Deep Learning with Tensorflow

In this section you will find all the Deep Learning Projects using Tensorflow 2.0. Projects such as Movie Poster Classification, Bank Customer Satisfaction, Human Activity Recognition,Customer Churn, among others.

¿What is Deep Learning?

The field of artificial intelligence is essentially when machines can do tasks that typically require human intelligence. It encompasses machine learning, where machines can learn by experience and acquire skills without human involvement. Deep learning is a subset of machine learning where artificial neural networks, algorithms inspired by the human brain, learn from large amounts of data. Similarly to how we learn from experience, the deep learning algorithm would perform a task repeatedly, each time tweaking it a little to improve the outcome. We refer to ‘deep learning’ because the neural networks have various (deep) layers that enable learning. Just about any problem that requires “thought” to figure out is a problem deep learning can learn to solve.

Deep learning allows machines to solve complex problems even when using a data set that is very diverse, unstructured and inter-connected. The more deep learning algorithms learn, the better they perform.

Projects

1. Human Activity Recognition

Human Activity Recognition (HAR) has been a challenging problem yet it needs to be solved. It will mainly be used for eldercare and healthcare as an assistive technology when ensemble with other technologies like Internet of Things(IoT). HAR can be done with the help of sensors, smartphones or images.

Dataset

The dataset contains data collected through controlled, laboratory conditions.

Raw Time Series Data

Number of examples: 1,098,207
Number of attributes: 6
Missing attribute values: None
Class Distribution:

Walking: 424,400 (38.6%)
Jogging: 342,177 (31.2%)
Upstairs: 122,869 (11.2%)
Downstairs: 100,427 (9.1%)
Sitting: 59,939 (5.5%)
Standing: 48,395 (4.4%)

In the following image you'll see the signals for walking

For this case, I built a Convolutional Neural Network to classify each class, and I got accuracy values for the training and the validation samples closer to 90 %.

And for the testing samples I got the following confusion matrix with 88 % of accuracy. Which means that the model classify 95 instances correctly and 12 instances incorrectly.

Link to Github Repository

2. Customer Churn Prediction

There’s no more vital metric for a SaaS company to keep track of than churn: the rate at which customers are leaving your business and taking their subscription dollars elsewhere. Churn can be powered by a number of factors, and even small month-on-month increases in churn percentage can be ruinous to planning, so understanding what churn is and how to analyze it is paramount.

¿What is Churn Analysis?

Churn analysis is the evaluation of a company’s customer loss rate in order to reduce it. Also referred to as customer attrition rate, churn can be minimized by assessing your product and how people use it.

Project Overview

This project uses an Artificial Neural Network to predict whether a customer will leave or not the company by analyzing features such as Tenure, Age, Balance, Number of Products, Estimated Salary, whether or not is an active member, among others. The created model predicts with an 86.7 % accuracy the customer churn.

Link to Github Repository

3. Google Stock Price Prediction

The stock's price only tells you a company's current value or its market value. So, the price represents how much the stock trades at—or the price agreed upon by a buyer and a seller. If there are more buyers than sellers, the stock's price will climb. If there are more sellers than buyers, the price will drop.

In this project I created a Recurrent Neural Network (RNN) with Long Short Term Memory (LSTM) to predict the Google stock price since 2004 to 2020.

The dataset was downloeaded from Yahoo Finances. Following the link.

Dataset

In the following image you'll see the performance of the Neural Network, in this case, the tomato color line correspond to the real stock price and the blue line corresponds to the predicted stock price by the model, as you can see, the prediction is quite good, in order to improve the performance of the model you can use a deeper neural network or change the hyperparameters of the model.

Link to Github Repository

More Projects

Link to Complete Github Repository