Saving and loading arrays

Introduction to NumPy

Izzy Weber

Curriculum Manager, DataCamp

RGB arrays

rgb = np.array([[[255, 0, 0], [255, 0, 0], [255, 0, 0]],
                [[0, 255, 0], [0, 255, 0], [0, 255, 0]],
                [[0, 0, 255], [0, 0, 255], [0, 0, 255]]])
plt.imshow(rgb)
plt.show()

Plotted RGB data with red on the top row, green in the middle, and blue at the bottom

RGB arrays

A color-coded code snippet showing how mixed colors such as pink and yellow are represented in RGB data Multicolored three by three grid of colors created by code in the previous image

Loading .npy files

Save arrays in many formats:

.csv
.txt
.pkl
.npy

with open("logo.npy", "rb") as f:
    logo_rgb_array = np.load(f)
plt.imshow(logo_rgb_array)
plt.show()

Image of blue NumPy logo on white background

Examining RGB data

red_array = logo_rgb_array[:, :, 0]
blue_array = logo_rgb_array[:, :, 1]
green_array = logo_rgb_array[:, :, 2]

A 3D RGB array sliced into three 2D arrays: one red, one green, and one blue

Examining RGB data

red_array[1], green_array[1], blue_array[1]

(array([255, 255, 255, ..., 255, 255, 255]),
 array([255, 255, 255, ..., 255, 255, 255]),
 array([255, 255, 255, ..., 255, 255, 255]))

Updating RGB data

dark_logo_array = np.where(logo_rgb_array == 255, 50, logo_rgb_array)
plt.imshow(dark_logo_array)
plt.show()

The same NumPy logo we saw earlier but with a dark gray background instead of a white one

Saving arrays as .npy files

with open("dark_logo.npy", "wb") as f:
    np.save(f, dark_logo_array)

If we need help()...

help(np.unique)

Help on function unique in module numpy:
unique(ar, return_index=False, return_inverse=False, return_counts=False,
        axis=None)

Find the unique elements of an array.

Returns the sorted unique elements of an array. There are three optional
outputs in addition to the unique elements:

* the indices of the input array that give the unique values...

A screen shot of the numpy.org documentation for np.unique()

help() with methods

help(np.ndarray.flatten)

Help on method_descriptor: flatten(...)
a.flatten(order='C')

Return a copy of the array collapsed into one dimension.

Parameters
<hr />-------
order : {'C', 'F', 'A', 'K'}, optional
    'C' means to flatten in row-major (C-style) order.
    'F' means to flatten in column-major (Fortran- ...

Let's practice!

Introduction to NumPy