Add documentation and README, use Upsampling2D rather than image Resizing layer

main.py

@@ -2,6 +2,7 @@ import fast_depth_functional as fd
 
 if __name__ == '__main__':
     fd.fix_windows_gpu()
-    model = fd.load_model('fast_depth_nyu_v2_224_224_3_e1')
+    model = fd.mobilenet_nnconv5_no_bn(weights='imagenet')
     fd.compile(model)
+    fd.train(existing_model=model, save_file='../fast-depth-experimental')
     fd.evaluate(model)