##VGG16 model for Keras
This is the Keras model of the 16-layer network used by the VGG team in the ILSVRC-2014 competition.
It has been obtained by directly converting the Caffe model provived by the authors.
Details about the network architecture can be found in the following arXiv paper:
Very Deep Convolutional Networks for Large-Scale Image Recognition
K. Simonyan, A. Zisserman
| /** | |
| * Creates a 'ghost' bitmap version of the given source drawable (ideally a BitmapDrawable). | |
| * In the ghost bitmap, the RGB values take on the values from the 'color' argument, while | |
| * the alpha values are derived from the source's grayscaled RGB values. The effect is that | |
| * you can see through darker parts of the source bitmap, while lighter parts show up as | |
| * the given color. The 'invert' argument inverts the computation of alpha values, and looks | |
| * best when the given color is a dark. | |
| */ | |
| private Bitmap createGhostIcon(Drawable src, int color, boolean invert) { | |
| int width = src.getIntrinsicWidth(); |