)
Handling pixel intensities I (6:18)
Handling pixel intensities II (5:16)
Why convolution is so important in image processing? (12:07)
Image processing - blur operation (5:08)
Image processing - edge detection kernel (5:38)
Image processing - sharpen operation (3:43)
Computer Vision Project I - Lane Detection Problem (Self-Driving Cars)
Lane detection - the problem (1:44)
Lane detection - handling videos (5:36)
Lane detection - first transformations (4:21)
What is Canny edge detection? (6:32)
Getting the useful region of the image - masking (13:20)
Detecting lines - what is Hough transformation? (10:30)
Drawing lines on video frames (9:05)
Testing lane detection algorithm (2:19)
Histogram of Oriented Gradients (HOG) Algorithm Theory
Histogram of oriented gradients basics (3:48)
Histogram of oriented gradients - gradient kernel (6:55)
Histogram of oriented gradients - magnitude and angle (7:42)
Histogram of oriented gradients - normalization (4:46)
Histogram of oriented gradients - big picture (3:14)
Histogram of Oriented Gradients (HOG) Implementation
Showing the HOG features programatically (10:31)
Face detection with HOG implementation I (6:12)
Face detection with HOG implementation II (13:03)
Face detection with HOG implementation III (5:32)
Face detection with HOG implementation IV (7:30)
Convolutional Neural Networks (CNNs) Based Approaches
The standard convolutional neural network (CNN) way (5:57)
Region proposals and convolutional neural networks (CNNs) (9:41)
Detecting bounding boxes with regression (6:32)
What is the Fast R-CNN model? (2:30)
What is the Faster R-CNN model? (1:49)
You Only Look Once (YOLO) Algorithm Theory
What is the YOLO approach? (5:26)
YOLO algorithm - grid cells (6:45)
YOLO algorithm - intersection over union (9:02)
How to train the YOLO algorithm? (7:21)
YOLO algorithm - loss function (4:53)
YOLO algorithm - non-max suppression (2:52)
Why to use the so-called anchor boxes? (6:10)
You Only Look Once (YOLO) Algorithm Implementation
YOLO algorithm implementation I (6:15)
YOLO algorithm implementation II (9:08)
YOLO algorithm implementation III (8:45)
YOLO algorithm implementation IV (12:47)
YOLO algorithm implementation V (12:07)
YOLO algorithm implementation VI (1:47)
YOLO algorithm implementation VII (3:47)
Single-Shot MultiBox Detector (SSD) Theory
What is the SSD algorithm? (3:50)
Basic concept behind SSD algorithm (architecture) (7:17)
Bounding boxes and anchor boxes (10:16)
Feature maps and convolution layers (4:40)
Hard negative mining during training (2:22)
Regularization (data augmentation) and non-max suppression during training (2:14)
SSD Algorithm Implementation
SSD implementation I (6:06)
SSD implementation II (2:15)
SSD implementation III (5:10)
SSD implementation IV (7:56)
SSD implementation V (2:57)
Course Materials
Slides
Source code
Datasets
Course Materials (DOWNLOADS)
Source code and slides
Why convolution is so important in image processing?
Lesson content locked
If you're already enrolled,
you'll need to login
.
Enroll in Course to Unlock