fix readme and modify config comments

Author: MengzhangLI

configs/maskformer/README.md

@@ -55,6 +55,6 @@ pip install "mmdet>=3.0.0rc4"
 Note:
 
 - All experiments of MaskFormer are implemented with 8 V100 (32G) GPUs with 2 samplers per GPU.
-- The results of MaskFormer are relatively not stable, variance of `R-101-D32` is 44.7 to 46.0, variance of `Swin-S` is 49.0 to 49.8.
+- The results of MaskFormer are relatively not stable.  The accuracy (mIoU) of model with `R-101-D32` is from 44.7 to 46.0, and with `Swin-S` is from 49.0 to 49.8.
 - The ResNet backbones utilized in MaskFormer models are standard `ResNet` rather than `ResNetV1c`.
-- `MultiScaleFlipAug` is not supported in MMSegmentation 1.x version yet, we would add "ms+flip" results as soon as possible.
+- Test time augmentation is not supported in MMSegmentation 1.x version yet, we would add "ms+flip" results as soon as possible.

configs/maskformer/maskformer_r50-d32_8xb2-160k_ade20k-512x512.py

@@ -31,7 +31,8 @@
         init_cfg=dict(type='Pretrained', checkpoint='torchvision://resnet50')),
     decode_head=dict(
         type='MaskFormerHead',
-        in_channels=[256, 512, 1024, 2048],  # pass to pixel_decoder inside
+        in_channels=[256, 512, 1024,
+                     2048],  # input channels of pixel_decoder modules
         feat_channels=256,
         in_index=[0, 1, 2, 3],
         num_classes=150,

configs/maskformer/maskformer_swin-s_upernet_8xb2-160k_ade20k-512x512.py

@@ -27,7 +27,8 @@
         init_cfg=dict(type='Pretrained', checkpoint=checkpoint_file)),
     decode_head=dict(
         type='MaskFormerHead',
-        in_channels=[96, 192, 384, 768],  # pass to pixel_decoder inside
+        in_channels=[96, 192, 384,
+                     768],  # input channels of pixel_decoder modules
     ))
 
 # optimizer

configs/maskformer/maskformer_swin-t_upernet_8xb2-160k_ade20k-512x512.py

@@ -28,7 +28,8 @@
         init_cfg=dict(type='Pretrained', checkpoint=checkpoint_file)),
     decode_head=dict(
         type='MaskFormerHead',
-        in_channels=[96, 192, 384, 768],  # pass to pixel_decoder inside
+        in_channels=[96, 192, 384,
+                     768],  # input channels of pixel_decoder modules
     ))
 
 # optimizer