[Enhancement] update segformer default pretrain to configs (open-mmlab#3153)

xiexinch · web-flow · commit cc74c5c3fee3 · 2023-06-30T16:40:13.000+08:00
diff --git a/configs/segformer/segformer_mit-b0_8xb1-160k_cityscapes-1024x1024.py b/configs/segformer/segformer_mit-b0_8xb1-160k_cityscapes-1024x1024.py
@@ -5,10 +5,10 @@
 ]
 crop_size = (1024, 1024)
 data_preprocessor = dict(size=crop_size)
+checkpoint = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/segformer/mit_b0_20220624-7e0fe6dd.pth'  # noqa
 model = dict(
     data_preprocessor=data_preprocessor,
-    backbone=dict(
-        init_cfg=dict(type='Pretrained', checkpoint='pretrain/mit_b0.pth')),
+    backbone=dict(init_cfg=dict(type='Pretrained', checkpoint=checkpoint)),
     test_cfg=dict(mode='slide', crop_size=(1024, 1024), stride=(768, 768)))
 
 optim_wrapper = dict(
diff --git a/configs/segformer/segformer_mit-b0_8xb2-160k_ade20k-512x512.py b/configs/segformer/segformer_mit-b0_8xb2-160k_ade20k-512x512.py
@@ -4,9 +4,10 @@
 ]
 crop_size = (512, 512)
 data_preprocessor = dict(size=crop_size)
+checkpoint = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/segformer/mit_b0_20220624-7e0fe6dd.pth'  # noqa
 model = dict(
     data_preprocessor=data_preprocessor,
-    pretrained='pretrain/mit_b0.pth',
+    backbone=dict(init_cfg=dict(type='Pretrained', checkpoint=checkpoint)),
     decode_head=dict(num_classes=150))
 
 optim_wrapper = dict(
diff --git a/configs/segformer/segformer_mit-b1_8xb1-160k_cityscapes-1024x1024.py b/configs/segformer/segformer_mit-b1_8xb1-160k_cityscapes-1024x1024.py
@@ -1,7 +1,9 @@
 _base_ = ['./segformer_mit-b0_8xb1-160k_cityscapes-1024x1024.py']
 
+checkpoint = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/segformer/mit_b1_20220624-02e5a6a1.pth'  # noqa
+
 model = dict(
     backbone=dict(
-        init_cfg=dict(type='Pretrained', checkpoint='pretrain/mit_b1.pth'),
+        init_cfg=dict(type='Pretrained', checkpoint=checkpoint),
         embed_dims=64),
     decode_head=dict(in_channels=[64, 128, 320, 512]))
diff --git a/configs/segformer/segformer_mit-b1_8xb2-160k_ade20k-512x512.py b/configs/segformer/segformer_mit-b1_8xb2-160k_ade20k-512x512.py
@@ -1,8 +1,12 @@
 _base_ = ['./segformer_mit-b0_8xb2-160k_ade20k-512x512.py']
 
+checkpoint = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/segformer/mit_b1_20220624-02e5a6a1.pth'  # noqa
+
 # model settings
 model = dict(
-    pretrained='pretrain/mit_b1.pth',
     backbone=dict(
-        embed_dims=64, num_heads=[1, 2, 5, 8], num_layers=[2, 2, 2, 2]),
+        init_cfg=dict(type='Pretrained', checkpoint=checkpoint),
+        embed_dims=64,
+        num_heads=[1, 2, 5, 8],
+        num_layers=[2, 2, 2, 2]),
     decode_head=dict(in_channels=[64, 128, 320, 512]))
diff --git a/configs/segformer/segformer_mit-b2_8xb1-160k_cityscapes-1024x1024.py b/configs/segformer/segformer_mit-b2_8xb1-160k_cityscapes-1024x1024.py
@@ -1,8 +1,10 @@
 _base_ = ['./segformer_mit-b0_8xb1-160k_cityscapes-1024x1024.py']
 
+checkpoint = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/segformer/mit_b2_20220624-66e8bf70.pth'  # noqa
+
 model = dict(
     backbone=dict(
-        init_cfg=dict(type='Pretrained', checkpoint='pretrain/mit_b2.pth'),
+        init_cfg=dict(type='Pretrained', checkpoint=checkpoint),
         embed_dims=64,
         num_layers=[3, 4, 6, 3]),
     decode_head=dict(in_channels=[64, 128, 320, 512]))
diff --git a/configs/segformer/segformer_mit-b2_8xb2-160k_ade20k-512x512.py b/configs/segformer/segformer_mit-b2_8xb2-160k_ade20k-512x512.py
@@ -1,8 +1,12 @@
 _base_ = ['./segformer_mit-b0_8xb2-160k_ade20k-512x512.py']
 
+checkpoint = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/segformer/mit_b2_20220624-66e8bf70.pth'  # noqa
+
 # model settings
 model = dict(
-    pretrained='pretrain/mit_b2.pth',
     backbone=dict(
-        embed_dims=64, num_heads=[1, 2, 5, 8], num_layers=[3, 4, 6, 3]),
+        init_cfg=dict(type='Pretrained', checkpoint=checkpoint),
+        embed_dims=64,
+        num_heads=[1, 2, 5, 8],
+        num_layers=[3, 4, 6, 3]),
     decode_head=dict(in_channels=[64, 128, 320, 512]))
diff --git a/configs/segformer/segformer_mit-b3_8xb1-160k_cityscapes-1024x1024.py b/configs/segformer/segformer_mit-b3_8xb1-160k_cityscapes-1024x1024.py
@@ -1,8 +1,10 @@
 _base_ = ['./segformer_mit-b0_8xb1-160k_cityscapes-1024x1024.py']
 
+checkpoint = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/segformer/mit_b3_20220624-13b1141c.pth'  # noqa
+
 model = dict(
     backbone=dict(
-        init_cfg=dict(type='Pretrained', checkpoint='pretrain/mit_b3.pth'),
+        init_cfg=dict(type='Pretrained', checkpoint=checkpoint),
         embed_dims=64,
         num_layers=[3, 4, 18, 3]),
     decode_head=dict(in_channels=[64, 128, 320, 512]))
diff --git a/configs/segformer/segformer_mit-b3_8xb2-160k_ade20k-512x512.py b/configs/segformer/segformer_mit-b3_8xb2-160k_ade20k-512x512.py
@@ -1,8 +1,12 @@
 _base_ = ['./segformer_mit-b0_8xb2-160k_ade20k-512x512.py']
 
+checkpoint = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/segformer/mit_b3_20220624-13b1141c.pth'  # noqa
+
 # model settings
 model = dict(
-    pretrained='pretrain/mit_b3.pth',
     backbone=dict(
-        embed_dims=64, num_heads=[1, 2, 5, 8], num_layers=[3, 4, 18, 3]),
+        init_cfg=dict(type='Pretrained', checkpoint=checkpoint),
+        embed_dims=64,
+        num_heads=[1, 2, 5, 8],
+        num_layers=[3, 4, 18, 3]),
     decode_head=dict(in_channels=[64, 128, 320, 512]))
diff --git a/configs/segformer/segformer_mit-b4_8xb1-160k_cityscapes-1024x1024.py b/configs/segformer/segformer_mit-b4_8xb1-160k_cityscapes-1024x1024.py
@@ -1,8 +1,10 @@
 _base_ = ['./segformer_mit-b0_8xb1-160k_cityscapes-1024x1024.py']
 
+checkpoint = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/segformer/mit_b4_20220624-d588d980.pth'  # noqa
+
 model = dict(
     backbone=dict(
-        init_cfg=dict(type='Pretrained', checkpoint='pretrain/mit_b4.pth'),
+        init_cfg=dict(type='Pretrained', checkpoint=checkpoint),
         embed_dims=64,
         num_layers=[3, 8, 27, 3]),
     decode_head=dict(in_channels=[64, 128, 320, 512]))
diff --git a/configs/segformer/segformer_mit-b4_8xb2-160k_ade20k-512x512.py b/configs/segformer/segformer_mit-b4_8xb2-160k_ade20k-512x512.py
@@ -1,8 +1,12 @@
 _base_ = ['./segformer_mit-b0_8xb2-160k_ade20k-512x512.py']
 
+checkpoint = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/segformer/mit_b4_20220624-d588d980.pth'  # noqa
+
 # model settings
 model = dict(
-    pretrained='pretrain/mit_b4.pth',
     backbone=dict(
-        embed_dims=64, num_heads=[1, 2, 5, 8], num_layers=[3, 8, 27, 3]),
+        init_cfg=dict(type='Pretrained', checkpoint=checkpoint),
+        embed_dims=64,
+        num_heads=[1, 2, 5, 8],
+        num_layers=[3, 8, 27, 3]),
     decode_head=dict(in_channels=[64, 128, 320, 512]))
diff --git a/configs/segformer/segformer_mit-b5_8xb1-160k_cityscapes-1024x1024.py b/configs/segformer/segformer_mit-b5_8xb1-160k_cityscapes-1024x1024.py
@@ -1,8 +1,10 @@
 _base_ = ['./segformer_mit-b0_8xb1-160k_cityscapes-1024x1024.py']
 
+checkpoint = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/segformer/mit_b5_20220624-658746d9.pth'  # noqa
+
 model = dict(
     backbone=dict(
-        init_cfg=dict(type='Pretrained', checkpoint='pretrain/mit_b5.pth'),
+        init_cfg=dict(type='Pretrained', checkpoint=checkpoint),
         embed_dims=64,
         num_layers=[3, 6, 40, 3]),
     decode_head=dict(in_channels=[64, 128, 320, 512]))
diff --git a/configs/segformer/segformer_mit-b5_8xb2-160k_ade20k-512x512.py b/configs/segformer/segformer_mit-b5_8xb2-160k_ade20k-512x512.py
@@ -1,8 +1,12 @@
 _base_ = ['./segformer_mit-b0_8xb2-160k_ade20k-512x512.py']
 
+checkpoint = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/segformer/mit_b5_20220624-658746d9.pth'  # noqa
+
 # model settings
 model = dict(
-    pretrained='pretrain/mit_b5.pth',
     backbone=dict(
-        embed_dims=64, num_heads=[1, 2, 5, 8], num_layers=[3, 6, 40, 3]),
+        init_cfg=dict(type='Pretrained', checkpoint=checkpoint),
+        embed_dims=64,
+        num_heads=[1, 2, 5, 8],
+        num_layers=[3, 6, 40, 3]),
     decode_head=dict(in_channels=[64, 128, 320, 512]))
diff --git a/configs/segformer/segformer_mit-b5_8xb2-160k_ade20k-640x640.py b/configs/segformer/segformer_mit-b5_8xb2-160k_ade20k-640x640.py
@@ -1,5 +1,7 @@
 _base_ = ['./segformer_mit-b0_8xb2-160k_ade20k-512x512.py']
 
+checkpoint = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/segformer/mit_b5_20220624-658746d9.pth'  # noqa
+
 # dataset settings
 crop_size = (640, 640)
 data_preprocessor = dict(size=crop_size)
@@ -31,7 +33,9 @@
 # model settings
 model = dict(
     data_preprocessor=data_preprocessor,
-    pretrained='pretrain/mit_b5.pth',
     backbone=dict(
-        embed_dims=64, num_heads=[1, 2, 5, 8], num_layers=[3, 6, 40, 3]),
+        init_cfg=dict(type='Pretrained', checkpoint=checkpoint),
+        embed_dims=64,
+        num_heads=[1, 2, 5, 8],
+        num_layers=[3, 6, 40, 3]),
     decode_head=dict(in_channels=[64, 128, 320, 512]))
