Choose your language:

France
Germany
Hong Kong
India
Ireland
Japan
Malaysia
Netherlands
New Zealand
Singapore
Sweden
United Kingdom
United States

TensorFlow Deep Learning

Course Code

BD73

Duration

3 Days

General knowledge of data stores, advanced math, analytics and a working knowledge of the Python language.
Deep learning is the intersection of statistics, artificial intelligence, and data to build accurate models and TensorFlow is one of the newest and most comprehensive libraries for implementing deep learning. Deep-learning networks are distinguished from ordinary neural networks having more hidden layers. These kinds of nets are capable of discovering hidden structures within unlabeled and unstructured data. The course will show you how to train and optimize basic neural networks, convolutional neural networks, and long short term memory networks. Complete learning systems in TensorFlow will be introduced via lectures and labs. You will learn to solve new classes of problems that were once thought prohibitively challenging, and come to better appreciate the complex nature of human intelligence as you solve these same problems effortlessly using the deep learning methods of TensorFlow.
This course is designed for Data Analysts, Data Scientists, Machine Learning Engineers, and anyone interested in Machine Learning/Deep Learning with TensorFlow.

In this course, participants will:

  • Understand Neural Networks and Deep Learning
  • Install TensorFlow
  • Build TensorFlow graphs for everyday computations
  • Apply logistic regression for classification with TensorFlow
  • Design and train a multilayer neural network with TensorFlow
  • Understand Convolutional Neural Networks
  • Program a Recurrent Neural Network
  • Use TensorFlow Linear Models
  • Learn to Optimize Input Pipelines
  • Apply Performance Best Practices
1. Overview
History
Understanding
Machine Learning
Neural Networks
The Neuron
Backpropagation
Long Short-Term Memory
Dataflow Graph
TensorFlow
CPU, GPU, TPU
Deep Learning Frameworks
TensorFlow Serving
Applications
Lab

2. Installing TensorFlow
Platforms
CPU Support Only
GPU Support
Python Versions
Installing TensorFlow on Ubuntu
Virtualenv
Pip
Docker
Anaconda
Installing TensorFlow on macOS
Virtualenv
Pip
Docker
Installing TensorFlow on Windows
Python 3.5.x and 3.6.x
Pip
Anaconda
Validate Installation
Hello TensorFlow
Lab

3. Fundamental Math
Scalars
Vectors
Matrices
Matrix Mathematics
Tensors
Tensorflow Computations
Imperative Coding
Declarative Coding
Tensor Arithmetic
Tensor Types
Tensorflow Sessions
Tensorflow Variables
Linear Regression
Lab

4. Deep Learning
Artificial Neural Networks
Deep Neural Networks
Neural Network Layers
Fully Connected Layer
Convolutional Layer
Recurrent Neural Network (RNN) Layers
Long Short-Term Memory
TensorFlow Architecture
Hidden Markov Models
Autoencoder
Gradient Descent
Lab

5. Estimators
Advantages of Estimators
Pre-made Estimators
tf.layers
Structure of a Pre-made Estimators
Benefits of Pre-made Estimators
Custom Estimators
Recommended Workflow
tf.estimator
Iris Dataset
DNNClassifier
Model Accuracy
Lab

6. Tensors & Variables
Properties
Rank
Shape
Changing the Shape
Data Types
Evaluating Tensors
Tensor.eval
Printing Tensors
tf.SparseTensor
tf.Variable
Creating a Variable
Variable Collections
Initializing Variables
Using Variables
Lab

7. TensorBoard
Visualizing Learning
Summary Data
Launching TensorBoard
Graph Visualization
Name Scoping
Nodes
Special Icons
Interaction
Tensor Shape
Runtime Statistics
Histogram Dashboard
tf.summary.histogram
Overlay Mode
Multimodal Distributions
Lab

8. Sessions & Graphs
Dataflow Graphs
tf.Graph
Building a Graph
Naming Operations
Tensor-like Objects
tf.Session
Executing
Creating a Session
tf.Session.run
Visualizing your Graph
Multiple Graphs
Eager Execution
Lab

9. Convolutional Neural Networks
Convolutional Neural Networks
Building a CNN MNIST Classifier
Input Layer
Convolutional Layer #1
Pooling Layer #1
Convolutional Layer #2
Pooling Layer #2
Dense Layer
Logits Layer
Generate Predictions
Calculate Loss
Configure the Training Op
Evaluation Metrics
Train the Model
Load Test Data
Create the Estimator
Evaluate the Model
Run the Model
Lab

10. Vector Representations of Words
word2vec Model
Word Embeddings
Scaling
Noise-Contrastive Training
The Skip-gram Model
Building the Graph
Training the Model
Visualizing the Learned Embeddings
Evaluating Embeddings
Analogical Reasoning
Optimizing
Lab

11. Recurrent Neural Networks
Identical Feedforward Neural Networks
RNN cell
Backpropagation
Long Short-Term Memory Networks
Character-Level Language Models
Data Files
Prepare the Data
The Model
LSTM
Truncated Backpropagation
Inputs
Loss Function
Stacking multiple LSTMs
Run the Code
Lab

12. TensorFlow Linear Model
Linear Model
tf.estimator
Feature Columns
Transformations
Sparse Columns
Feature Crosses
Bucketization
Linear Estimators
Logistic Regression
Census Data
Converting Data into Tensors
Selecting Features
Engineering Features
Logistic Regression Model
Regularization
Training The Model
Evaluating The Model
Lab

13.TensorFlow Wide & Deep Learning
Define Base Feature Columns
The Wide Model
Linear Model with Crossed Feature Columns
The Deep Model
Neural Network with Embeddings
Combining Wide and Deep Models
DNNLinearCombinedClassifier
Training The Model
Evaluating The Model
Detail Code
Lab

14. Input Pipelines
Importing Data
Dataset Objects
Dataset Structure
Iterator Objects
One-shot,
Initializable
Reinitializable
Feedable
Consuming Values
Reading Input Data
Consuming NumPy Arrays
Consuming TFRecord Data
Consuming Text Data
Preprocessing Data
Dataset.map()
Lab

15. Performance Best Practices
General Best Practices
Input Pipeline Optimization
Preprocessing on the CPU
Using the Dataset API
Use Large Files
Data Formats
Common Fused Ops
Optimizing for GPU
Optimizing for CPU
Parallelize I/O Reads
Parallelize CPU-to-GPU Data Transfer
Software Pipelining
Variable Distribution
Gradient Aggregation
Lab
Send Us a Message
Choose one