From ff85f85dcca877a41340903ef6bd884b58af62e0 Mon Sep 17 00:00:00 2001
From: ZiyuanTonyZhang <ziyuantony@gmail.com>
Date: Sun, 17 Jun 2018 02:31:42 -0400
Subject: [PATCH 1/2] fix scaling issue by size change from net, i.e. Issues -
 error in predict.py #12

Former-commit-id: 4af6e54bd4e1c642282fe512d97aa5a09ce16fca
---
 predict.py | 32 +++++++++++++++++++++++++-------
 1 file changed, 25 insertions(+), 7 deletions(-)

diff --git a/predict.py b/predict.py
index d963b04..d3ef938 100644
--- a/predict.py
+++ b/predict.py
@@ -11,6 +11,8 @@ from unet import UNet
 from utils import resize_and_crop, normalize, split_img_into_squares, hwc_to_chw, merge_masks, dense_crf
 from utils import plot_img_and_mask
 
+from torchvision import transforms
+
 def predict_img(net,
                 full_img,
                 scale_factor=0.5,
@@ -40,11 +42,25 @@ def predict_img(net,
         output_left = net(X_left)
         output_right = net(X_right)
 
-        left_probs = F.sigmoid(output_left)
-        right_probs = F.sigmoid(output_right)
+        left_probs = F.sigmoid(output_left).squeeze(0)
+        right_probs = F.sigmoid(output_right).squeeze(0)
+
+        tf = transforms.Compose(
+            [
+                transforms.ToPILImage(),
+                transforms.Resize(img_height),
+                transforms.ToTensor()
+            ]
+        )
+
+        left_probs = tf(left_probs.cpu()).unsqueeze(0)
+        right_probs = tf(right_probs.cpu()).unsqueeze(0)
+
+
+        # left_probs = F.upsample(left_probs, size=(img_height, img_height))
+        # right_probs = F.upsample(right_probs, size=(img_height, img_height))
+
 
-        left_probs = F.upsample(left_probs, size=(img_height, img_height))
-        right_probs = F.upsample(right_probs, size=(img_height, img_height))
 
         left_mask_np = left_probs.squeeze().cpu().numpy()
         right_mask_np = right_probs.squeeze().cpu().numpy()
@@ -64,10 +80,11 @@ def get_args():
                         metavar='FILE',
                         help="Specify the file in which is stored the model"
                              " (default : 'MODEL.pth')")
-    parser.add_argument('--input', '-i', metavar='INPUT', nargs='+',
-                        help='filenames of input images', required=True)
+    parser.add_argument('--input', '-i', default=['test.jpg'], metavar='INPUT', nargs='+',
+                        help='filenames of input images', required=False)
+
     parser.add_argument('--output', '-o', metavar='INPUT', nargs='+',
-                        help='filenames of ouput images')
+                        help='filenames of ouput images',default=['opt.jpg'])
     parser.add_argument('--cpu', '-c', action='store_true',
                         help="Do not use the cuda version of the net",
                         default=False)
@@ -91,6 +108,7 @@ def get_args():
 
 def get_output_filenames(args):
     in_files = args.input
+    # in_files = 'img.jpg'
     out_files = []
 
     if not args.output:

From bc465b2d91a970e4d325e470d1a700722dccf5cd Mon Sep 17 00:00:00 2001
From: milesial <milesial@users.noreply.github.com>
Date: Sat, 23 Jun 2018 19:57:53 +0200
Subject: [PATCH 2/2] Fixed some details on predict + requirements

Former-commit-id: d04263be694bba9dc1cfeb96389a3ad2a4015841
---
 predict.py       | 21 +++++++--------------
 requirements.txt |  3 ++-
 2 files changed, 9 insertions(+), 15 deletions(-)
 mode change 100644 => 100755 predict.py

diff --git a/predict.py b/predict.py
old mode 100644
new mode 100755
index d3ef938..176e7a1
--- a/predict.py
+++ b/predict.py
@@ -33,7 +33,7 @@ def predict_img(net,
 
     X_left = torch.from_numpy(left_square).unsqueeze(0)
     X_right = torch.from_numpy(right_square).unsqueeze(0)
-
+    
     if use_gpu:
         X_left = X_left.cuda()
         X_right = X_right.cuda()
@@ -52,15 +52,9 @@ def predict_img(net,
                 transforms.ToTensor()
             ]
         )
-
-        left_probs = tf(left_probs.cpu()).unsqueeze(0)
-        right_probs = tf(right_probs.cpu()).unsqueeze(0)
-
-
-        # left_probs = F.upsample(left_probs, size=(img_height, img_height))
-        # right_probs = F.upsample(right_probs, size=(img_height, img_height))
-
-
+        
+        left_probs = tf(left_probs.cpu())
+        right_probs = tf(right_probs.cpu())
 
         left_mask_np = left_probs.squeeze().cpu().numpy()
         right_mask_np = right_probs.squeeze().cpu().numpy()
@@ -80,11 +74,11 @@ def get_args():
                         metavar='FILE',
                         help="Specify the file in which is stored the model"
                              " (default : 'MODEL.pth')")
-    parser.add_argument('--input', '-i', default=['test.jpg'], metavar='INPUT', nargs='+',
-                        help='filenames of input images', required=False)
+    parser.add_argument('--input', '-i', metavar='INPUT', nargs='+',
+                        help='filenames of input images', required=True)
 
     parser.add_argument('--output', '-o', metavar='INPUT', nargs='+',
-                        help='filenames of ouput images',default=['opt.jpg'])
+                        help='filenames of ouput images')
     parser.add_argument('--cpu', '-c', action='store_true',
                         help="Do not use the cuda version of the net",
                         default=False)
@@ -108,7 +102,6 @@ def get_args():
 
 def get_output_filenames(args):
     in_files = args.input
-    # in_files = 'img.jpg'
     out_files = []
 
     if not args.output:
diff --git a/requirements.txt b/requirements.txt
index 2d8ee43..d22f05c 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -2,4 +2,5 @@ matplotlib
 pydensecrf
 numpy
 Pillow
-torch
\ No newline at end of file
+torch
+torchvision