[Docs] Fix some unconsistency (#51)

* del ceph setting in configs

* fix some unconsistency between algo's README and configs

configs/_base_/datasets/mmcls/imagenet_bs128_colorjittor.py

@@ -2,17 +2,8 @@
 dataset_type = 'ImageNet'
 img_norm_cfg = dict(
     mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
-file_client_args = dict(
-    backend='petrel',
-    path_mapping=dict({
-        'data/imagenet/':
-        'openmmlab:s3://openmmlab/datasets/classification/imagenet/',
-        'data/imagenet/':
-        'openmmlab:s3://openmmlab/datasets/classification/imagenet/'
-    }))
-
 train_pipeline = [
-    dict(type='LoadImageFromFile', file_client_args=file_client_args),
+    dict(type='LoadImageFromFile'),
     dict(type='RandomResizedCrop', size=224),
     dict(type='ColorJitter', brightness=0.4, contrast=0.4, saturation=0.4),
     dict(type='RandomFlip', flip_prob=0.5, direction='horizontal'),
@@ -22,7 +13,7 @@ train_pipeline = [
     dict(type='Collect', keys=['img', 'gt_label'])
 ]
 test_pipeline = [
-    dict(type='LoadImageFromFile', file_client_args=file_client_args),
+    dict(type='LoadImageFromFile'),
     dict(type='Resize', size=(256, -1)),
     dict(type='CenterCrop', crop_size=224),
     dict(type='Normalize', **img_norm_cfg),

configs/nas/detnas/README.md

@@ -32,15 +32,15 @@ Object detectors are usually equipped with backbone networks designed for image
 ## Getting Started
 ### Step 1: Supernet pre-training on ImageNet
 ```bash
-python ./tools/mmdet/train_mmdet.py \
-  configs/nas/detnas/detnas_shufflenet_supernet_imagenet.py \
+python ./tools/mmcls/train_mmcls.py \
+  configs/nas/detnas/detnas_supernet_shufflenetv2_8xb128_in1k.py \
   --work-dir $WORK_DIR
 ```
 
 ### Step 2: Supernet fine-tuning on COCO
 ```bash
 python ./tools/mmdet/train_mmdet.py \
-  configs/nas/detnas/detnas_frcnn_shufflenet_fpn_supernet_coco_1x.py \
+  configs/nas/detnas/detnas_supernet_frcnn_shufflenetv2_fpn_1x_coco.py \
   --work-dir $WORK_DIR \
   --cfg-options load_from=$STEP1_CKPT
 ```
@@ -48,15 +48,15 @@ python ./tools/mmdet/train_mmdet.py \
 ### Step 3: Search for subnet on the trained supernet
 ```
 python ./tools/mmdet/search_mmdet.py \
-  configs/nas/detnas/detnas_frcnn_shufflenet_fpn_evolution_search_coco.py \
+  configs/nas/detnas/detnas_evolution_search_frcnn_shufflenetv2_fpn_coco.py \
   $STEP2_CKPT \
   --work-dir $WORK_DIR
 ```
 
 ### Step 4: Subnet retraining on ImageNet
 ```
-python ./tools/mmdet/train_mmdet.py \
-  configs/nas/detnas/detnas_shufflenet_subnet_imagenet.py \
+python ./tools/mmcls/train_mmcls.py \
+  configs/nas/detnas/detnas_subnet_shufflenetv2_8xb128_in1k.py \
   --work-dir $WORK_DIR \
   --cfg-options algorithm.mutable_cfg=$STEP3_SUBNET_YAML
 ```
@@ -64,7 +64,7 @@ python ./tools/mmdet/train_mmdet.py \
 ### Step 5: Subnet fine-tuning on COCO
 ```
 python ./tools/mmdet/train_mmdet.py \
-  configs/nas/detnas/detnas_frcnn_shufflenet_fpn_subnet_coco_1x.py \
+  configs/nas/detnas/detnas_subnet_frcnn_shufflenetv2_fpn_1x_coco.py \
   --work-dir $WORK_DIR \
-  --cfg-options algorithm.mutable_cfg=$STEP3_SUBNET_YAML load_from=$STEP$_CKPT
+  --cfg-options algorithm.mutable_cfg=$STEP3_SUBNET_YAML load_from=$STEP4_CKPT
 ```

configs/nas/detnas/detnas_evolution_search_frcnn_shufflenetv2_fpn_coco.py

@@ -1,4 +1,4 @@
-_base_ = ['./detnas_frcnn_shufflenet_fpn_supernet_coco_1x.py']
+_base_ = ['./detnas_supernet_frcnn_shufflenetv2_fpn_1x_coco.py']
 
 data = dict(
     samples_per_gpu=128,

configs/nas/detnas/detnas_subnet_shufflenetv2_8xb128_in1k.py

@@ -1,3 +1,3 @@
 _base_ = [
-    '../spos/spos_shufflenetv2_subnet_8xb128_in1k.py',
+    '../spos/spos_subnet_shufflenetv2_8xb128_in1k.py',
 ]

configs/nas/detnas/detnas_supernet_shufflenetv2_8xb128_in1k.py

@@ -1,5 +1,5 @@
 _base_ = [
-    '../spos/spos_shufflenetv2_supernet_8xb128_in1k.py',
+    '../spos/spos_supernet_shufflenetv2_8xb128_in1k.py',
 ]
 
 runner = dict(max_iters=300000)

configs/nas/spos/README.md

@@ -24,7 +24,7 @@ Comprehensive experiments verify that our approach is flexible and effective. It
 ## Results and models
 |Dataset|       Supernet      | Subnet | Params(M) | Flops(G) | Top-1 (%) | Top-5 (%) | Config | Download | Remarks |
 |:---------------------:|:---------------------:|:------:|:---------:|:--------:|:---------:|:---------:|:------:|:---------|:---------:|
-|ImageNet|   ShuffleNetV2      |[mutable](https://openmmlab-share.oss-cn-hangzhou.aliyuncs.com/mmrazor/v0.1/nas/spos/spos_shufflenetv2_subnet_8xb128_in1k/spos_shufflenetv2_subnet_8xb128_in1k_flops_0.33M_acc_73.87_20211222-454627be_mutable_cfg.yaml?versionId=CAEQHxiBgICw5b6I7xciIGY5MjVmNWFhY2U5MjQzN2M4NDViYzI2YWRmYWE1YzQx)|    3.35    |  0.33        |     73.87  |    91.6   |[config](./spos_shufflenetv2_subnet_8xb128_in1k.py)|[model](https://openmmlab-share.oss-cn-hangzhou.aliyuncs.com/mmrazor/v0.1/nas/spos/spos_shufflenetv2_subnet_8xb128_in1k/spos_shufflenetv2_subnet_8xb128_in1k_flops_0.33M_acc_73.87_20211222-1f0a0b4d.pth?versionId=CAEQHxiBgIDK5b6I7xciIDM1YjIwZjQxN2UyMDRjYjA5YTM5NTBlMGNhMTdkNjI2) | [log](https://openmmlab-share.oss-cn-hangzhou.aliyuncs.com/mmrazor/v0.1/nas/spos/spos_shufflenetv2_subnet_8xb128_in1k/spos_shufflenetv2_subnet_8xb128_in1k_flops_0.33M_acc_73.87_20211222-1f0a0b4d.log.json?versionId=CAEQHxiBgIDr9cuL7xciIDBmOTZiZGUyYjRiMDQ5NzhhZjY0NWUxYmUzNDlmNTg5)| MMRazor searched
+|ImageNet|   ShuffleNetV2      |[mutable](https://openmmlab-share.oss-cn-hangzhou.aliyuncs.com/mmrazor/v0.1/nas/spos/spos_shufflenetv2_subnet_8xb128_in1k/spos_shufflenetv2_subnet_8xb128_in1k_flops_0.33M_acc_73.87_20211222-454627be_mutable_cfg.yaml?versionId=CAEQHxiBgICw5b6I7xciIGY5MjVmNWFhY2U5MjQzN2M4NDViYzI2YWRmYWE1YzQx)|    3.35    |  0.33        |     73.87  |    91.6   |[config](./spos_subnet_shufflenetv2_8xb128_in1k.py)|[model](https://openmmlab-share.oss-cn-hangzhou.aliyuncs.com/mmrazor/v0.1/nas/spos/spos_shufflenetv2_subnet_8xb128_in1k/spos_shufflenetv2_subnet_8xb128_in1k_flops_0.33M_acc_73.87_20211222-1f0a0b4d.pth?versionId=CAEQHxiBgIDK5b6I7xciIDM1YjIwZjQxN2UyMDRjYjA5YTM5NTBlMGNhMTdkNjI2) | [log](https://openmmlab-share.oss-cn-hangzhou.aliyuncs.com/mmrazor/v0.1/nas/spos/spos_shufflenetv2_subnet_8xb128_in1k/spos_shufflenetv2_subnet_8xb128_in1k_flops_0.33M_acc_73.87_20211222-1f0a0b4d.log.json?versionId=CAEQHxiBgIDr9cuL7xciIDBmOTZiZGUyYjRiMDQ5NzhhZjY0NWUxYmUzNDlmNTg5)| MMRazor searched
 
 **Note**:
 1. There are some small differences in our experiment in order to be consistent with other repos in OpenMMLab. For example,
@@ -34,13 +34,13 @@ normalize images in data preprocessing; resize by cv2 rather than PIL in trainin
 ### Supernet pre-training on ImageNet
 ```bash
 python ./tools/mmcls/train_mmcls.py \
-  configs/nas/spos/spos_shufflenet_supernet_imagenet.py \
+  configs/nas/spos/spos_supernet_shufflenetv2_8xb128_in1k.py \
   --work-dir $WORK_DIR
 ```
 ### Search for subnet on the trained supernet
 ```bash
 python ./tools/mmcls/search_mmcls.py \
-  configs/nas/spos/spos_shufflenet_evolution_search_imagenet.py \
+  configs/nas/spos/spos_evolution_search_shufflenetv2_8xb2048_in1k.py \
   $STEP1_CKPT \
   --work-dir $WORK_DIR
 ```
@@ -48,7 +48,7 @@ python ./tools/mmcls/search_mmcls.py \
 ### Subnet retraining on ImageNet
 ```bash
 python ./tools/mmcls/train_mmcls.py \
-  configs/nas/spos/spos_shufflenet_subnet_imagenet.py \
+  configs/nas/spos/spos_subnet_shufflenetv2_8xb128_in1k.py \
   --work-dir $WORK_DIR \
   --cfg-options algorithm.mutable_cfg=$STEP2_SUBNET_YAML
 ```

configs/nas/spos/spos_evolution_search_shufflenetv2_8xb2048_in1k.py

@@ -1,4 +1,4 @@
-_base_ = ['./spos_shufflenetv2_supernet_8xb128_in1k.py']
+_base_ = ['./spos_supernet_shufflenetv2_8xb128_in1k.py']
 
 data = dict(
     samples_per_gpu=2048,

configs/nas/spos/spos_subnet_shufflenetv2_8xb128_in1k.py

@@ -1,5 +1,5 @@
 _base_ = [
-    './spos_shufflenetv2_supernet_8xb128_in1k.py',
+    './spos_supernet_shufflenetv2_8xb128_in1k.py',
 ]
 
 algorithm = dict(retraining=True)