Skip to main content

EasyNN is a python package designed to provide an easy-to-use neural network. The package is designed to work right out of the box, while also allowing the user to customize features as they see fit.

Project description

EasyNN - Neural Networks made Easy

EasyNN is a python package designed to provide an easy-to-use Neural Network. The package is designed to work right out of the box with multiple datasets, while also allowing the user to customize features as they see fit.

EasyNN requires Python version 3.9.7 or greater.

See our wiki for more information and Datasets.

Installation:

Run python's pip3 to install:

pip3 install EasyNN

Model:

from EasyNN.examples.mnist.number.trained import model

# Classify an image.
print(model.classify(image))

Dataset Example:

from EasyNN.examples.mnist.number.trained import model
from EasyNN.examples.mnist.number.data import dataset

images, labels = dataset

# Classify what the second image is in the dataset.
print(model.classify(images[1]))

# Show the image.
model.show(images[1])

Dataset example output:

Downloading - number_parameters.npz:
[################################] 1769/1769 - 00:00:00
Downloading - number_structure.pkl:
[################################] 10700/10700 - 00:00:00
Downloading - number_dataset.npz:
[################################] 11221/11221 - 00:00:00
0

Full example:

More info can be found about converting images in the utilities section.

from EasyNN.examples.mnist.number.trained import model
from EasyNN.utilities import Preprocess, download

# Download an example image.
download("three.jpg","https://bit.ly/3dbO1eV")

format_options = dict(
    grayscale=True,
    invert=True,
    process=True,
    contrast=30,
    resize=(28, 28),
    rotate=3,
)

# Converting your image into the correct format for the mnist number dataset.
image = Preprocess("three.jpg").format(**format_options)

# Classify what the image is using the pretrained model.
print(model.classify(image))

# Show the image after it has been processed.
model.show(image)

Output:

Downloading - four.jpg:
[################################] 1371/1371 - 00:00:00
3

Image output:

Trained Models

Use the trained models section to see EasyNN's datasets and pre-trained neural networks ready to run.

MNIST Number Classifier network for images of handwritten single digits between 0 and 9.
MNIST Fashion Classifier network for ten classes of human clothing images of the size 28x28 pixels.
Cifar 10 Classifier network for ten types of images varying from airplane, cat, dog, etc - 32x32 RGB images.

To see more examples with many other datasets. Please visit our wiki.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

EasyNN-0.2.2.tar.gz (31.1 kB view details)

Uploaded Source

Built Distribution

EasyNN-0.2.2-py3-none-any.whl (54.1 kB view details)

Uploaded Python 3

File details

Details for the file EasyNN-0.2.2.tar.gz.

File metadata

  • Download URL: EasyNN-0.2.2.tar.gz
  • Upload date:
  • Size: 31.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.4.2 importlib_metadata/4.8.1 pkginfo/1.7.1 requests/2.26.0 requests-toolbelt/0.9.1 tqdm/4.62.3 CPython/3.9.7

File hashes

Hashes for EasyNN-0.2.2.tar.gz
Algorithm Hash digest
SHA256 cca91b005e30eaf234028364caec98ff2a4dcca823bdef5e5a0dbe213abc6ae0
MD5 8c1d7a6eaf057b7bf063ecb77b615bce
BLAKE2b-256 932bca52f80d2fc2584810e8481ed0652a962e8518e9d997686f5b3c27be3d6e

See more details on using hashes here.

File details

Details for the file EasyNN-0.2.2-py3-none-any.whl.

File metadata

  • Download URL: EasyNN-0.2.2-py3-none-any.whl
  • Upload date:
  • Size: 54.1 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.4.2 importlib_metadata/4.8.1 pkginfo/1.7.1 requests/2.26.0 requests-toolbelt/0.9.1 tqdm/4.62.3 CPython/3.9.7

File hashes

Hashes for EasyNN-0.2.2-py3-none-any.whl
Algorithm Hash digest
SHA256 2553f4d3ab7596fa6720eaad5ffbf34896abc21be3c0f93ede44bb0564a70b4b
MD5 57e58395c1bfa8e9d85076ab30cb0d59
BLAKE2b-256 ba604b530d06d96fb7040c0e5a1d0bb423980511c971d5925fe8b21ed957362b

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page