Revert maskrcnn_r50_tv due to overflow issue

sungchul2 · Jul 10, 2024 · 2ea6ee8 · 2ea6ee8
1 parent 6157217
commit 2ea6ee8
Show file tree

Hide file tree

Showing 2 changed files with 87 additions and 28 deletions.
diff --git a/src/otx/recipe/instance_segmentation/maskrcnn_r50_tv.yaml b/src/otx/recipe/instance_segmentation/maskrcnn_r50_tv.yaml
@@ -1,23 +1,42 @@
-engine: ../_base_/engine/instance_segmentation.yaml
-model: ../_base_/model/instance_segmentation.yaml
-data: ../_base_/data/instance_segmentation.yaml
+model:
+  class_path: otx.algo.instance_segmentation.maskrcnn_tv.TVMaskRCNNR50
+  init_args:
+    label_info: 80
+
+    optimizer:
+      class_path: torch.optim.SGD
+      init_args:
+        lr: 0.007
+        momentum: 0.9
+        weight_decay: 0.001
+
+    scheduler:
+      class_path: otx.core.schedulers.LinearWarmupSchedulerCallable
+      init_args:
+        num_warmup_steps: 100
+        main_scheduler_callable:
+          class_path: lightning.pytorch.cli.ReduceLROnPlateau
+          init_args:
+            mode: max
+            factor: 0.1
+            patience: 4
+            monitor: val/map_50
+
+engine:
+  task: INSTANCE_SEGMENTATION
+  device: auto
 
 callback_monitor: val/map_50
 
+data: ../_base_/data/torchvision_base.yaml
 overrides:
   max_epochs: 100
   gradient_clip_val: 35.0
-  model:
-    class_path: otx.algo.instance_segmentation.maskrcnn_tv.TVMaskRCNNR50
-    init_args:
-      optimizer:
-        class_path: torch.optim.SGD
-        init_args:
-          lr: 0.007
-          momentum: 0.9
-          weight_decay: 0.001
-
   data:
+    task: INSTANCE_SEGMENTATION
+    stack_images: true
+    data_format: coco_instances
+    include_polygons: true
     train_subset:
       batch_size: 4
       num_workers: 8
@@ -41,6 +60,7 @@ overrides:
         - class_path: torchvision.transforms.v2.ToDtype
           init_args:
             dtype: ${as_torch_dtype:torch.float32}
+            scale: False
         - class_path: torchvision.transforms.v2.Normalize
           init_args:
             mean: [123.675, 116.28, 103.53]
@@ -52,16 +72,20 @@ overrides:
         - class_path: otx.core.data.transform_libs.torchvision.Resize
           init_args:
             keep_ratio: true
+            transform_bbox: false
+            transform_mask: false
             scale:
               - 1024
               - 1024
         - class_path: otx.core.data.transform_libs.torchvision.Pad
           init_args:
             pad_to_square: true
+            transform_mask: false
             is_numpy_to_tvtensor: true
         - class_path: torchvision.transforms.v2.ToDtype
           init_args:
             dtype: ${as_torch_dtype:torch.float32}
+            scale: False
         - class_path: torchvision.transforms.v2.Normalize
           init_args:
             mean: [123.675, 116.28, 103.53]
@@ -73,17 +97,21 @@ overrides:
         - class_path: otx.core.data.transform_libs.torchvision.Resize
           init_args:
             keep_ratio: true
+            transform_bbox: false
+            transform_mask: false
             scale:
               - 1024
               - 1024
         - class_path: otx.core.data.transform_libs.torchvision.Pad
           init_args:
             pad_to_square: true
+            transform_mask: false
             is_numpy_to_tvtensor: true
         - class_path: torchvision.transforms.v2.ToDtype
           init_args:
             dtype: ${as_torch_dtype:torch.float32}
+            scale: False
         - class_path: torchvision.transforms.v2.Normalize
           init_args:
             mean: [123.675, 116.28, 103.53]
-            std: [58.395, 57.12, 57.375]
+            std: [58.395, 57.12, 57.375]
diff --git a/src/otx/recipe/instance_segmentation/maskrcnn_r50_tv_tile.yaml b/src/otx/recipe/instance_segmentation/maskrcnn_r50_tv_tile.yaml
@@ -1,32 +1,52 @@
-engine: ../_base_/engine/instance_segmentation.yaml
-model: ../_base_/model/instance_segmentation.yaml
-data: ../_base_/data/instance_segmentation.yaml
+model:
+  class_path: otx.algo.instance_segmentation.maskrcnn_tv.TVMaskRCNNR50
+  init_args:
+    label_info: 80
+
+    optimizer:
+      class_path: torch.optim.SGD
+      init_args:
+        lr: 0.007
+        momentum: 0.9
+        weight_decay: 0.001
+
+    scheduler:
+      class_path: otx.core.schedulers.LinearWarmupSchedulerCallable
+      init_args:
+        num_warmup_steps: 100
+        main_scheduler_callable:
+          class_path: lightning.pytorch.cli.ReduceLROnPlateau
+          init_args:
+            mode: max
+            factor: 0.1
+            patience: 4
+            monitor: val/map_50
+
+engine:
+  task: INSTANCE_SEGMENTATION
+  device: auto
 
 callback_monitor: val/map_50
 
+data: ../_base_/data/torchvision_base.yaml
 overrides:
   max_epochs: 100
   gradient_clip_val: 35.0
-  model:
-    class_path: otx.algo.instance_segmentation.maskrcnn_tv.TVMaskRCNNR50
-    init_args:
-      optimizer:
-        class_path: torch.optim.SGD
-        init_args:
-          lr: 0.007
-          momentum: 0.9
-          weight_decay: 0.001
-
   data:
+    task: INSTANCE_SEGMENTATION
+    stack_images: true
     tile_config:
       enable_tiler: true
       enable_adaptive_tiling: true
+    data_format: coco_instances
+    include_polygons: true
     train_subset:
       batch_size: 4
       num_workers: 8
       transforms:
         - class_path: otx.core.data.transform_libs.torchvision.Resize
           init_args:
+            keep_ratio: false
             transform_bbox: true
             transform_mask: true
             scale:
@@ -43,6 +63,7 @@ overrides:
         - class_path: torchvision.transforms.v2.ToDtype
           init_args:
             dtype: ${as_torch_dtype:torch.float32}
+            scale: False
         - class_path: torchvision.transforms.v2.Normalize
           init_args:
             mean: [123.675, 116.28, 103.53]
@@ -53,16 +74,21 @@ overrides:
       transforms:
         - class_path: otx.core.data.transform_libs.torchvision.Resize
           init_args:
+            keep_ratio: false
+            transform_bbox: false
+            transform_mask: false
             scale:
               - 512
               - 512
         - class_path: otx.core.data.transform_libs.torchvision.Pad
           init_args:
             size_divisor: 32
+            transform_mask: false
             is_numpy_to_tvtensor: true
         - class_path: torchvision.transforms.v2.ToDtype
           init_args:
             dtype: ${as_torch_dtype:torch.float32}
+            scale: False
         - class_path: torchvision.transforms.v2.Normalize
           init_args:
             mean: [123.675, 116.28, 103.53]
@@ -73,17 +99,22 @@ overrides:
       transforms:
         - class_path: otx.core.data.transform_libs.torchvision.Resize
           init_args:
+            keep_ratio: false
+            transform_bbox: false
+            transform_mask: false
             scale:
               - 512
               - 512
         - class_path: otx.core.data.transform_libs.torchvision.Pad
           init_args:
             size_divisor: 32
+            transform_mask: false
             is_numpy_to_tvtensor: true
         - class_path: torchvision.transforms.v2.ToDtype
           init_args:
             dtype: ${as_torch_dtype:torch.float32}
+            scale: False
         - class_path: torchvision.transforms.v2.Normalize
           init_args:
             mean: [123.675, 116.28, 103.53]
-            std: [58.395, 57.12, 57.375]
+            std: [58.395, 57.12, 57.375]