-
Notifications
You must be signed in to change notification settings - Fork 2
Image Utilitie Show
Daniel Wilczak edited this page Dec 3, 2021
·
14 revisions
Every trained model allows for a model.show()
function that will output the data in a format that follows the dataset's form. A few examples can be seen below.
To learn more about the MNIST handwritten dataset.
from EasyNN.examples.mnist.number.trained import model
from EasyNN.examples.mnist.number.data import dataset
images, labels = dataset
model.show(images[0])
This functionality can also be paired up with downloaded images that are properly sized.
from EasyNN.examples.mnist.fashion.trained import model
from EasyNN.utilities import Preprocess, download
download("dress.jpg","https://bit.ly/3b7rsXF")
format_options = dict(
grayscale=True,
invert=True,
process=True,
contrast=5,
resize=(28, 28),
rotate=0,
)
# Converting your image into the correct format for the mnist fashion dataset.
image = Preprocess("dress.jpg").format(**format_options)
# Show the image after it has been processed.
model.show(image)
# Classify what the image is using the pretrained model.
print(model.classify(image))