# Numpy All, Explained

This article is originally published at https://www.sharpsightlabs.com

In this tutorial, I’ll explain how to use the Numpy all function (AKA, np.all).

I’ll explain the what this function does, how the syntax works, and I’ll show you clear, step-by-step examples of how to use it.

If you need something specific, just click on any of the following links. The link will take you to that specific part of the tutorial.

**Table of Contents:**

Having said that, if you’re somewhat new to Numpy or want all of the details, you should probably read the tutorial start to finish.

## A Quick Introduction to Numpy All

Ok. Let’s start off with a quick introduction to the Numpy all function and how it works.

At a high level, **the Numpy all function tests a Numpy array and evaluates if all of the elements evaluate as True**.

This explanation might be a little confusing, so let’s quickly review a few things about Numpy to help you understand.

#### A Quick Review of Numpy Arrays

First, since `np.all()`

works on Numpy arrays, let’s quickly review what Numpy arrays are.

You probably know that a Numpy array is a data structure that we use in Python.

More specifically, a Numpy array is a row-and-column data structure that we use to store numeric data.

Typically Numpy arrays contain numeric data, like integers and floating point numbers. But technically, it’s possible for a Numpy array to hold other data types like boolean data.

This will be relevant later as we discuss what `numpy.all`

does.

##### Numpy Arrays have Axes

Another important feature of Numpy arrays is that Numpy arrays have *axes*.

Axes often confuse beginners, but there’s sort of a simple way to think about them.

Numpy axes are like *directions* along edges of a Numpy array.

For 1-dimensional arrays, there is only one axis, axis-0. And for 1D arrays, axis-0 points horizontally.

Things are a little different for 2-dimensional arrays. For 2D arrays, axis-0 points downward and axis-1 points horizontally.

This is relevant, because when we use `np.all()`

, we can use the `axis`

parameter, which will cause the function to operate in one direction or another. I’ll show you how this works in the syntax section and show you examples in the examples section.

### The Numpy All Function Tests if All Elements Evaluate as True

Now, let’s return to the Numpy all function.

The `np.all()`

function tests if all elements in an array are `True`

. There are some more complicated applications of this, but the simplest way to see it is with a small Numpy array that contains boolean data. (Remember that I mentioned earlier that a Numpy array can contain boolean data.)

If we have a Numpy array with boolean, True or False data, we can use np.all() to check if all of the elements are `True`

.

Or if we have a Numpy array with numeric data, we can test if all of the numbers meet certain conditions.

At a high level, this is what `np.all`

does: it tests Numpy arrays to check if *all elements* meet some condition or evaluate as `True`

.

Having said that, there are a few details about how it operates that can change depending on the syntax.

That being the case, lets look at the syntax.

## The Syntax of Numpy All

Here, we’ll look at the syntax of the `all()`

function, including some important parameters.

#### A quick note

One quick note before we dive in.

The exact form of the syntax will depend on how we import Numpy.

The common convention among Python programmers and data scientists is to import Numpy with the alias `np`

. You can do that with the following code:

import numpy as np

After you import Numpy this way, you can call Numpy functions starting with the prefix `np`

.

### np.all syntax

Let’s start with a look at the basic syntax.

To call the function, you simply type the syntax `np.all()`

. Again, this assumes that you’ve imported Numpy with the alias `np`

.

Then, inside the parenthesis, the first argument is the name of the Numpy array that you want to operate on. In the above picture, I’ve called that array `input_array`

.

After the name of the input array, there are several parameters that enable you to slightly change how the function behaves.

### The Parameters of np.all

The np.all function has a few parameters and arguments that control its behavior:

`input_array`

`axis`

`out`

`keepdims`

Let’s take a look at each of them individually.

`input_array`

(required)

The first argument to the function is `input_array`

. (Note that in the official documentation, this is called `a`

.)

This parameter enables you to specify the input to the function.

Importantly, do don’t actually need to type this parameter name. Rather, you simply type the name of your array as the first argument to the function. I’ll show you an example of this in the examples section.

Additionally, this parameter will accept several types of inputs as valid inputs. Numpy arrays are obviously valid inputs, but this function will also accept array-like objects such as Python lists and tuples.

Also note that this is required. You must provide some kind of array-like input for this parameter.

`axis`

(optional)

The `axis`

parameter enables you to specify an axis along which to use `np.all`

.

By default, this parameter is set to `axis = None`

. In this case, the `all()`

function will operate on all of the array elements collectively, without regard to any particular axis.

But you can also use this parameter such that `np.all`

will operate along a particular axis.

If you set `axis = 0`

, the function will operate in the axis-0 direction. This will not have an effect on a 1-dimensional array. But for a 2-dimensional array, this will cause `np.all`

to operate *downward*. So effectively, setting `axis = 0`

causes the function to operate on the *columns* of a 2D array.

Similarly, if you set `axis = 1`

, the function will operate in the axis-1 direction. For a 2D array or n-dimensional array, axis-1 points horizontally. So effectively, if you set `axis = 1`

, `np.all`

will test the rows for a particular condition.

I realize that this is confusing, so I’ll show you examples of this in the examples section.

If you’re confused about Numpy axes, you should read our tutorial that explains how axes work.

`out`

(optional)

The `out`

parameter enable you to specify a particular array in which to put the output.

`keepdims`

(optional)

You can use the `keepdims`

parameter to “keep” the original number of dimensions when you apply `np.all`

.

Essentially, when you set `keepdims = True`

, the output will have the same number of dimensions as the input.

### The output of np.all

By default, the `np.all`

returns a boolean value (i.e., `True`

or `False`

) that indicates whether *all* of the elements of the input are true, or meet a particular condition.

However, if you use `np.all`

with the `axis`

parameter on a multi-dimensional array, there are many instances where the function will output an *array* of boolean values instead. (I’ll show you an example of this in the examples section.)

Ok. Now that we’ve looked at the syntax of the function, let’s look at some examples.

## Examples of How to Use Numpy All

Here, I’ll show you a few step-by-step examples of how to use Numpy all.

You can click on any of these links, and it will take you directly to the appropriate example.

**Examples:**

- Use np.all on a 1-dimensional array
- Test an array for a specific condition
- Use np.all on a 2-dimensional array
- Apply np.all along axis-0
- Apply np.all along axis-1

#### Run this code first

One quick thing before we get started.

As I mentioned in the syntax section, in order to use this function, you first need to import Numpy.

You can do that with this code:

import numpy as np

Once you import Numpy, you’ll be ready to run the examples.

### EXAMPLE 1: Use np.all on a 1-dimensional array

First, let’s start with a simple example.

Here, we’ll apply the `all()`

function to a 1-dimensional list of boolean values.

Note that we could use a proper Numpy array instead, but a Python list makes it easier to directly see what’s going on in the syntax.

Specifically, we’re going to apply `np.all`

to the list `[True, True, True]`

.

Let’s take a look.

np.all([True, True, True])

OUT:

`True`

###### Explanation

This is really simple.

The Numpy `all()`

function evaluates if all of the input elements are `True`

.

In this case, we passed in the list `[True, True, True]`

.

Clearly, all of the values of the input *are* true.

As a result, `np.all`

produced the output value `True`

, indicating that all of the values in the input are true.

Note: if you want to try this with a proper Numpy array, you can run the following code:

my_true_array = np.array([True, True, True]) np.all(my_true_array)

### EXAMPLE 2: Test an array for a specific condition

Next, let’s test an array for a specific condition.

To do this, we can actually input a *boolean expression* into `np.all`

, and it will evaluate if that condition is value for all values of the Numpy array.

Let me show you what I mean.

###### Create array

First, let’s just create a Numpy array.

Here, instead of creating an array with boolean values, we’ll create a traditional array with numeric data.

my_1d_array = np.array([1,2,3])

And we can print it out to see the contents:

print(my_1d_array)

OUT:

[1 2 3]

As you can see, `my_1d_array`

is a simple 1D array with three numeric values.

###### Run function

Next, we’re going to use Numpy’s `all()`

function to check if a logical expression is valid for every element of the array.

np.all(my_1d_array > 2)

OUT:

False

###### Explanation

When we use the `all()`

function like this, and input a *boolean expression* that’s operating on a Numpy array, the `all()`

function will test if that condition is valid for *every element* of the array.

Here, the conditional expression `my_1d_array > 2`

is testing if each particular value is greater than 2.

In fact, the conditional expression inside the function actually produces a Numpy array with True/False values. I recommend that you run the following code, so you can see what’s going on:

my_1d_array > 2

OUT:

array([False, False, True])

When you run that conditional logic by itself, you can see that it outputs a Numpy array with True/False values. These True/False values show if the condition is valid for *particular elements* of `my_1d_array`

.

Then `np.all`

operates on this intermediate array, and checks if the logic is true for *all* of the values. In this case the values of `my_1d_array`

are *not* all greater than 2, so the code `np.all(my_1d_array > 2)`

returns `False`

.

### EXAMPLE 3: Use np.all on a 2-dimensional array

Now, we’ll use np.all on a 2-dimensional array.

This is really very similar to example 2, but here, we’ll use a 2D input instead of a 1D input.

###### Create array

First, let’s just create a 2D Numpy array. To do this, we’ll use Numpy arrange to generate a range of numbers, and then we’ll use the Numpy reshape method to reshape that array into a 2-dimensional shape.

my_2d_array = np.arange(start = 1, stop = 7).reshape((2,3))

And let’s print it out so you can see the contents.

print(my_2d_array)

OUT:

[[1 2 3] [4 5 6]]

As you can see, `my_2d_array`

contains the values from 1 to 6, arranged into a 2×3 shape.

###### Use Function

Now, let’s use the `all()`

function to test a condition for the whole 2D array.

np.all(my_2d_array > 2)

OUT:

False

###### Explanation

Here, the output is False, indicating that all of the elements are *not* greater than 2.

When we use Numpy all on a 2D array like this, it simply identifies if the condition applies to *all* of the elements.

Also, one other quick note. Behind the scenes, what’s actually going on is that the conditional logic itself, `my_2d_array > 2`

, is creating a Numpy array filled with boolean values.

To see this, just type that code into Python by itself:

my_2d_array > 2

OUT:

array([[False, False, True], [ True, True, True]])

So what’s happening, is that the conditional logic (i.e., the greater than operation) creates a Numpy array filled with boolean values. Then `np.all`

operates on that to make a final `True`

/`False`

determination.

### EXAMPLE 4: Apply np.all along axis-0

In this example, we’ll make things a little more complicated. Here, we’re going to use the `axis`

parameter to apply the function along the axis-0 direction.

###### Create array

First, let’s just create a 2D Numpy array.

Here, we’ll create a 2-dimensional Numpy array that contains the values from 1 to 6, arranged into a 2×3 shape.

(This is the same array that we created in example 3, so if you created it there, you don’t need to create it again.)

my_2d_array = np.arange(start = 1, stop = 7).reshape((2,3))

And let’s print it out so you can see the contents.

print(my_2d_array)

OUT:

[[1 2 3] [4 5 6]]

So `my_2d_array`

just has the values from 1 to 6 arranged into a 2D shape.

###### Use function

Now, we’re going to test some conditional logic in the axis-0 direction.

Here, we’re going to test if all of the values are greater than 2, *in the axis-0 direction*.

np.all(my_2d_array > 2, axis = 0)

OUT:

array([False, False, True])

###### Explanation

What’s going on here?

In this example, the `np.all`

function tested the conditional logic in the axis-0 direction.

Remember that for a 2D array, axis-0 points downward.

When we use the function with the `axis`

parameter set to `axis = 0`

, the function will test if that condition is true for all of the elements, along that direction.

Effectively, when we set `axis = 0`

, `np.all`

tests if the logic is true for each column.

### EXAMPLE 5: Apply np.all along axis-1

Finally, let’s apply the function along the axis-1 direction.

This will effectively apply the function along the rows.

###### Create array

First, let’s just create our 2D Numpy array.

(This is the same array that we created in example 3 and example 4, so if you created it already, you don’t need to create it again.)

my_2d_array = np.arange(start = 1, stop = 7).reshape((2,3))

And let’s print it out so you can see the contents.

print(my_2d_array)

OUT:

[[1 2 3] [4 5 6]]

`my_2d_array`

has the values from 1 to 6 arranged into a 2D shape.

###### Use function

Next, we’ll apply some conditional logic in the axis-1 direction.

Here again, we’re going to test if the values are greater than 2, but now we’ll do it in the *in the axis-1 direction*.

np.all(my_2d_array > 2, axis = 1)

OUT:

array([False, True])

###### Explanation

What happened here?

In this example, the function tested the conditional logic in the axis-1 direction.

Remember that for a 2D array, axis-1 points horizontally.

When we use this function with the `axis`

parameter set to `axis = 1`

, the function tests if that condition is true for all of the elements *along the axis-1 direction*.

Effectively, when we set `axis = 1`

, `np.all`

tests if the logic is true for each *row*.

##### Leave your other questions in the comments below

Do you have other questions about Numpy all?

If so, leave your questions in the comments section near the bottom of the page.

## Join our course to learn more about Numpy

Are you interested in learning more about Numpy?

This tutorial should have shown you how to use the Numpy `all()`

function, but if you want to master Numpy, there’s a lot more to learn.

If you’re ready to master Numpy you should join our premium course, Numpy Mastery.

In this course, you’ll learn everything you need to know about Numpy.

- How to create Numpy arrays
- How to use the Numpy random functions
- What Numpy axes are, and how to use them
- What the “Numpy random seed” function does
- How to reshape, split, and combine your Numpy arrays
- and much more …

Additionally, you’ll discover our unique practice system that will enable you to memorize all of the syntax that you learn. Once you take this course and practice like we show you, you’ll be able to write Numpy code fluently, accurately, and 100% from memory.

Find out more here:

Learn More About Numpy Mastery

The post Numpy All, Explained appeared first on Sharp Sight.

Thanks for visiting r-craft.org

This article is originally published at https://www.sharpsightlabs.com

Please visit source website for post related comments.