Add resnet and resnet-v2 18/34 weights trained with mnv4 small based recipe

Author: rwightman

timm/models/resnet.py

@@ -711,6 +711,9 @@ def _gcfg(url='', **kwargs):
         hf_hub_id='timm/',
         url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet18d_ra2-48a79e06.pth',
         first_conv='conv1.0'),
+    'resnet18d.ra4_e3600_r224_in1k': _rcfg(
+        hf_hub_id='timm/',
+        mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5), crop_pct=0.9),
     'resnet34.a1_in1k': _rcfg(
         hf_hub_id='timm/',
         url='https://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet34_a1_0-46f8f793.pth'),
@@ -724,6 +727,9 @@ def _gcfg(url='', **kwargs):
     'resnet34.bt_in1k': _ttcfg(
         hf_hub_id='timm/',
         url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet34-43635321.pth'),
+    'resnet34.ra4_e3600_r224_in1k': _rcfg(
+        hf_hub_id='timm/',
+        mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5), crop_pct=0.9),
     'resnet34d.ra2_in1k': _ttcfg(
         hf_hub_id='timm/',
         url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet34d_ra2-f8dcfcaf.pth',

timm/models/resnetv2.py

@@ -696,14 +696,20 @@ def _cfg(url='', **kwargs):
         hf_hub_id='timm/',
         num_classes=21843, custom_load=True),
 
-    'resnetv2_18.untrained': _cfg(
-        interpolation='bicubic', crop_pct=0.95),
-    'resnetv2_18d.untrained': _cfg(
-        interpolation='bicubic', crop_pct=0.95, first_conv='stem.conv1'),
-    'resnetv2_34.untrained': _cfg(
-        interpolation='bicubic', crop_pct=0.95),
-    'resnetv2_34d.untrained': _cfg(
-        interpolation='bicubic', crop_pct=0.95, first_conv='stem.conv1'),
+    'resnetv2_18.ra4_e3600_r224_in1k': _cfg(
+        hf_hub_id='timm/',
+        interpolation='bicubic', crop_pct=0.9, test_input_size=(3, 288, 288), test_crop_pct=1.0),
+    'resnetv2_18d.ra4_e3600_r224_in1k': _cfg(
+        hf_hub_id='timm/',
+        interpolation='bicubic', crop_pct=0.9, test_input_size=(3, 288, 288), test_crop_pct=1.0,
+        first_conv='stem.conv1'),
+    'resnetv2_34.ra4_e3600_r224_in1k': _cfg(
+        hf_hub_id='timm/',
+        interpolation='bicubic', crop_pct=0.9, test_input_size=(3, 288, 288), test_crop_pct=1.0),
+    'resnetv2_34d.ra4_e3600_r224_in1k': _cfg(
+        hf_hub_id='timm/',
+        interpolation='bicubic', crop_pct=0.9, test_input_size=(3, 288, 288), test_crop_pct=1.0,
+        first_conv='stem.conv1'),
     'resnetv2_50.a1h_in1k': _cfg(
         hf_hub_id='timm/',
         interpolation='bicubic', crop_pct=0.95, test_input_size=(3, 288, 288), test_crop_pct=1.0),