Courses

Machine Learning

---

In this course you will understand the basics of Machine Learning techniques and apply them. You will learn the concept of statistical learning vs. traditional programming, and the differences between AI, Machine and Learning. The course covers supervised learning: regression and classification (linear and non linear), and some unsupervised techniques: clustering and dimensionality reduction. The code will be based on Python and Keras.

Pre-requisities

• Python

• Probability

• Linear Algebra

Session details

#	Session	Duration (hrs)
1	Introduction to Statistical and Machine Learning	3
2	Linear Regression	3
3	Linear Classification: Logistic Regression	3
4	Evaluation protocols and ML process	3
5	Linear Classification: Large-margin classifiers (SVM) and Multi-class classification	3
6	Non linear Classifiers: Kernel SVMs and Neural Networks	3
7	Meta Algorithms: Ensemble methods, Boosting and BAgging	3
8	Unsupervised learning 1: Clustering	3
9	Unsupervised learning 2: Dimensionality reduction	3

What you will learn

• Traditional programming vs. Statistical learning

• AI vs. Machine learning vs. Deep learning

• Different ML types: Supervised learning vs. Unsupervised learning vs. Self-supervised learning vs. Reinforcement learning

• Linear Regression

• Linear Classifiers

• Non-Linear Classifiers

• Evaluation protocols

• Meta-Algorithms

• Unsupervised learning: clustering and dimensionality reduction

Enroll Now

Close

Deep Learning

---

In this course you will understand the basics of Deep Learning and Deep Neural Networks and apply them. You will learn the the basic ingredients of Deep Learning frameworks, and apply DNNs to basic examples. The code will be based on Python and Keras.

Pre-requisities

• Python

• Probability

• Linear Algebra

Session details

#	Session	Duration (hrs)
1	Introduction to AI, Machine Learning and Deep Learning	3
2	The mathematical Building Blocks of Neural Networks	3
3	Optimization algorithms and Backpropagation	3
4	Introduction to DL frameworks (Keras, TensorFlow, Pytorch)	3
5	Evaluation protocols and ML process	3

What you will learn

• Traditional programming vs. Statistical learning

• AI vs. Machine learning vs. Deep learning

• Different ML types: Supervised learning vs. Unsupervised learning vs. Self-supervised learning vs. Reinforcement learning

• The mathematical building blocks of Neural Nets

• The basics of DL programming and frameworks

• Evaluation protocols and DL/ML process

Enroll Now

Close

Deep Learning for NLP

---

In this course you will understand the basics of Deep Learning application in Natural Language Processing. You will learn about word, sentence and document representations, word vectors and word embeddings, Neural Language Modeling and Text classification with Recurrent Neural Networks and Convolutional Neural Networks. The code will be based on Python and Keras.

Pre-requisities

• Python

• Probability

• Linear Algebra

• Deep Learning

Session details

#	Session	Duration (hrs)
1	Introduction to Deep Learning in NLP	3
2	Word vectors 1: Embeddings and Word2Vec	3
3	Word vectors 2: Coocurrence matrix, SVD and GloVe	3
4	Language Modeling	3
5	Recurrent Neural Networks and Neural Language Models (NLM)	3
6	Text classification with RNNs and text pre-processing	3
7	Text classification with ConvNets and word vectors	3

What you will learn

• Word embeddings and Embeddings

• Sentence and document representations

• Language modeling and Recurrent Neural Nets

• Text classification with RNN and ConvNets

Enroll Now

Close

Deep Learning for Computer Vision

---

In this course you will understand the basics of Deep Learning application in Computer Vision. You will learn about Convolutional Neural Nets, and various ConvNets architectures for computer vision. Also, you will learn practical aspects like: data augmentation and how to use ConvNets with small datasets, applying transfer learning and debugging ConvNets by visualizing the learnt features. The code will be based on Python and Keras.

Pre-requisities

• Python

• Probability

• Linear Algebra

• Deep Learning

Session details

#	Session	Duration (hrs)
1	Introduction to Deep Learning in Computer Vision	3
2	Convolutional Neural Networks	3
3	ConvNets on small datasets	3
4	ConvNets Meta-architectures, Pre-trained Nets and Transfer learning	3
5	Visualization of ConvNets	3

What you will learn

• The basics of ConvNets in computer vision

• The practical aspects of DL in CV, like data augmentation and transfer learning

• ConvNets Architectures and Pre-trained ConvNets

• Debugging ConvNets by visualization of ConvNets filters and features

Enroll Now

Close

Computer Vision Applications with Deep Learning

---

In this course you will get introduced to some important Computer Vision applications using ConvNets. You will learn about Semantic Segmentation, Single and Multiple Object detection. The code will be based on Python and Keras.

Pre-requisities

• Python

• Probability

• Linear Algebra

• Deep Learning

• Basics of Computer Vision

• Convolutional Neural Networks

Session details

#	Session	Duration (hrs)
1	Review of ConvNets Architectures	3
2	Semantic Segmentation with Deep Learning	3
3	Signle Object Detection	1
4	Multiple Object Detection	3
5	Video processing and Spatio-Temporal moddels	3
6	Introduction to 3D Deep Learning	1

What you will learn

• Signle and Multiple Object Detection with Deep Learning

• Semantic Segmentation with Deep Learning

• Video handling and Spatio-Temporal models

• 3D Deep Learning

Enroll Now

Close

NLP Applications with Deep Learning

---

In this course you will get introduced to some important NLP applications using ConvNets. You will learn about Neural Machine Translation using seq2seq models, attention models and Transformers. You will also learn about modern Neural architectures and transfer learning in NLP like ULMFiT, BERT, XLTransformers and GPT. The code will be based on Python and Keras.

Pre-requisities

• Python

• Probability

• Linear Algebra

• Deep Learning

• Basics of NLP

• Convolutional Neural Networks

• Recurrent Neural Networks

Session details

#	Session	Duration (hrs)
1	Neural Machine Translation with seq2seq models	3
2	Full Attention models and Transformers	3
3	Transfer learning in NLP: ULMFiT	3
4	Modern Neural NLP: BERT, GPT and XLTransformers	3

What you will learn

• Neural Machine Translation and seq2seq models

• Attention Mechanisms and Transformer networks

• Modern NLP networks: BERT, XLTrans and BERT

Enroll Now

Close

Deep Reinforcement Learning

---

In this course you will get introduced to Reinforcement Learning. We will start by the basic formulations of RL problems for finite state space using Markov Decision Process (MDP). Then we will move to the continous state space using Deep Reinforcement Learning. We will discuss Q-Learning, DQN, Policy Gradients, DDPG, Actor Critic, A3C and others.

Pre-requisities

• Python

• Probability

• Linear Algebra

• Deep Learning

Session details

#	Session	Duration (hrs)
1	Reinforcement Learning Problem Definition	1
2	Markov Decision Process Framework (MDP): Dynamic Programming solutions and Bellman equation, Policy and Value iteration algorithms	1
3	Model-based vs. Model-free RL	30min
4	TD Learning	30min
5	Exploration vs. Exploitation	15min
6	Q-Learning	30min
7	SARSA	30min
8	On-policy vs Off-policy	15min
9	Deep Reinforcement Learning and DQN	30min
10	Policy Gradients and DDPG	1
11	REINFORCE algorithm	30min
12	Actor Critic Algorithms: AC, A2C, A3C	1
13	Imitation Learning: Inverse RL (IRL), Apprenticeship Learning	1
14	RL vs. Optimal Control	30min

What you will learn

• RL problem space and solutions

• Deep RL problem space and solutions

Close

Deployment of Deep Learning Models

---

In this course you will learn about the different deployment scenarios of deep learning models. We will cover client side scenarios including Mobile, Browser and Embedded devices like Rasperry Pi, and also client-server scenarios using API's like TFHub, Torch-hub and Hugging face. We will demonstrate the scenarios for different applications like image classification, multiple objects detection and semantic segmentation.

Pre-requisities

• Python

• Probability

• Linear Algebra

• Deep Learning

• Computer Vision

• NLP

Session details

#	Session	Duration (hrs)
1	Deployment scenarios overview of DL models	1
2	Network compression overview	1
3	Pruning	1
4	Knowledge distillation	1
5	Quantization	1
6	Model compression pipelines	1
7	Optimized and light weight architectures (MobileNet, EfficientNet...etc)	3
8	TFLite overview	1
9	TFLite for Mobile Devices	1
10	TFLite for Edge and Embedded Devices (Rasperry Pi)	1
11	TFJs overview for browser deployment	1
12	AWS and Django deployment of Computer Vision server based models: image classification, object detection, semantic segmentation	1
13	Muli-object detection deployment from scratch and finetuning scernarios: Client side (TFJS, TFLite, Native)	1

What you will learn

• Identify the DL deployment scenarios and solutions space to choose the best fitting one.

• Practice the different deployment scenarios in client, server and edge devices.

• Apply what you have learnt for practical problems like image classification, multiple object detection and semantic segmentation

Enroll Now

Close

Transfomers in Computer Vision

---

In this course you will learn about the state-of-the art Transformer networks, and their application in Computer Vision. We will start by introducing the idea of attention as a general DL concept, then move to the Transformer network in NLP. From there, we discuss the application of Transformers in CV, for object detection. We compare and visualize the DEtection TRansformer (DETR) network vs. YOLOv5 as a ConvNet-based architecture. We then extend our discussion to the use of Transformers in image classification using Vision Transfer (ViT) and Semantic Segmentation (SETR).

Pre-requisities

• Python

• Probability

• Linear Algebra

• Deep Learning

• Computer Vision

Session details

#	Session	Duration (hrs)
1	Attention as a General DL concept	1
2	Transformers in NLP	1
3	Transoformers in Computer Vision: DETR example for traffice signs object detection	3
4	Transformers for Image Classification: Vision Transfer (ViT)	1
5	Transformers for Semantic Segmentation (SETR)	1

What you will learn

• Deep understanding of attention mechanisms.

• Understand the state-of-the art Transformer networks in CV for all the major 3 CV tasks: classification, detection and segmentation.

• Apply what you have learnt for practical problems like image classification, multiple object detection and semantic segmentation

Close

Computer Vision Engineer Bootcamp

---

This bootcamp is a condensed and focused series of sessions that prepares the candidate to become a professional Computer Vision Engineer, by following a well-designed, project-based learning track. By the end of the bootcamp, the candidate will have strong practical knowledge, and also a rich set of projects that enrich her/his portfolio.

Pre-requisities

• Python

• Probability

• Linear Algebra

• Deep Learning

Session details

#	Session	Duration
1	Basics of Computer Vision and ConvNet	1-day
2	Practical ConvNets: Transfer Learning, Data augmentation and Visualization	1-day
3	Computer Vision Applications: Image Classification, Object Detection and Semantic Segmentation	1-day
4	Optimized ConvNet architectures: MobileNet, SqueezeNet, ShuffleNet, EfficientNet	1-day
5	Advanced Deployment Scenarios: Client (Browser, Edge, Mobile) and Server (AWS, TFHub, Torchhub)	1-day

What you will learn

• Deep understanding of computer vision and ConvNets.

• Deep understanding of Practical deployment of ConvNets.

• Project-based practice of different deployment scenarios (Client: Edge, Mobile and Browser and Server) of ConvNets.

Enroll Now

Close

NLP Engineer Bootcamp

---

In this course you will learn about the be qualified to be a professional NLP Engineer through a condensed series of project-based sessions. The bootcamp covers the basics of traditional NLP and text preprocessing, DL in NLP, NLP applications like text classification, machine translation, transfomer networks and transfer learning in NLP (BERT, GPT,...etc).

Pre-requisities

• Python

• Probability

• Linear Algebra

• Deep Learning

Session details

#	Session	Duration
1	DL in NLP and Text preprocessing	1-day
2	Word Embeddings	1-day
3	Recommender systems	3hrs
4	Sequence models: Text Classification, Language Models	1-day
5	Sequence-to-seqeunce models: Machine Translation	1-day
6	Transfomers and Transfer Learning in NLP (BERT, GPT-1,2,3)	1-day

What you will learn

• Deep understanding of DL in NLP and how to deal with text data

• Deep understanding of sequence models like RNN/LSTM/GRU and their applications in text classification and language modeling

• Build seq2seq models for machine translation

• Understand Transformer models and attention and use them to build state-of-the art machine translation system, chatbot and QA system.

• Understand Transfer Learning application in NLP, and use pre-trained models like BERT and GPT to build text classifier and chatbots.

Enroll Now

Close

Graph Neural Networks

---

Coming Soon