Tensorflow on Google’s Cloud Platform for Data Engineers
Welcome to Tensorflow on the Google Cloud Platform for Data Engineers This is the fourth course in a series of courses designed to help you attain the coveted Google Certified Data Engineer.
Additionally, the series of courses is going to show you the role of the data engineer on the Google Cloud Platform.
NOTE: This is not a course on how to develop machine learning models with TensorFlow. This is a very targeted course on TensorFlow for data engineers. My goal is to give data engineers what they need to know for the exam and provide learners with the foundations of TensorFlow on Google’s Cloud Platform.
At this juncture the Google Certified Data Engineer is the only real world certification for data and machine learning engineers.
TensorFlow is an open source software library created by Goggle for doing graph-based computations quickly. It does this by utilizing the GPU(Graphics Processing Unit) and also making it easy to distribute the work across multiple GPUs and computers.
Tensors, in general, are simply arrays of numbers, or functions, that transform according to certain rules under a change of conditions. Nodes in the graphs represent mathematical operations, while the graph edges represent the multidimensional data arrays (tensors) communicated between them.
In the course you’ll discover how to apply TensorFlow to machine learning, the concept of a Tensor, the anatomy of a simple program, basic constructs such as constants, variables, placeholders, sessions and the computation graph.
You’ll work with basic math operations and image transformations to see how common computations are performed.
You’ll learn TensorFlow within the context of the Google Cloud Platform.
*Five Reasons to take this Course.*
1) You Want to be a Data Engineer
the number one job in the world. (not just within the computer space) The
growth potential career wise is second to none. You want the freedom to move
anywhere you’d like. You want to be compensated for your efforts. You want to
be able to work remotely. The list of benefits goes on.
2) The Google Certified Data
is always ahead of the game. If you were to look back at a timeline of their
accomplishments in the data space you might believe they have a crystal ball.
They’ve been a decade ahead of everyone. Now, they are the first and the
only cloud vendor to have a data engineering certification. With their track
record I’ll go with Google.
3) The Growth of Data is
percent of all the world’s data has been created in the last two years.
Business around the world generate approximately 450 billions transactions a
day. The amount of data collected by all organizations is approximately 2.5
Exabytes a day. That number doubles every month.
4) TensorFlow in Plain English
TensorFlow is a low level language. The basic concept of a tenor is hard to grasp if you aren’t familiar with neural networks. In the course we will break down TensorFlow in to bite sized pieces ensuring you learn the fundamentals first. After we’ve built a base understanding of tensors and how they flow we will move on to more complicated examples.
5) You want to be ahead of the Curve
The data engineer role is fairly new. While your learning, building your skills and becoming certified you are also the first to be part of this burgeoning field. You know that the first to be certified means the first to be hired and first to receive the top compensation package.
Thank you for your interest in Tensorflow on the Google Cloud Platform for Data Engineers and we will see you in the course!!
Welcome to TensorFlow
This course is about TensorFlow.
This is Google's computational engine for machine learning.
In this lecture I discuss created a machine learning course for data engineers.
This course will be very targeted and specific to questions on the exam.
Is this course right for you?
Let's find out in this lesson.
In this lesson I answer a few questions about the course and the exam.
Let's define what an array is.
The very basics.
Tensors are multi-dimensional arrays.
Let's find out what that is in this lesson.
Tensors and nodes together create a TensorFlow network.
These networks are represented as graphs in TensorFlow.
In this lesson let's add some real numbers to our graph.
This will help show how tensor move through the graph with data in them.
Every programming language has hello world and TensorFlow is no different.
Let's walk through the most basic TensorFlow example we can.
Up and Running in Cloud Datalab
Let's learn how to create a datalab vm for the course.
Once we close our datalab VM we will need to be able to log back into it in order to build our models.
Let's see how simple it is to call our existing datalab notebook from cloud shell.
In this lecture let's learn how to upload and download our cloud datalab notebooks.
In this lesson let's set up a virtual machine that we will use throughout the course to host our datalab notebooks.
Let's learn about the TensorFlow toolkit in this lesson.
All TensorFlow graphs feed in one direction and that is forward through the graph.
Let's learn how this operates in this lesson.
Machine learning involves lots of iteration but TensorFlow doesn't.
Let's learn how Tensorflow handles that in this lesson.
Every TensorFlow program has two steps.
In this lesson let's learn what they are.
Real world machine learning is machine learning at scale.
Let's learn how TensorFlow handles that in this lesson.
In this lesson let's add more resources to our datalab vm.
In this example let's walk through a very simple end to end model in TensorFlow.
Let's learn about the dimension of Tensors in this lesson.
In this lesson let's learn what placeholders are and how to use them.
In this lesson let's learn how to pass two core parameters into our session.
Let's learn what a node's life cycle in this brief lesson.
A tensor have three properties.
Let's learn what they are in this lesson.
Let's convert our non-tensored arrays to tensors.
In this lesson let's learn how to enable logging on our TensorFlow models.
In this lesson let's work through the most simple TensorFlow example we can.
In this lesson let's view the difference between executing a numpy array in Python and a tensor in TensorFlow.
We will do all of this in our datalab notebooks.
We are going to do this in our datalab notebooks but our data will be in BigQuery.
Let's massage our data and create clean datasets for our machine learning models.
In this lesson let'walk through a linear regression model in TensorFlow.
A strange example from Google but it's been on a few exams so let's cover it here.
In this lesson let's learn about overfitting and how to avoid it.
Real world machine learning is all about scale.
In this lesson let's learn about scale on GCP.
Once we have our model it's time to pack it up.
Let's learn how to package a completed model in this lesson.
In this lesson let's learn the basics of creating a prediction service.
In this lesson let's complete a lab specific to linear regression in TensorFlow.
Let's walk through the lab step by step.