[Fix] Fix several config file errors in 2.0 (New) (#1994)

* [Fix] Fix several config file errors in 2.0

* change _base_ config file name in configs

configs/bisenetv1/bisenetv1_r101-d32_4xb4-160k_coco-stuff164k-512x512.py

@@ -5,6 +5,7 @@ _base_ = [
 ]
 crop_size = (512, 512)
 data_preprocessor = dict(size=crop_size)
+norm_cfg = dict(type='SyncBN', requires_grad=True)
 model = dict(
     data_preprocessor=data_preprocessor,
     backbone=dict(
@@ -14,8 +15,30 @@ model = dict(
         backbone_cfg=dict(type='ResNet', depth=101)),
     decode_head=dict(in_channels=1024, channels=1024, num_classes=171),
     auxiliary_head=[
-        dict(in_channels=512, channels=256, num_classes=171),
+        dict(
-        dict(in_channels=512, channels=256, num_classes=171),
+            type='FCNHead',
+            in_channels=512,
+            channels=256,
+            num_convs=1,
+            num_classes=171,
+            in_index=1,
+            norm_cfg=norm_cfg,
+            concat_input=False,
+            align_corners=False,
+            loss_decode=dict(
+                type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0)),
+        dict(
+            type='FCNHead',
+            in_channels=512,
+            channels=256,
+            num_convs=1,
+            num_classes=171,
+            in_index=2,
+            norm_cfg=norm_cfg,
+            concat_input=False,
+            align_corners=False,
+            loss_decode=dict(
+                type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0)),
     ])
 param_scheduler = [
     dict(type='LinearLR', by_epoch=False, start_factor=0.1, begin=0, end=1000),

configs/bisenetv1/bisenetv1_r18-d32_4xb4-160k_coco-stuff164k-512x512.py

@@ -5,12 +5,35 @@ _base_ = [
 ]
 crop_size = (512, 512)
 data_preprocessor = dict(size=crop_size)
+norm_cfg = dict(type='SyncBN', requires_grad=True)
 model = dict(
     data_preprocessor=data_preprocessor,
     decode_head=dict(num_classes=171),
     auxiliary_head=[
-        dict(num_classes=171),
+        dict(
-        dict(num_classes=171),
+            type='FCNHead',
+            in_channels=128,
+            channels=64,
+            num_convs=1,
+            num_classes=171,
+            in_index=1,
+            norm_cfg=norm_cfg,
+            concat_input=False,
+            align_corners=False,
+            loss_decode=dict(
+                type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0)),
+        dict(
+            type='FCNHead',
+            in_channels=128,
+            channels=64,
+            num_convs=1,
+            num_classes=171,
+            in_index=2,
+            norm_cfg=norm_cfg,
+            concat_input=False,
+            align_corners=False,
+            loss_decode=dict(
+                type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0)),
     ])
 param_scheduler = [
     dict(type='LinearLR', by_epoch=False, start_factor=0.1, begin=0, end=1000),

configs/bisenetv1/bisenetv1_r50-d32_4xb4-160k_coco-stuff164k-512x512.py

@@ -5,6 +5,7 @@ _base_ = [
 ]
 crop_size = (512, 512)
 data_preprocessor = dict(size=crop_size)
+norm_cfg = dict(type='SyncBN', requires_grad=True)
 model = dict(
     data_preprocessor=data_preprocessor,
     backbone=dict(
@@ -14,8 +15,30 @@ model = dict(
         backbone_cfg=dict(type='ResNet', depth=50)),
     decode_head=dict(in_channels=1024, channels=1024, num_classes=171),
     auxiliary_head=[
-        dict(in_channels=512, channels=256, num_classes=171),
+        dict(
-        dict(in_channels=512, channels=256, num_classes=171),
+            type='FCNHead',
+            in_channels=512,
+            channels=256,
+            num_convs=1,
+            num_classes=171,
+            in_index=1,
+            norm_cfg=norm_cfg,
+            concat_input=False,
+            align_corners=False,
+            loss_decode=dict(
+                type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0)),
+        dict(
+            type='FCNHead',
+            in_channels=512,
+            channels=256,
+            num_convs=1,
+            num_classes=171,
+            in_index=2,
+            norm_cfg=norm_cfg,
+            concat_input=False,
+            align_corners=False,
+            loss_decode=dict(
+                type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0)),
     ])
 param_scheduler = [
     dict(type='LinearLR', by_epoch=False, start_factor=0.1, begin=0, end=1000),

configs/deeplabv3plus/deeplabv3plus_r101-d8_4xb4-80k_loveda-512x512.py

@@ -1,6 +1,2 @@
 _base_ = './deeplabv3plus_r50-d8_4xb4-80k_loveda-512x512.py'
-model = dict(
+model = dict(pretrained='open-mmlab://resnet101_v1c', backbone=dict(depth=101))
-    backbone=dict(
-        depth=101,
-        init_cfg=dict(
-            type='Pretrained', checkpoint='open-mmlab://resnet101_v1c')))

configs/deeplabv3plus/deeplabv3plus_r18-d8_4xb4-80k_loveda-512x512.py

@@ -1,9 +1,7 @@
 _base_ = './deeplabv3plus_r50-d8_4xb4-80k_loveda-512x512.py'
 model = dict(
-    backbone=dict(
+    pretrained='open-mmlab://resnet18_v1c',
-        depth=18,
+    backbone=dict(depth=18),
-        init_cfg=dict(
-            type='Pretrained', checkpoint='open-mmlab://resnet18_v1c')),
     decode_head=dict(
         c1_in_channels=64,
         c1_channels=12,

configs/hrnet/fcn_hr18s_4xb4-80k_loveda-512x512.py

@@ -1,9 +1,7 @@
 _base_ = './fcn_hr18_4xb4-80k_loveda-512x512.py'
 model = dict(
+    pretrained='open-mmlab://msra/hrnetv2_w18_small',
     backbone=dict(
-        init_cfg=dict(
-            type='Pretrained',
-            checkpoint='open-mmlab://msra/hrnetv2_w18_small'),
         extra=dict(
             stage1=dict(num_blocks=(2, )),
             stage2=dict(num_blocks=(2, 2)),

configs/hrnet/fcn_hr48_4xb4-80k_loveda-512x512.py

@@ -1,8 +1,7 @@
 _base_ = './fcn_hr18_4xb4-80k_loveda-512x512.py'
 model = dict(
+    pretrained='open-mmlab://msra/hrnetv2_w48',
     backbone=dict(
-        init_cfg=dict(
-            type='Pretrained', checkpoint='open-mmlab://msra/hrnetv2_w48'),
         extra=dict(
             stage2=dict(num_channels=(48, 96)),
             stage3=dict(num_channels=(48, 96, 192)),

configs/pspnet/pspnet_r101-d8_4xb4-80k_loveda-512x512.py

@@ -1,6 +1,2 @@
 _base_ = './pspnet_r50-d8_4xb4-80k_loveda-512x512.py'
-model = dict(
+model = dict(pretrained='open-mmlab://resnet101_v1c', backbone=dict(depth=101))
-    backbone=dict(
-        depth=101,
-        init_cfg=dict(
-            type='Pretrained', checkpoint='open-mmlab://resnet101_v1c')))

configs/pspnet/pspnet_r18-d8_4xb4-80k_loveda-512x512.py

@@ -1,9 +1,7 @@
 _base_ = './pspnet_r50-d8_4xb4-80k_loveda-512x512.py'
 model = dict(
-    backbone=dict(
+    pretrained='open-mmlab://resnet18_v1c',
-        depth=18,
+    backbone=dict(depth=18),
-        init_cfg=dict(
-            type='Pretrained', checkpoint='open-mmlab://resnet18_v1c')),
     decode_head=dict(
         in_channels=512,
         channels=128,

configs/swin/swin-large-patch4-window12-in22k-384x384-pre_upernet_8xb2-160k_ade20k-512x512.py

@@ -0,0 +1,10 @@
+_base_ = [
+    'swin-large-patch4-window7-in22k-pre_upernet_'
+    '8xb2-160k_ade20k-512x512.py'
+]
+checkpoint_file = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/swin/swin_large_patch4_window12_384_22k_20220412-6580f57d.pth'  # noqa
+model = dict(
+    backbone=dict(
+        init_cfg=dict(type='Pretrained', checkpoint=checkpoint_file),
+        pretrain_img_size=384,
+        window_size=12))

configs/swin/swin-large-patch4-window7-in22k-pre_upernet_8xb2-160k_ade20k-512x512.py

@@ -0,0 +1,15 @@
+_base_ = [
+    'swin-tiny-patch4-window7-in1k-pre_upernet_8xb2-160k_'
+    'ade20k-512x512.py'
+]
+checkpoint_file = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/swin/swin_large_patch4_window7_224_22k_20220412-aeecf2aa.pth'  # noqa
+model = dict(
+    backbone=dict(
+        init_cfg=dict(type='Pretrained', checkpoint=checkpoint_file),
+        pretrain_img_size=224,
+        embed_dims=192,
+        depths=[2, 2, 18, 2],
+        num_heads=[6, 12, 24, 48],
+        window_size=7),
+    decode_head=dict(in_channels=[192, 384, 768, 1536], num_classes=150),
+    auxiliary_head=dict(in_channels=768, num_classes=150))

configs/upernet/upernet_r18_4xb2-40k_cityscapes-512x1024.py

@@ -0,0 +1,6 @@
+_base_ = './upernet_r50_4xb2-40k_cityscapes-512x1024.py'
+model = dict(
+    pretrained='open-mmlab://resnet18_v1c',
+    backbone=dict(depth=18),
+    decode_head=dict(in_channels=[64, 128, 256, 512]),
+    auxiliary_head=dict(in_channels=256))

configs/upernet/upernet_r18_4xb2-80k_cityscapes-512x1024.py

@@ -0,0 +1,6 @@
+_base_ = './upernet_r50_4xb2-80k_cityscapes-512x1024.py'
+model = dict(
+    pretrained='open-mmlab://resnet18_v1c',
+    backbone=dict(depth=18),
+    decode_head=dict(in_channels=[64, 128, 256, 512]),
+    auxiliary_head=dict(in_channels=256))

configs/upernet/upernet_r18_4xb4-160k_ade20k-512x512.py

@@ -0,0 +1,9 @@
+_base_ = [
+    '../_base_/models/upernet_r50.py', '../_base_/datasets/ade20k.py',
+    '../_base_/default_runtime.py', '../_base_/schedules/schedule_160k.py'
+]
+model = dict(
+    pretrained='open-mmlab://resnet18_v1c',
+    backbone=dict(depth=18),
+    decode_head=dict(in_channels=[64, 128, 256, 512], num_classes=150),
+    auxiliary_head=dict(in_channels=256, num_classes=150))

configs/upernet/upernet_r18_4xb4-20k_voc12aug-512x512.py

@@ -0,0 +1,10 @@
+_base_ = [
+    '../_base_/models/upernet_r50.py',
+    '../_base_/datasets/pascal_voc12_aug.py', '../_base_/default_runtime.py',
+    '../_base_/schedules/schedule_20k.py'
+]
+model = dict(
+    pretrained='open-mmlab://resnet18_v1c',
+    backbone=dict(depth=18),
+    decode_head=dict(in_channels=[64, 128, 256, 512], num_classes=21),
+    auxiliary_head=dict(in_channels=256, num_classes=21))

configs/upernet/upernet_r18_4xb4-40k_voc12aug-512x512.py

@@ -0,0 +1,10 @@
+_base_ = [
+    '../_base_/models/upernet_r50.py',
+    '../_base_/datasets/pascal_voc12_aug.py', '../_base_/default_runtime.py',
+    '../_base_/schedules/schedule_40k.py'
+]
+model = dict(
+    pretrained='open-mmlab://resnet18_v1c',
+    backbone=dict(depth=18),
+    decode_head=dict(in_channels=[64, 128, 256, 512], num_classes=21),
+    auxiliary_head=dict(in_channels=256, num_classes=21))

configs/upernet/upernet_r18_4xb4-80k_ade20k-512x512.py

@@ -0,0 +1,9 @@
+_base_ = [
+    '../_base_/models/upernet_r50.py', '../_base_/datasets/ade20k.py',
+    '../_base_/default_runtime.py', '../_base_/schedules/schedule_80k.py'
+]
+model = dict(
+    pretrained='open-mmlab://resnet18_v1c',
+    backbone=dict(depth=18),
+    decode_head=dict(in_channels=[64, 128, 256, 512], num_classes=150),
+    auxiliary_head=dict(in_channels=256, num_classes=150))