Merge pull request #3777 from harimkang/harimkan/refactor-recipe-anomaly

Refine recipes for anomaly tasks

src/otx/recipe/_base_/data/anomaly.yaml

@@ -0,0 +1,76 @@
+task: ANOMALY_CLASSIFICATION
+input_size: 256
+data_format: mvtec
+mem_cache_size: 1GB
+mem_cache_img_max_size: null
+image_color_channel: RGB
+stack_images: false
+unannotated_items_ratio: 0.0
+train_subset:
+  subset_name: train
+  transform_lib_type: TORCHVISION
+  to_tv_image: true
+  batch_size: 32
+  num_workers: 4
+  transforms:
+    - class_path: otx.core.data.transform_libs.torchvision.ResizetoLongestEdge
+      init_args:
+        size: $(input_size)
+        antialias: true
+    - class_path: otx.core.data.transform_libs.torchvision.PadtoSquare
+    - 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]
+  sampler:
+    class_path: torch.utils.data.RandomSampler
+
+val_subset:
+  subset_name: test
+  transform_lib_type: TORCHVISION
+  to_tv_image: true
+  batch_size: 32
+  num_workers: 4
+  transforms:
+    - class_path: otx.core.data.transform_libs.torchvision.ResizetoLongestEdge
+      init_args:
+        size: $(input_size)
+        antialias: true
+    - class_path: otx.core.data.transform_libs.torchvision.PadtoSquare
+    - 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]
+  sampler:
+    class_path: torch.utils.data.RandomSampler
+
+test_subset:
+  subset_name: test
+  transform_lib_type: TORCHVISION
+  to_tv_image: true
+  batch_size: 32
+  num_workers: 4
+  transforms:
+    - class_path: otx.core.data.transform_libs.torchvision.ResizetoLongestEdge
+      init_args:
+        size: $(input_size)
+        antialias: true
+    - class_path: otx.core.data.transform_libs.torchvision.PadtoSquare
+    - 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]
+  sampler:
+    class_path: torch.utils.data.RandomSampler

src/otx/recipe/anomaly_classification/padim.yaml

@@ -13,78 +13,11 @@ engine:
 
 callback_monitor: image_F1Score # this has no effect as Padim does not need to be trained
 
-data: ../_base_/data/torchvision_base.yaml
+data: ../_base_/data/anomaly.yaml
 overrides:
-  reset:
-    - data.train_subset.transforms
-    - data.val_subset.transforms
-    - data.test_subset.transforms
-
   precision: 32
   max_epochs: 1
   callbacks:
     - class_path: otx.algo.callbacks.adaptive_train_scheduling.AdaptiveTrainScheduling
       init_args:
         max_interval: 1
-
-  data:
-    task: ANOMALY_CLASSIFICATION
-    input_size: 256
-    data_format: mvtec
-    vpm_config:
-      use_bbox: true
-      use_point: false
-    train_subset:
-      batch_size: 32
-      num_workers: 4
-      transforms:
-        - class_path: otx.core.data.transform_libs.torchvision.ResizetoLongestEdge
-          init_args:
-            size: $(input_size)
-            antialias: true
-        - class_path: otx.core.data.transform_libs.torchvision.PadtoSquare
-        - 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]
-
-    val_subset:
-      batch_size: 32
-      num_workers: 4
-      subset_name: test
-      transforms:
-        - class_path: otx.core.data.transform_libs.torchvision.ResizetoLongestEdge
-          init_args:
-            size: $(input_size)
-            antialias: true
-        - class_path: otx.core.data.transform_libs.torchvision.PadtoSquare
-        - 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]
-
-    test_subset:
-      batch_size: 32
-      num_workers: 4
-      transforms:
-        - class_path: otx.core.data.transform_libs.torchvision.ResizetoLongestEdge
-          init_args:
-            size: $(input_size)
-            antialias: true
-        - class_path: otx.core.data.transform_libs.torchvision.PadtoSquare
-        - 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]

src/otx/recipe/anomaly_classification/stfpm.yaml

@@ -11,13 +11,8 @@ engine:
 
 callback_monitor: image_F1Score
 
-data: ../_base_/data/torchvision_base.yaml
+data: ../_base_/data/anomaly.yaml
 overrides:
-  reset:
-    - data.train_subset.transforms
-    - data.val_subset.transforms
-    - data.test_subset.transforms
-
   precision: 32
   max_epochs: 100
   callbacks:
@@ -27,62 +22,3 @@ overrides:
     - class_path: otx.algo.callbacks.adaptive_train_scheduling.AdaptiveTrainScheduling
       init_args:
         max_interval: 1
-
-  data:
-    task: ANOMALY_CLASSIFICATION
-    input_size: 256
-    data_format: mvtec
-    train_subset:
-      batch_size: 32
-      num_workers: 4
-      transforms:
-        - class_path: otx.core.data.transform_libs.torchvision.ResizetoLongestEdge
-          init_args:
-            size: $(input_size)
-            antialias: true
-        - class_path: otx.core.data.transform_libs.torchvision.PadtoSquare
-        - 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]
-
-    val_subset:
-      batch_size: 32
-      num_workers: 4
-      subset_name: test
-      transforms:
-        - class_path: otx.core.data.transform_libs.torchvision.ResizetoLongestEdge
-          init_args:
-            size: $(input_size)
-            antialias: true
-        - class_path: otx.core.data.transform_libs.torchvision.PadtoSquare
-        - 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]
-
-    test_subset:
-      batch_size: 32
-      num_workers: 4
-      transforms:
-        - class_path: otx.core.data.transform_libs.torchvision.ResizetoLongestEdge
-          init_args:
-            size: $(input_size)
-            antialias: true
-        - class_path: otx.core.data.transform_libs.torchvision.PadtoSquare
-        - 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]

src/otx/recipe/anomaly_detection/padim.yaml

@@ -13,13 +13,8 @@ engine:
 
 callback_monitor: image_F1Score # this has no effect as Padim does not need to be trained
 
-data: ../_base_/data/torchvision_base.yaml
+data: ../_base_/data/anomaly.yaml
 overrides:
-  reset:
-    - data.train_subset.transforms
-    - data.val_subset.transforms
-    - data.test_subset.transforms
-
   precision: 32
   max_epochs: 1
   callbacks:
@@ -29,62 +24,3 @@ overrides:
 
   data:
     task: ANOMALY_DETECTION
-    input_size: 256
-    data_format: mvtec
-    vpm_config:
-      use_bbox: true
-      use_point: false
-    train_subset:
-      batch_size: 32
-      num_workers: 4
-      transforms:
-        - class_path: otx.core.data.transform_libs.torchvision.ResizetoLongestEdge
-          init_args:
-            size: $(input_size)
-            antialias: true
-        - class_path: otx.core.data.transform_libs.torchvision.PadtoSquare
-        - 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]
-
-    val_subset:
-      batch_size: 32
-      num_workers: 4
-      subset_name: test
-      transforms:
-        - class_path: otx.core.data.transform_libs.torchvision.ResizetoLongestEdge
-          init_args:
-            size: $(input_size)
-            antialias: true
-        - class_path: otx.core.data.transform_libs.torchvision.PadtoSquare
-        - 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]
-
-    test_subset:
-      batch_size: 32
-      num_workers: 4
-      transforms:
-        - class_path: otx.core.data.transform_libs.torchvision.ResizetoLongestEdge
-          init_args:
-            size: $(input_size)
-            antialias: true
-        - class_path: otx.core.data.transform_libs.torchvision.PadtoSquare
-        - 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]

src/otx/recipe/anomaly_detection/stfpm.yaml

@@ -11,7 +11,7 @@ engine:
 
 callback_monitor: pixel_F1Score
 
-data: ../_base_/data/torchvision_base.yaml
+data: ../_base_/data/anomaly.yaml
 overrides:
   reset:
     - data.train_subset.transforms
@@ -30,59 +30,3 @@ overrides:
 
   data:
     task: ANOMALY_DETECTION
-    input_size: 256
-    data_format: mvtec
-    train_subset:
-      batch_size: 32
-      num_workers: 4
-      transforms:
-        - class_path: otx.core.data.transform_libs.torchvision.ResizetoLongestEdge
-          init_args:
-            size: $(input_size)
-            antialias: true
-        - class_path: otx.core.data.transform_libs.torchvision.PadtoSquare
-        - 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]
-
-    val_subset:
-      batch_size: 32
-      num_workers: 4
-      subset_name: test
-      transforms:
-        - class_path: otx.core.data.transform_libs.torchvision.ResizetoLongestEdge
-          init_args:
-            size: $(input_size)
-            antialias: true
-        - class_path: otx.core.data.transform_libs.torchvision.PadtoSquare
-        - 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]
-
-    test_subset:
-      batch_size: 32
-      num_workers: 4
-      transforms:
-        - class_path: otx.core.data.transform_libs.torchvision.ResizetoLongestEdge
-          init_args:
-            size: $(input_size)
-            antialias: true
-        - class_path: otx.core.data.transform_libs.torchvision.PadtoSquare
-        - 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]