View Code? Open in Web Editor
NEW
Application of Convolutional Neural Networks (ConvNet) for image classification.
License: GNU General Public License v3.0
Jupyter Notebook 99.99%
Shell 0.01%
cnn-playground's Introduction
Using Convolutional Neural Networks (CNN, ConvNet) for Image Classification.
ImageNet/LSVRC 2012 : 1.2 million images in 1,000 classes. Additional 150,000 validation and test images. Dataset @Kaggle
COCO - Common Objects in Context
CIFAR
CIFAR-10 : 60,000 32x32 colour images in 10 classes . 6,000 images per class, 5,0000 training, 10,000 test images.
CIFAR-10@Kaggle
CIFAR-100 : 60,000 32x32 colour images in 100 classes . 600 images each. There are 500 training images and 100 testing images per class. 20 superclasses.
PASCAL VOC : 11,530 images in 20 classes.
MNIST - handwritten digits, 70,000 28x28 greyscale images in 10 classes. 60,000 training and 10,000 test examples. (1998)
@Kaggle Cats & Dogs 25,000 images
@Kaggle Fashion MNIST Zalando : 70,000 greyscale images in 10 classes. 60,000 training and 10,000 test images
@Kaggle Fruits-360 : 62,037 100x100 images in 90 classes (one fruit per image). 46,371 training and 15,563 test images.
VGG datasets
ResNet - Deep Residual Learning for Image Recognition . Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun. Dec 2015.
MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications . Andrew G. Howard, Menglong Zhu, Bo Chen, Dmitry Kalenichenko, Weijun Wang, Tobias Weyand, Marco Andreetto, Hartwig Adam. April 2017.
MobileNetV2: Inverted Residuals and Linear Bottlenecks . Mark Sandler, Andrew Howard, Menglong Zhu, Andrey Zhmoginov, Liang-Chieh Chen. Apr 2018.
MnasNet: Platform-Aware Neural Architecture Search for Mobile . Mingxing Tan, Bo Chen, Ruoming Pang, Vijay Vasudevan, Mark Sandler, Andrew Howard, Quoc V. Le. Jul. 2018.
EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks . Mingxing Tan, Quoc Le. May 2019.
cnn-playground's People
Contributors
Stargazers