Limber Cheng
Limber Cheng
I think U-Net, MobileNet, Xception, Inceptionv3, flownet(both 1.0 and 2.0) has real good performance.
chart_studio is required by examples but has not yet added in requirements. Add chart_studio.
By just setting arguments on kernel and other relating arguments, we implement 3D conv and 3d deconv net. But I got trouble binding random number with Generator. Could anyone give...
Updating the examples using the Animation class instead.
Conda could install the tensorflow-gpu in a very easy way which without any single setting. Add some new function: In utils: 1. Sequence 2. convert_all_kernels_in_model Add argument description for plot_model