enable automatic mixed precision training

demo.py

@@ -20,6 +20,13 @@
 parser.add_argument('--style_seg_path', default=[])
 parser.add_argument('--output_image_path', default='./results/example1.png')
 parser.add_argument('--cuda', type=int, default=1, help='Enable CUDA.')
+parser.add_argument("--export_onnx", type=str, help="export ONNX model to a given file")
+parser.add_argument("--engine", type=str, help="run serialized TRT engine")
+parser.add_argument("--onnx", type=str, help="run ONNX model via TRT")
+parser.add_argument('--verbose', action='store_true', default = False, help='toggles verbose')
+parser.add_argument("-d", "--data_type", default=32, type=int, choices=[8, 16, 32], help="Supported data type i.e. 8, 16, 32 bit")
+
+
 args = parser.parse_args()
 
 # Load model
@@ -32,7 +39,7 @@
 
 if args.cuda:
     p_wct.cuda(0)
-
+    
 process_stylization.stylization(
     p_wct=p_wct,
     content_image_path=args.content_image_path,
@@ -41,4 +48,5 @@
     style_seg_path=args.style_seg_path,
     output_image_path=args.output_image_path,
     cuda=args.cuda,
+    args=args
 )

demo.sh

@@ -7,4 +7,4 @@ axel -n 1 https://vignette.wikia.nocookie.net/strangerthings8338/images/e/e0/Wik
 convert -resize 25% content1.png content1.png;
 convert -resize 50% style1.png style1.png;
 cd ..;
-python demo.py;
+python demo.py $@; 

photo_wct.py

@@ -23,7 +23,7 @@ def __init__(self):
         self.d3 = VGGDecoder(3)
         self.e4 = VGGEncoder(4)
         self.d4 = VGGDecoder(4)
-
+                
     def transform(self, cont_img, styl_img, cont_seg, styl_seg):
         self.__compute_label_info(cont_seg, styl_seg)
 
@@ -53,8 +53,15 @@ def transform(self, cont_img, styl_img, cont_seg, styl_seg):
         csF1 = self.__feature_wct(cF1, sF1, cont_seg, styl_seg)
         Im1 = self.d1(csF1)
         return Im1
+
+    def forward(self, args):
+        [cont_img, styl_img, cont_seg, styl_seg] = args
+        print (cont_img, styl_img, cont_seg, styl_seg)
+        self.transform(cont_img, styl_img, cont_seg, styl_seg)
 
     def __compute_label_info(self, cont_seg, styl_seg):
+        cont_seg=cont_seg.numpy()
+        styl_seg=styl_seg.numpy()
         if cont_seg.size == False or styl_seg.size == False:
             return
         max_label = np.max(cont_seg) + 1
@@ -69,6 +76,8 @@ def __compute_label_info(self, cont_seg, styl_seg):
             self.label_indicator[l] = is_valid(o_cont_mask[0].size, o_styl_mask[0].size)
 
     def __feature_wct(self, cont_feat, styl_feat, cont_seg, styl_seg):
+        cont_seg = cont_seg.numpy()
+        styl_seg = styl_seg.numpy()
         cont_c, cont_h, cont_w = cont_feat.size(0), cont_feat.size(1), cont_feat.size(2)
         styl_c, styl_h, styl_w = styl_feat.size(0), styl_feat.size(1), styl_feat.size(2)
         cont_feat_view = cont_feat.view(cont_c, -1).clone()

process_stylization.py

@@ -8,8 +8,10 @@
 import time
 
 import numpy as np
+import torch
 from PIL import Image
 from torch.autograd import Variable
+from torch.onnx import export
 import torchvision.transforms as transforms
 import torchvision.utils as utils
 
@@ -32,8 +34,7 @@ def __exit__(self, exc_type, exc_value, exc_tb):
         print(self.msg % (time.time() - self.start_time))
 
 
-def stylization(p_wct, content_image_path, style_image_path, content_seg_path, style_seg_path, output_image_path,
-                cuda):
+def stylization(p_wct, content_image_path, style_image_path, content_seg_path, style_seg_path, output_image_path, cuda, args):
     # Load image
     cont_img = Image.open(content_image_path).convert('RGB')
     styl_img = Image.open(style_image_path).convert('RGB')
@@ -52,12 +53,16 @@ def stylization(p_wct, content_image_path, style_image_path, content_seg_path, s
         styl_img = styl_img.cuda(0)
         p_wct.cuda(0)
 
-    cont_img = Variable(cont_img, volatile=True)
-    styl_img = Variable(styl_img, volatile=True)
-
-    cont_seg = np.asarray(cont_seg)
-    styl_seg = np.asarray(styl_seg)
-
+    cont_img = Variable(cont_img, requires_grad=False)
+    styl_img = Variable(styl_img, requires_grad=False)
+    cont_seg = torch.FloatTensor(np.asarray(cont_seg))
+    styl_seg = torch.FloatTensor(np.asarray(styl_seg))
+
+    if args.export_onnx:
+        assert args.export_onnx.endswith(".onnx"), "Export model file should end with .onnx"
+        export(p_wct, [cont_img, styl_img, cont_seg, styl_seg],
+               f=args.export_onnx, verbose=args.verbose)
+
     with Timer("Elapsed time in stylization: %f"):
         stylized_img = p_wct.transform(cont_img, styl_img, cont_seg, styl_seg)
     utils.save_image(stylized_img.data.cpu().float(), output_image_path, nrow=1)