Comparing VGG16, VGG19, ResNet50 and Xception on the Cifar10 dataset. The experiments are divided into three groups:
- no data augmentation
- standard augmentations (translation, rotation, etc.)
- data augmentations using Histogram equalization.
Check Evaluation.pdf for results.
python 3.6^
jupyter
python3 -m pip install --upgrade pip
python3 -m pip install jupyter
jupyter-tensorboard
pip install jupyter-tensorboard
tensorflow
pip install tensorflow
numpy
pip install numpy
scikit-image
pip install scikit-image
keras
pip install Keras