Deep Learning and Neural Network Regression with Tensorflow

Duration: Hours

Training Mode: Online


Deep Learning with TensorFlow is an end-to-end open-source platform for machine learning. The developers easily build and deploy ML-powered applications.

Objectives :

a). Learn to pass Google’s official TensorFlow Developer Certificate exam (and add it to your resume)
b). Build TensorFlow models using Computer Vision, Convolutional Neural Networks and Natural Language Processing
c). Complete access to ALL interactive notebooks and ALL course slides as downloadable guides
d). Increase your skills in Machine Learning and Deep Learning, to test your abilities with the TensorFlow assessment exam
e). Understand how to integrate Machine Learning into tools and applications
f). Learn to build all types of Machine Learning Models using the latest TensorFlow 2
g). Build image recognition, object detection, text recognition algorithms with deep neural networks and convolutional neural networks.
h). Using real world images to visualize the journey of an image through convolutions to understand how a computer “sees” information, plot loss and accuracy.
i). Applying Deep Learning for Time Series Forecasting
j). Gain the skills you need to become a TensorFlow Certified Developer
k). Be recognized as a top candidate for recruiters seeking TensorFlow developers

a). Introduction

  1. Course outline

b). Deep Learning and Tensorflow Fundamentals

  1. What is deep Learning ?
  2. Why use deep learning ?
  3. What are neural networks ?
  4. What is deep learning already being used for?
  5. What is and why use TensorFlow ?
  6. What is a Tensor ?
  7. Need a Refresher ?
  8. Creating your first tensors with TensorFlow and tf.constant()
  9. Creating tensors with Deep Learning with Deep Learning with TensorFlow and tf.Variable()
  10. Creating random tensors with TensorFlow
  11. Shuffling the order of tensors
  12. Creating tensors from NumPy arrays
  13. Getting information from your tensors (tensor attributes)
  14. Indexing and expanding tensors
  15. Manipulating Tensors with basic operations
  16. Matrix multiplication with tensors
  17. Changing the datatype of tensors
  18. Tensor aggregation (finding the min, max, mean & more)
  19. Tensor troubleshooting example (updating tensor datatypes)
  20. Finding the positional minimum and maximum of a tensor (argmin and argmax)
  21. Squeezing a tensor (removing all 1-dimension axes)
  22. One-hot encoding tensors
  23. Trying out more tensor math operations
  24. Exploring TensorFlow and NumPy’s compatibility
  25. Making sure our tensor operations run really fast on GPUs

c). Neural network regression with Tensorflow

  1. Introduction to Neural Network Regression with TensorFlow
  2. Inputs and outputs of a neural network regression model
  3. Anatomy and architecture of a neural network regression model
  4. Creating sample regression data (so we can model it)
  5. The major steps in modelling with TensorFlow
  6. Steps in improving a model with TensorFlow
  7. Evaluating a Deep Learning with TensorFlow model
  8. Setting up TensorFlow modelling experiments
  9. Comparing and tracking your TensorFlow modelling experiments
  10. How to save a TensorFlow model
  11. How to load and use a saved TensorFlow model
  12. How to save and download files from Google Colab
  13. Preprocessing data with feature scaling

d). Neural network classification with Tensorflow

  1. Introduction to neural network classification in TensorFlow
  2. Example classification problems (and their inputs and outputs)
  3. Input and output tensors of classification problems
  4. Typical architecture of neural network classification models with TensorFlow
  5. Creating and viewing classification data to model
  6. Checking the input and output shapes of our classification data
  7. Building a not very good classification model with TensorFlow
  8. Trying to improve our not very good classification model
  9. Creating a function to view our model’s not so good predictions
  10. Make our poor classification model work for a regression dataset
  11. Non-linearity part 1: Straight lines and non-straight lines
  12. Building our first neural network with non-linearity
  13. Upgrading our non-linear model with more layers
  14. Modelling our non-linear data once and for all
  15. Replicating non-linear activation functions from scratch
  16. Getting great results in less time by tweaking the learning rate
  17. Using the TensorFlow History object to plot a model’s loss curves

e). Computer Vision and Convolution Neural Network in tensorflow

  1. Introduction to Computer Vision with TensorFlow
  2. Introduction to Convolutional Neural Networks (CNNs) with TensorFlow
  3. Downloading an image dataset for our first Food Vision model
  4. Becoming One With Data
  5. Building an end to end CNN Model
  6. Using a GPU to run our CNN model 5x faster
  7. Trying a non-CNN model on our image data
  8. Improving our non-CNN model by adding more layers
  9. Breaking our CNN model down part 1: Becoming one with the data
  10. Preparing to load our data
  11. Loading our data with ImageDataGenerator
  12. Building a baseline CNN model
  13. Looking inside a Conv2D layer
  14. Compiling and fitting our baseline CNN
  15. Evaluating our CNN’s training curves
  16. Reducing overfitting with Max Pooling
  17. Reducing overfitting with data augmentation
  18. Visualizing our augmented data

f). Transfer learning in Tensorflow : Feature extraction

  1. What is and why use transfer learning?
  2. Downloading and preparing data for our first transfer learning model
  3. Introducing Callbacks in Deep Learning with TensorFlow and making a callback to track our models
  4. Exploring the TensorFlow Hub website for pretrained models
  5. Building and compiling a TensorFlow Hub feature extraction model
  6. Blowing our previous models out of the water with transfer learning
  7. Plotting the loss curves of our ResNet feature extraction model

g). Transfer learning in Tensorflow : Fine tuning

  1. Introduction to Transfer Learning : Fine-tuning
  2. Importing a script full of helper functions (and saving lots of space)
  3. Downloading and turning our images into a BatchDataset
  4. Discussing the four (actually five) modelling experiments we’re running
  5. Comparing the  Keras Sequential API versus the Functional API
  6. Creating our first model with the Keras Functional API
  7. Compiling and fitting our first Functional API model
  8. Getting a feature vector from our trained model
  9. Drilling into the concept of a feature vector (a learned representation)
  10. Downloading and preparing the data for Model 1 (1 percent of training data)
  11. Building a data augmentation layer to use inside our model

h). NLP Fundamentals in Tensorflow

  1. Introduction to Natural Language Processing (NLP) and Sequence Problems
  2. Example NLP inputs and outputs
  3. The typical architecture of a Recurrent Neural Network (RNN)
  4. Preparing a notebook for our first NLP with project
  5. Becoming one with the data and visualising a text dataset
  6. Splitting data into training and validation sets
  7. Converting text data to numbers using tokenisation and embeddings (overview)
  8. Setting up a TextVectorization layer to convert text to numbers
  9. Mapping the TextVectorization layer to text data and turning it into numbers
  10. Creating an Embedding layer to turn tokenised text into embedding vectors
  11. Discussing the various modelling experiments we’re going to run


Locus IT has decade long industry experience in “Deep Learning with Tensorflow Developer” consulting, staffing & training services. For more inputs on Tensorflow Developer Training/staffing  you can connect here.
Contact the L&D Specialist at Locus IT
Locus Academy has more than a decade experience in delivering the training/staffing on Tensorflow Development for corporates across the globe. The participants for the training/staffing on Tensorflow Development are extremely satisfied and are able to implement the learnings in their on going projects.

Other useful references



TensorFlow is an end-to-end open source platform for machine learning. It has a comprehensive and flexible ecosystem of tools, libraries and also community resources that lets researchers push the state-of-the-art in ML. The developers easily build and deploy ML powered applications.