diff --git a/configs/benchmarks/classification/_base_/datasets/cifar10.py b/configs/benchmarks/classification/_base_/datasets/cifar10.py
deleted file mode 100644
index 633ba92a..00000000
--- a/configs/benchmarks/classification/_base_/datasets/cifar10.py
+++ /dev/null
@@ -1,49 +0,0 @@
-# dataset settings
-data_source = 'CIFAR10'
-dataset_type = 'SingleViewDataset'
-img_norm_cfg = dict(mean=[0.4914, 0.4822, 0.4465], std=[0.2023, 0.1994, 0.201])
-train_pipeline = [
-    dict(type='RandomCrop', size=32, padding=4),
-    dict(type='RandomHorizontalFlip'),
-]
-test_pipeline = []
-
-# prefetch
-prefetch = False
-if not prefetch:
-    train_pipeline.extend(
-        [dict(type='ToTensor'),
-         dict(type='Normalize', **img_norm_cfg)])
-    test_pipeline.extend(
-        [dict(type='ToTensor'),
-         dict(type='Normalize', **img_norm_cfg)])
-
-# dataset summary
-data = dict(
-    samples_per_gpu=128,
-    workers_per_gpu=2,
-    train=dict(
-        type=dataset_type,
-        data_source=dict(
-            type=data_source,
-            data_prefix='data/cifar10',
-        ),
-        pipeline=train_pipeline,
-        prefetch=prefetch),
-    val=dict(
-        type=dataset_type,
-        data_source=dict(
-            type=data_source,
-            data_prefix='data/cifar10',
-        ),
-        pipeline=test_pipeline,
-        prefetch=prefetch),
-    test=dict(
-        type=dataset_type,
-        data_source=dict(
-            type=data_source,
-            data_prefix='data/cifar10',
-        ),
-        pipeline=test_pipeline,
-        prefetch=prefetch))
-evaluation = dict(interval=10, topk=(1, 5))
diff --git a/configs/benchmarks/classification/_base_/datasets/imagenet-swin.py b/configs/benchmarks/classification/_base_/datasets/imagenet-swin.py
index 1b96cf14..4b15785b 100644
--- a/configs/benchmarks/classification/_base_/datasets/imagenet-swin.py
+++ b/configs/benchmarks/classification/_base_/datasets/imagenet-swin.py
@@ -1,5 +1,6 @@
 # dataset settings
 dataset_type = 'ImageNet'
+data_root = 'data/imagenet/'
 preprocess_cfg = dict(
     # RGB format normalization parameters
     mean=[123.675, 116.28, 103.53],
@@ -57,7 +58,7 @@ train_dataloader = dict(
     num_workers=8,
     dataset=dict(
         type=dataset_type,
-        data_root='data/imagenet',
+        data_root=data_root,
         ann_file='meta/train.txt',
         data_prefix='train',
         pipeline=train_pipeline),
@@ -70,7 +71,7 @@ val_dataloader = dict(
     num_workers=5,
     dataset=dict(
         type=dataset_type,
-        data_root='data/imagenet',
+        data_root=data_root,
         ann_file='meta/val.txt',
         data_prefix='val',
         pipeline=test_pipeline),
diff --git a/configs/benchmarks/classification/_base_/datasets/imagenet.py b/configs/benchmarks/classification/_base_/datasets/imagenet.py
index 5bc9e260..60d4049c 100644
--- a/configs/benchmarks/classification/_base_/datasets/imagenet.py
+++ b/configs/benchmarks/classification/_base_/datasets/imagenet.py
@@ -21,7 +21,7 @@ train_dataloader = dict(
     num_workers=4,
     dataset=dict(
         type=dataset_type,
-        data_root='data/imagenet',
+        data_root=data_root,
         ann_file='meta/train.txt',
         data_prefix='train',
         pipeline=train_pipeline),
@@ -34,7 +34,7 @@ val_dataloader = dict(
     num_workers=4,
     dataset=dict(
         type=dataset_type,
-        data_root='data/imagenet',
+        data_root=data_root,
         ann_file='meta/val.txt',
         data_prefix='val',
         pipeline=test_pipeline),
diff --git a/configs/benchmarks/classification/_base_/datasets/imagenet_mocov3_ft.py b/configs/benchmarks/classification/_base_/datasets/imagenet_mocov3_ft.py
index f9ca06d2..28f449b5 100644
--- a/configs/benchmarks/classification/_base_/datasets/imagenet_mocov3_ft.py
+++ b/configs/benchmarks/classification/_base_/datasets/imagenet_mocov3_ft.py
@@ -57,7 +57,7 @@ train_dataloader = dict(
     num_workers=5,
     dataset=dict(
         type=dataset_type,
-        data_root='data/imagenet',
+        data_root=data_root,
         ann_file='meta/train.txt',
         data_prefix='train',
         pipeline=train_pipeline),
@@ -70,7 +70,7 @@ val_dataloader = dict(
     num_workers=5,
     dataset=dict(
         type=dataset_type,
-        data_root='data/imagenet',
+        data_root=data_root,
         ann_file='meta/val.txt',
         data_prefix='val',
         pipeline=test_pipeline),
diff --git a/configs/benchmarks/classification/cifar/resnet50_8xb128-steplr-350e_cifar10.py b/configs/benchmarks/classification/cifar/resnet50_8xb128-steplr-350e_cifar10.py
index c4294e48..b4e21c13 100644
--- a/configs/benchmarks/classification/cifar/resnet50_8xb128-steplr-350e_cifar10.py
+++ b/configs/benchmarks/classification/cifar/resnet50_8xb128-steplr-350e_cifar10.py
@@ -1,15 +1,20 @@
 _base_ = [
     '../_base_/models/resnet50.py',
-    '../_base_/datasets/cifar10.py',
+    'mmcls::_base_/datasets/cifar10_bs16.py',
     '../_base_/schedules/sgd_steplr-100e.py',
     '../_base_/default_runtime.py',
 ]
 
+# dataset settings
+train_dataloader = dict(batch_size=128)
+val_dataloader = dict(batch_size=128)
+
 # model settings
 model = dict(head=dict(num_classes=10))
 
 # optimizer
 optimizer = dict(type='SGD', lr=0.1, momentum=0.9, weight_decay=5e-4)
+optim_wrapper = dict(type='OptimWrapper', optimizer=optimizer)
 
 # learning rate scheduler
 param_scheduler = [
@@ -17,5 +22,6 @@ param_scheduler = [
 ]
 
 # runtime settings
-runner = dict(type='EpochBasedRunner', max_epochs=350)
-checkpoint_config = dict(interval=50)
+train_cfg = dict(type='EpochBasedTrainLoop', max_epochs=350)
+default_hooks = dict(
+    checkpoint=dict(type='CheckpointHook', interval=50, max_keep_ckpts=3))