Welcome back. In this course, you'll learn how to use
TensorFlow to process texts for natural
language processing. Unlike images, which come in these regular shaped tensors
of pixel intensity values, Text is messier, there
are long sentences, there are short sentences. Do you want to process
individual characters alphabets or do you
wanna process words? So in this course, you learn how to
deal with all that. Yes. So like in our first week as well we're going to take
a look at what it takes for you to process text because neural networks
generally deal in numbers. The zero functions, calculating weights and
biases is all numbers. So how are we going to convert our texts into numbers
in a sensible way? Given a word like cat, how do you turn
that into a set of numbers that you can feed
into neural network? Exactly. Then if I have cat, then what's dog
going to look like? All of those kind of things. Then multiple length
sentences as well. How do we deal with padding them? Or if you've got like a body of words that
you use for training, and then you have another body of words that you want to
actually predict on, but you're going to
have some words in this body that
aren't in that one. How do you deal with out
of vocabulary tokens and that kind of things? There's going to be a lot of fun. Yeah. So in this first week, you learn how to
load in the texts, pre-process it and
set up your data so it can be fed to
a neural network. I'm really excited.
In this course, you'll learn how to deal
with texts using TensorFlow. So let's go on to
the next video to get started.