Internal change

PiperOrigin-RevId: 481231095

Author: tensorflower-gardener

official/projects/panoptic/tasks/panoptic_maskrcnn.py

@@ -244,7 +244,8 @@ def build_metrics(self, training: bool = True) -> List[
             dtype=tf.float32)
 
     else:
-      self._build_coco_metrics()
+      if self.task_config.use_coco_metrics:
+        self._build_coco_metrics()
 
       rescale_predictions = (not self.task_config.validation_data.parser
                              .segmentation_resize_eval_groundtruth)
@@ -366,24 +367,25 @@ def validation_step(self,
         training=False)
 
     logs = {self.loss: 0}
-    coco_model_outputs = {
-        'detection_masks': outputs['detection_masks'],
-        'detection_boxes': outputs['detection_boxes'],
-        'detection_scores': outputs['detection_scores'],
-        'detection_classes': outputs['detection_classes'],
-        'num_detections': outputs['num_detections'],
-        'source_id': labels['groundtruths']['source_id'],
-        'image_info': labels['image_info']
-    }
+    if self._task_config.use_coco_metrics:
+      coco_model_outputs = {
+          'detection_masks': outputs['detection_masks'],
+          'detection_boxes': outputs['detection_boxes'],
+          'detection_scores': outputs['detection_scores'],
+          'detection_classes': outputs['detection_classes'],
+          'num_detections': outputs['num_detections'],
+          'source_id': labels['groundtruths']['source_id'],
+          'image_info': labels['image_info']
+      }
+      logs.update(
+          {self.coco_metric.name: (labels['groundtruths'], coco_model_outputs)})
+
     segmentation_labels = {
         'masks': labels['groundtruths']['gt_segmentation_mask'],
         'valid_masks': labels['groundtruths']['gt_segmentation_valid_mask'],
         'image_info': labels['image_info']
     }
 
-    logs.update(
-        {self.coco_metric.name: (labels['groundtruths'], coco_model_outputs)})
-
     self.segmentation_perclass_iou_metric.update_state(
         segmentation_labels, outputs['segmentation_outputs'])
 
@@ -400,15 +402,18 @@ def validation_step(self,
 
   def aggregate_logs(self, state=None, step_outputs=None):
     if state is None:
-      self.coco_metric.reset_states()
       self.segmentation_perclass_iou_metric.reset_states()
-      state = [self.coco_metric, self.segmentation_perclass_iou_metric]
+      state = [self.segmentation_perclass_iou_metric]
+      if self.task_config.use_coco_metrics:
+        self.coco_metric.reset_states()
+        state.append(self.coco_metric)
       if self.task_config.model.generate_panoptic_masks:
-        state += [self.panoptic_quality_metric]
+        self.panoptic_quality_metric.reset_states()
+        state.append(self.panoptic_quality_metric)
 
-    self.coco_metric.update_state(
-        step_outputs[self.coco_metric.name][0],
-        step_outputs[self.coco_metric.name][1])
+    if self.task_config.use_coco_metrics:
+      self.coco_metric.update_state(step_outputs[self.coco_metric.name][0],
+                                    step_outputs[self.coco_metric.name][1])
 
     if self.task_config.model.generate_panoptic_masks:
       self.panoptic_quality_metric.update_state(