fixed merge

src/evaluate.py

@@ -109,7 +109,6 @@ def getCellsEvalutions(points, pointsPrediction, thresholds=10, minDist=20):
     # the cost is the distance between the points
     # the assignment problem is solved using the Hungarian algorithm
     # the Hungarian algorithm is a special case of the linear sum assignment problem
-
     # the Hungarian algorithm returns two arrays
     # row_ind is an array of row indices
     # col_ind is an array of column indices

src/models.py

@@ -94,20 +94,20 @@ class SegModel(pl.LightningModule):
 
         self.log('val_loss', loss)
 
-        if batch_nb == 0:
-            out = torch.sigmoid(out)[0].cpu().numpy()
+        # if batch_nb == 0:
+        #     out = torch.sigmoid(out)[0].cpu().numpy()
 
-            thresholds = [0.5, 0.5, 0.6, 0.6, 0.7]
-            thresholdedMasks = np.zeros_like(out)
-            for j in range(len(thresholds)):
-                thresholdedMasks[j] = out[j] > thresholds[j]
+        #         thresholds = [0.5, 0.5, 0.6, 0.6, 0.7]
+        #         thresholdedMasks = np.zeros_like(out)
+        #         for j in range(len(thresholds)):
+        #         thresholdedMasks[j] = out[j] > thresholds[j]
 
-            fig, axes = plt.subplots(1, 5, figsize=(10, 2), dpi=200)
-            axes = axes.flatten()
+        #         fig, axes = plt.subplots(1, 5, figsize=(10, 2), dpi=200)
+        #         axes = axes.flatten()
 
             # create new masks thresholded
 
-            for j, mask in enumerate(out):
+            # for j, mask in enumerate(out):
                 # print(mask.shape)
                 # print(f"{dataset_test.id2cat[j]}: {mask.min()}, {mask.max()}")
 
@@ -124,14 +124,14 @@ class SegModel(pl.LightningModule):
 #                     mask = cv2.threshold(mask, 127, 255, cv2.THRESH_BINARY)[1]
 #                     # mask = cv2.morphologyEx(mask, cv2.MORPH_OPEN, kernel)
 
-                axes[j].imshow(mask)
-                axes[j].axis('off')
+                # axes[j].imshow(mask)
+                # axes[j].axis('off')
                 # axes[j].set_title(f"{dataset_test.id2cat[j]}")
 
-            self.logger.experiment.add_figure("masks",
-                                              fig, global_step=self.global_step)
+            # self.logger.experiment.add_figure("masks",
+            #                                   fig, global_step=self.global_step)
 
-            plt.close(fig)
+            # plt.close(fig)
 
         return {'val_loss': loss}