Deep Learning Mega Course
Introduction to Deep Learning
•
• Introduction to Theano
• Introduction to TensorFlow
• Introduction to Keras
• Project: Develop Large Models on GPUs Cheaply In the Cloud
•
Multilayer Perceptrons
•
• Develop Your First Neural Network With Keras
• Evaluate The Performance of Deep Learning Models
• Use Keras Models With Scikit-Learn For General Machine Learning
• Project: Multiclass Classification Of Flower Species
• Project: Binary Classification Of Sonar Returns
• Project: Regression Of Boston House Prices
• Advanced Multilayer Perceptrons and Keras
• Reduce Overfitting With Dropout Regularization
• Lift Performance With Learning Rate Schedules
RNN & Long Short-Term Memory(LSTM)
Networks With Python
Develop Sequence Prediction
Models With Deep Learning
1-Foundations
What are LSTMs
Sequence Prediction Problems
Limitations of Multilayer Perceptrons
Promise of Recurrent Neural Networks
The Long Short-Term Memory Network
2- How to Train LSTMs
2.1 Backpropagation Training Algorithm
2.2 Unrolling Recurrent Neural Networks
2.3 Configurations for Truncated BPTT
3- How to Prepare Data for LSTMs
4- How to Develop LSTMs in Keras
5- Models for Sequence Prediction
6- Models
7- How to Develop Vanilla LSTMs
7.1 Echo Sequence Prediction Problem
8- How to Develop Stacked LSTMs
8.1 Damped Sine Wave Prediction Problem
9- How to Develop CNN LSTMs
Moving Square Video Prediction Problem
10- How to Develop Encoder-Decoder LSTMs
11- How to Develop Bidirectional LSTMs
12- How to Develop Generative LSTMs
13- Advanced
13.1- How to Diagnose and Tune LSTMs
13.2 Diagnosing Underfitting and Overfitting
14- How to Make Predictions with LSTMs
14.1 Finalize a LSTM Model
14.2 Save LSTM Models to File
14.3 Make Predictions on New Data
15- How to Update LSTM Models
Deep Learning for Computer Vision
(CNN- Convolution Neural Network)
Introduction to Computer Vision
Image Data Preparation
How to Load and Manipulate Images With PIL/Pillow
Manually Scale Image Pixel Data
Load and Manipulate Images with Keras
Scale Image Pixel Data with Keras
Load Large Datasets From Directories with Keras
Image Data Augmentation in Keras
Use Different Color Channel Ordering Formats
Convolutions and Pooling
Filter Size, Padding, and Stride
ImageNet, ILSVRC, and Milestone Architectures
Milestone Model Architectural Innovations Work
LeNet-5
AlexNet
VGG
Inception and GoogLeNet
Residual Network or ResNet
Pre-Trained Models and Transfer Learning
Image Classification
Classify Small Photos of Objects
Classify Photographs of Dogs and Cats
Label Satellite Photographs of the Amazon Rainforest
Object Detection
R-CNN Model Family
YOLO Model Family
Mask R-CNN
Face Recognition
Detect Faces in Photographs
Face Detection With OpenCV
Face Identification and Verification with VGGFace2
Face Classification with FaceNet
Deep Learning
For
Natural Language Processing
1- Foundations
1.1 Natural Language Processing
2- Deep Learning
2.1 - Deep Learning for Natural Language
3- Develop Deep Learning Models with Keras
3.1 Keras Model Life-Cycle
3.2 Keras Functional Models
4- Data Preparation
4.1 How to Clean Text Manually and with NLTK
4.2 Metamorphosis by Franz Kafka
4.3 Tokenization and Cleaning with NLTK
4.4 Additional Text Cleaning
5- Prepare Text Data with scikit-learn
5.1 The Bag-of-Words Model
5.2 Word Counts with CountVectorizer
5.3 Word Frequencies with TfidfVectorizer
5.4 Hashing with HashingVectorizer
6- Prepare Text Data with Keras
6.1 Split Words with text to word sequence
6.2 Encoding with one hot
6.3 Hash Encoding with hashing trick
6.4 Tokenizer API
7- Bag-of-Words
8 The Bag-of-Words Model
8.4 Example of the Bag-of-Words Model
8.5 Managing Vocabulary
8.6 Scoring Words
9 - Prepare Movie Review Data for Sentiment Analysis
10- Project: Develop a Neural Bag-of-Words Model for Sentiment Analysis
11- Word Embeddings
11.1 How to Develop Word Embedding with Genism
11.2 Genism Python Library
11.3 Develop Word2Vec Embedding
11.4 Visualize Word Embedding
11.5 Load Google's Word2Vec Embedding
11.6 Load Stanford's GloVe Embedding
11.7 Learn and Load Word Embeddings in Keras
12 - Text Classification
12.1 Neural Models for Document Classification
12.2 Word Embeddings + CNN = Text Classification
13- Project: Develop an Embedding + CNN Model for Sentiment Analysis
14- Project: Develop an n-gram CNN Model for Sentiment Analysis
15- Language Modeling
15.1 Neural Language Modeling
15.2 Statistical Language Modeling
15.3 Neural Language Models
16- Develop a Character-Based Neural Language Model
16.1 Sing a Song of Sixpence
16.2 Generate Text
16.3 Develop a Word-Based Neural Language Model
16.4 Jack and Jill Nursery Rhyme
16.5 Model 1: One-Word-In, One-Word-Out Sequences
16.6 Model 2: Line-by-Line Sequence
16.7 Model 3: Two-Words-In, One-Word-Out Sequence
17- Project: Develop a Neural Language Model for Text Generation
17.1 The Republic by Plato
18- Image Captioning
18.1 Neural Image Caption Generation
18.2 Describing an Image with Text
18.3 Encoder-Decoder Architecture
19- Neural Network Models for Caption Generation
19.1 Image Caption Generation
19.2 Load and Use a Pre-Trained Object Recognition Model
19.3 ImageNet
19.4 The Oxford VGG Models
19.5 Load the VGG Model in Keras
19.5 Develop a Simple Photo Classifier
20- BLEU Score
20.1 Bilingual Evaluation Understudy Score
20.2 Calculate BLEU Scores
20.3 Cumulative and Individual BLEU Scores
20.4 Prepare a Photo Caption Dataset for Modeling
21- Project: Develop a Neural Image Caption Generation Model
21.1- Machine Translation
21.2 Statistical Machine Translation
21.3 Neural Machine Translation
21.4 Encoder-Decoder Models for Neural Machine Translation
21.5 Encoder-Decoder Architecture for NMT
21.6 Sutskever NMT Model
21.7 Cho NMT Model
22- Configure Encoder-Decoder Models for Machine Translation
22.1 Encoder-Decoder Model for Neural Machine Translation
23- Project: Develop a Neural Machine Translation Model
23.1 German to English Translation Dataset
Deep Learning – Time Series Forecasting
Multilayer Perceptrons for Time Series
Convolutional Neural Networks for Time Series
Recurrent Neural Networks for Time Series
Taxonomy of Time Series Forecasting Problems
Inputs vs. Outputs
Endogenous vs. Exogenous
Regression vs. Classication
Unstructured vs. Structured
Univariate vs. Multivariate
Single-step vs. Multi-step
Static vs. Dynamic
How to Transform Time Series to a Supervised Learning Problem
Sliding Window
Sliding Window With Multiple Variates
Review of Simple and Classical Forecasting Methods
Simple Forecasting Methods
Autoregressive Methods
Exponential Smoothing Methods
How to Develop ETS Models for Univariate Forecasting
How to Develop SARIMA Models for Univariate Forecasting
How to Develop MLPs, CNNs and LSTMs for Univariate Forecasting
How to Grid Search Deep Learning Models for Univariate Forecasting
Multi-step Forecasting
How to Load and Explore Household Energy Usage Data
How to Develop ARIMA Models for Multi-step Energy Usage Forecasting
How to Develop CNNs for Multi-step Energy Usage Forecasting
How to Develop LSTMs for Multi-step Energy Usage Forecasting
Time Series Classication
Deep Learning Models for Human Activity Recognition
How to Develop ML Models for Human Activity Recognition
How to Develop CNNs for Human Activity Recognition
How to Develop LSTMs for Human Activity Recognition
Generative Adversarial Network – GAN
Upsample with Convolutional Neural Networks
GAN Training Algorithm
GAN Hacks to Train Stable Models
Develop a 1D GAN from Scratch
Discriminator Model
Generator Model
DCGAN for Grayscale Handwritten Digits
Develop a DCGAN for Small Color Photographs
the Latent Space When Generating Faces
Identify and Diagnose GAN Failure Modes
Evaluate Generative Adversarial Networks
Implement the Inception Score
Frechet Inception Distance
Different GAN Loss Functions
Least Squares GAN (LSGAN)
Wasserstein GAN (WGAN)
Conditional GAN (cGAN)
Information Maximizing GAN (InfoGAN)
Auxiliary Classifier GAN (AC-GAN)
Semi-Supervised GAN (SGAN)
Image Translation
Pix2Pix Models
CycleGAN
Advanced GANs
BigGAN
Progressive Growing GAN
StyleGAN
Reinforcement Learning
What is Reinforcement Learning
supervised, unsupervised, and reinforcement
RL formalisms and relations
OpenAI Gym - The anatomy of the agent
Hardware and software requirements
The random CartPole agent
Deep Learning with PyTorch & Tensorflow
Tabular Learning and the Bellman Equation
Deep Q-Networks
Tabular Q-learning
Deep Q-learning
DQN on Pong
Stocks Trading Using RL
Policy Gradients – An Alternative
The Actor-Critic Method
Chatbots Training with RL
Web Navigation
Continuous Action Space
Trust Regions – TRPO, PPO, and ACKTR
Black-Box Optimization in RL
Beyond Model-Free – Imagination
AlphaGo Zero
Unsupervised Deep Learning
Autoencoders
Restricted Boltzman Machine
########################################################################################################
**Tensorflow Guide and one project written in tensorflow
**AWS – SageMaker – Complete deployment of ML and DL models on Cloud
** Spark with python
** How to Participate in Kaggle Competitions
********************************************************************************************************
Major Projects
Chatbot using Deep Learning
Recommender System
Human Body Activity Recognition
Self Driving Car
Deep Learning All in One Course
Ubuntu OS –
• Installations
• Fundamentals of ubuntu
Introduction to Deep Learning
•
• Introduction to Theano
• Introduction to TensorFlow
• Introduction to Keras
• Project: Develop Large Models on GPUs Cheaply In the Cloud
•
Multilayer Perceptrons
• Develop Your First Neural Network With Keras
• Evaluate The Performance of Deep Learning Models
• Use Keras Models With Scikit-Learn For General Machine Learning
• Project: Multiclass Classi_cation Of Flower Species
• Project: Binary Classi_cation Of Sonar Returns
• Project: Regression Of Boston House Prices
• Advanced Multilayer Perceptrons and Keras
• Reduce Overfitting With Dropout Regularization
• Lift Performance With Learning Rate Schedules
•
Convolutional Neural Networks
• Project: Handwritten Digit Recognition
• Improve Model Performance With Image Augmentation
• Project Object Recognition in Photographs
• Project: Predict Sentiment From Movie Reviews
• Image Detection
• Amazon RainForest image Recognition
• Face Recognition
Recurrent Neural Networks
• Long Short-Term Memory Networks
• Time Series Prediction with Multilayer Perceptrons
• Time Series Prediction with LSTM Recurrent Neural Networks
• Project: Sequence Classification of Movie Reviews
• Understanding Stateful LSTM Recurrent Neural Networks
• Project: Text Generation With Alice in Wonderland
**Note:**
The syllabus is subject to adjustments based on the pace of the class and emerging developments in the field of data engineering with PySpark, Scala, and Shell scripting.