SqueezeNet implementation with Keras Framework
Switch branches/tags
Nothing to show
Clone or download
Latest commit 2b52ab1 Nov 28, 2017
Failed to load latest commit information.
.github update template Nov 28, 2017
examples new version prepare Nov 28, 2017
images new version prepare Nov 28, 2017
keras_squeezenet new version prepare Nov 28, 2017
.gitignore gitignore update Nov 28, 2017
.travis.yml new version prepare Nov 28, 2017
LICENSE init Oct 8, 2016
README.md update readme Nov 28, 2017
setup.py new version prepare Nov 28, 2017
test.py update test Nov 28, 2017


keras-squeezenet Build Status

SqueezeNet v1.1 Implementation using Keras Functional Framework 2.0

This network model has AlexNet accuracy with small footprint (5.1 MB) Pretrained models are converted from original Caffe network.

pip install keras_squeezenet


  • Project is now up-to-date with the new Keras version (2.0).

  • Old Implementation is still available at 'keras1' branch but not updated.

Library Versions

  • Keras v2.1.1
  • Tensorflow v1.4

Example Usage

import numpy as np
from keras_squeezenet import SqueezeNet
from keras.applications.imagenet_utils import preprocess_input, decode_predictions
from keras.preprocessing import image

model = SqueezeNet()

img = image.load_img('../images/cat.jpeg', target_size=(227, 227))
x = image.img_to_array(img)
x = np.expand_dims(x, axis=0)
x = preprocess_input(x)

preds = model.predict(x)
print('Predicted:', decode_predictions(preds))


  1. Keras Framework

  2. SqueezeNet Official Github Repo

  3. SqueezeNet Paper


MIT License

Note: If you find this project useful, please include reference link in your work.