Merge branch 'develop' into android_demo_doc

MANIFEST.in

@@ -2,7 +2,7 @@ include LICENSE.txt
 include README.md
 include docs/en/whl_en.md
 recursive-include deploy/python *.py
+recursive-include deploy/utils *.py
+recursive-include ppcls/arch *.py
+recursive-include ppcls/utils *.py *.txt
 recursive-include deploy/configs *.yaml
-recursive-include deploy/utils get_image_list.py config.py logger.py predictor.py
-
-recursive-include ppcls/ *.py *.txt
\ No newline at end of file

__init__.py

@@ -14,4 +14,4 @@
 
 __all__ = ['PaddleClas']
 from .paddleclas import PaddleClas
-from ppcls.arch.backbone import *
+from .ppcls.arch.backbone import *

deploy/python/build_gallery.py

@@ -12,10 +12,6 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import os
-import sys
-
-__dir__ = os.path.dirname(os.path.abspath(__file__))
-sys.path.append(os.path.abspath(os.path.join(__dir__, '../')))
 
 import cv2
 import faiss
@@ -23,10 +19,9 @@ import numpy as np
 from tqdm import tqdm
 import pickle
 
-from python.predict_rec import RecPredictor
-
-from utils import logger
-from utils import config
+from paddleclas.deploy.utils import logger, config
+from paddleclas.deploy.python.predict_rec import RecPredictor
+from paddleclas.deploy.python.predict_rec import RecPredictor
 
 
 def split_datafile(data_file, image_root, delimiter="\t"):

deploy/python/predict_cls.py

@@ -11,21 +11,17 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
-import os
-import sys
 
-__dir__ = os.path.dirname(os.path.abspath(__file__))
-sys.path.append(os.path.abspath(os.path.join(__dir__, '../')))
+import os
 
 import cv2
 import numpy as np
 
-from utils import logger
-from utils import config
-from utils.predictor import Predictor
-from utils.get_image_list import get_image_list
-from python.preprocess import create_operators
-from python.postprocess import build_postprocess
+from paddleclas.deploy.utils import logger, config
+from paddleclas.deploy.utils.predictor import Predictor
+from paddleclas.deploy.utils.get_image_list import get_image_list
+from paddleclas.deploy.python.preprocess import create_operators
+from paddleclas.deploy.python.postprocess import build_postprocess
 
 
 class ClsPredictor(Predictor):

deploy/python/predict_det.py

@@ -11,29 +11,23 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
-import os
-import sys
-
-__dir__ = os.path.dirname(os.path.abspath(__file__))
-sys.path.append(os.path.abspath(os.path.join(__dir__, '../')))
-
-from utils import logger
-from utils import config
-from utils.predictor import Predictor
-from utils.get_image_list import get_image_list
-from det_preprocess import det_preprocess
-from preprocess import create_operators
-
 import os
 import argparse
 import time
+from functools import reduce
+
 import yaml
 import ast
-from functools import reduce
-import cv2
 import numpy as np
+import cv2
 import paddle
 
+from paddleclas.deploy.utils import logger, config
+from paddleclas.deploy.utils.predictor import Predictor
+from paddleclas.deploy.utils.get_image_list import get_image_list
+from paddleclas.deploy.python.preprocess import create_operators
+from paddleclas.deploy.python.det_preprocess import det_preprocess
+
 
 class DetPredictor(Predictor):
     def __init__(self, config):

deploy/python/predict_rec.py

@@ -12,20 +12,15 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import os
-import sys
-
-__dir__ = os.path.dirname(os.path.abspath(__file__))
-sys.path.append(os.path.abspath(os.path.join(__dir__, '../')))
 
 import cv2
 import numpy as np
 
-from utils import logger
-from utils import config
-from utils.predictor import Predictor
-from utils.get_image_list import get_image_list
-from preprocess import create_operators
-from postprocess import build_postprocess
+from paddleclas.deploy.utils import logger, config
+from paddleclas.deploy.utils.predictor import Predictor
+from paddleclas.deploy.utils.get_image_list import get_image_list
+from paddleclas.deploy.python.preprocess import create_operators
+from paddleclas.deploy.python.postprocess import build_postprocess
 
 
 class RecPredictor(Predictor):

deploy/python/predict_system.py

@@ -12,24 +12,18 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import os
-import sys
-
-__dir__ = os.path.dirname(os.path.abspath(__file__))
-sys.path.append(os.path.abspath(os.path.join(__dir__, '../')))
-
 import copy
-import cv2
+
 import numpy as np
+import cv2
 import faiss
 import pickle
 
-from python.predict_rec import RecPredictor
-from python.predict_det import DetPredictor
-
-from utils import logger
-from utils import config
-from utils.get_image_list import get_image_list
-from utils.draw_bbox import draw_bbox_results
+from paddleclas.deploy.utils import logger, config
+from paddleclas.deploy.utils.get_image_list import get_image_list
+from paddleclas.deploy.utils.draw_bbox import draw_bbox_results
+from paddleclas.deploy.python.predict_rec import RecPredictor
+from paddleclas.deploy.python.predict_det import DetPredictor
 
 
 class SystemPredictor(object):

deploy/python/preprocess.py

@@ -29,7 +29,7 @@ import importlib
 from PIL import Image
 from paddle.vision.transforms import ToTensor, Normalize
 
-from python.det_preprocess import DetNormalizeImage, DetPadStride, DetPermute, DetResize
+from paddleclas.deploy.python.det_preprocess import DetNormalizeImage, DetPadStride, DetPermute, DetResize
 
 
 def create_operators(params):

deploy/utils/config.py

@@ -17,7 +17,7 @@ import copy
 import argparse
 import yaml
 
-from utils import logger
+from . import logger
 
 __all__ = ['get_config']
 

docs/en/algorithm_introduction/ImageNet_models_en.md

@@ -27,7 +27,7 @@
 - [20. DLA series](#20)
 - [21. RedNet series](#21)
 - [22. TNT series](#22)
-- [23. CSwinTransformer series](#23)
+- [23. CSWinTransformer series](#23)
 - [24. PVTV2 series](#24)
 - [25. MobileViT series](#25)
 - [26. Other models](#26)
@@ -400,14 +400,14 @@ The accuracy and speed indicators of SwinTransformer series models are shown in
 
 | Model       | Top-1 Acc | Top-5 Acc | time(ms)<br>bs=1 | time(ms)<br>bs=4 | time(ms)<br/>bs=8 | FLOPs(G) | Params(M) | Pretrained Model Download Address                                               | Inference Model Download Address                                      |
 | ---------- | --------- | --------- | ---------------- | ---------------- | -------- | --------- | ------------------------------------------------------------ | ------------------------------------------------------------ | ------------------------------------------------------------ |
-| SwinTransformer_tiny_patch4_window7_224    | 0.8069 | 0.9534 | 6.59 | 9.68 | 16.32 | 4.35  | 28.26   | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_tiny_patch4_window7_224_pretrained.pdparams) | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_tiny_patch4_window7_224_infer.tar) |
-| SwinTransformer_small_patch4_window7_224   | 0.8275 | 0.9613 | 12.54 | 17.07 | 28.08 | 8.51  | 49.56   | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_small_patch4_window7_224_pretrained.pdparams) | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_small_patch4_window7_224_infer.tar) |
-| SwinTransformer_base_patch4_window7_224    | 0.8300 | 0.9626 | 13.37 | 23.53 | 39.11 | 15.13 | 87.70   | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_base_patch4_window7_224_pretrained.pdparams) | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_base_patch4_window7_224_infer.tar) |
-| SwinTransformer_base_patch4_window12_384   | 0.8439 | 0.9693 | 19.52 | 64.56 | 123.30 | 44.45 | 87.70   | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_base_patch4_window12_384_pretrained.pdparams) | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_base_patch4_window12_384_infer.tar) |
-| SwinTransformer_base_patch4_window7_224<sup>[1]</sup>     | 0.8487 | 0.9746 | 13.53 | 23.46 | 39.13 | 15.13 | 87.70   | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_base_patch4_window7_224_22kto1k_pretrained.pdparams) | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_base_patch4_window7_224_infer.tar) |
-| SwinTransformer_base_patch4_window12_384<sup>[1]</sup>    | 0.8642 | 0.9807 | 19.65 | 64.72 | 123.42 | 44.45 | 87.70   | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_base_patch4_window12_384_22kto1k_pretrained.pdparams) | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_base_patch4_window12_384_infer.tar) |
-| SwinTransformer_large_patch4_window7_224<sup>[1]</sup>    | 0.8596 | 0.9783 | 15.74 | 38.57 | 71.49 | 34.02 | 196.43  | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_large_patch4_window7_224_22kto1k_pretrained.pdparams) | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_large_patch4_window7_224_22kto1k_infer.tar) |
-| SwinTransformer_large_patch4_window12_384<sup>[1]</sup>   | 0.8719 | 0.9823 | 32.61 | 116.59 | 223.23 | 99.97 | 196.43 | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_large_patch4_window12_384_22kto1k_pretrained.pdparams) | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_large_patch4_window12_384_22kto1k_infer.tar) |
+| SwinTransformer_tiny_patch4_window7_224    | 0.8069 | 0.9534 | 6.59 | 9.68 | 16.32 | 4.35  | 28.26   | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_tiny_patch4_window7_224_pretrained.pdparams) | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_tiny_patch4_window7_224_infer.tar) |
+| SwinTransformer_small_patch4_window7_224   | 0.8275 | 0.9613 | 12.54 | 17.07 | 28.08 | 8.51  | 49.56   | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_small_patch4_window7_224_pretrained.pdparams) | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_small_patch4_window7_224_infer.tar) |
+| SwinTransformer_base_patch4_window7_224    | 0.8300 | 0.9626 | 13.37 | 23.53 | 39.11 | 15.13 | 87.70   | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_base_patch4_window7_224_pretrained.pdparams) | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_base_patch4_window7_224_infer.tar) |
+| SwinTransformer_base_patch4_window12_384   | 0.8439 | 0.9693 | 19.52 | 64.56 | 123.30 | 44.45 | 87.70   | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_base_patch4_window12_384_pretrained.pdparams) | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_base_patch4_window12_384_infer.tar) |
+| SwinTransformer_base_patch4_window7_224<sup>[1]</sup>     | 0.8487 | 0.9746 | 13.53 | 23.46 | 39.13 | 15.13 | 87.70   | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_base_patch4_window7_224_22kto1k_pretrained.pdparams) | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_base_patch4_window7_224_infer.tar) |
+| SwinTransformer_base_patch4_window12_384<sup>[1]</sup>    | 0.8642 | 0.9807 | 19.65 | 64.72 | 123.42 | 44.45 | 87.70   | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_base_patch4_window12_384_22kto1k_pretrained.pdparams) | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_base_patch4_window12_384_infer.tar) |
+| SwinTransformer_large_patch4_window7_224<sup>[1]</sup>    | 0.8596 | 0.9783 | 15.74 | 38.57 | 71.49 | 34.02 | 196.43  | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_large_patch4_window7_224_22kto1k_pretrained.pdparams) | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_large_patch4_window7_224_22kto1k_infer.tar) |
+| SwinTransformer_large_patch4_window12_384<sup>[1]</sup>   | 0.8719 | 0.9823 | 32.61 | 116.59 | 223.23 | 99.97 | 196.43 | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_large_patch4_window12_384_22kto1k_pretrained.pdparams) | [Download link](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_large_patch4_window12_384_22kto1k_infer.tar) |
 
 [1]：It is pre-trained based on the ImageNet22k dataset, and then transferred and learned from the ImageNet1k dataset.
 

docs/en/installation/install_paddleclas_en.md

@@ -23,7 +23,13 @@ git clone https://gitee.com/paddlepaddle/PaddleClas.git -b develop
 
 <a name='2'></a>
 
-## 2. Install requirements
+## 2. Install PaddleClas and requirements
+
+It is recommanded that installing from PyPI:
+
+```shell
+pip install paddleclas
+```
 
 PaddleClas dependencies are listed in file `requirements.txt`, you can use the following command to install the dependencies.
 

docs/zh_CN/algorithm_introduction/ImageNet_models.md

@@ -31,7 +31,7 @@
 - [DLA 系列](#DLA)
 - [RedNet 系列](#RedNet)
 - [TNT 系列](#TNT)
-- [CSwinTransformer 系列](#CSwinTransformer)
+- [CSWinTransformer 系列](#CSWinTransformer)
 - [PVTV2 系列](#PVTV2)
 - [MobileViT 系列](#MobileViT)
 - [其他模型](#Others)
@@ -426,14 +426,14 @@ ViT(Vision Transformer) 与 DeiT（Data-efficient Image Transformers）系列模
 
 | 模型       | Top-1 Acc | Top-5 Acc | time(ms)<br>bs=1 | time(ms)<br>bs=4 | time(ms)<br/>bs=8 | FLOPs(G) | Params(M) | 预训练模型下载地址                                               | inference模型下载地址                                      |
 | ---------- | --------- | --------- | ---------------- | ---------------- | -------- | --------- | ------------------------------------------------------------ | ------------------------------------------------------------ | ------------------------------------------------------------ |
-| SwinTransformer_tiny_patch4_window7_224    | 0.8069 | 0.9534 | 6.59 | 9.68 | 16.32 | 4.35  | 28.26   | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_tiny_patch4_window7_224_pretrained.pdparams) | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_tiny_patch4_window7_224_infer.tar) |
-| SwinTransformer_small_patch4_window7_224   | 0.8275 | 0.9613 | 12.54 | 17.07 | 28.08 | 8.51  | 49.56   | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_small_patch4_window7_224_pretrained.pdparams) | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_small_patch4_window7_224_infer.tar) |
-| SwinTransformer_base_patch4_window7_224    | 0.8300 | 0.9626 | 13.37 | 23.53 | 39.11 | 15.13 | 87.70   | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_base_patch4_window7_224_pretrained.pdparams) | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_base_patch4_window7_224_infer.tar) |
-| SwinTransformer_base_patch4_window12_384   | 0.8439 | 0.9693 | 19.52 | 64.56 | 123.30 | 44.45 | 87.70   | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_base_patch4_window12_384_pretrained.pdparams) | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_base_patch4_window12_384_infer.tar) |
-| SwinTransformer_base_patch4_window7_224<sup>[1]</sup>     | 0.8487 | 0.9746 | 13.53 | 23.46 | 39.13 | 15.13 | 87.70   | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_base_patch4_window7_224_22kto1k_pretrained.pdparams) | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_base_patch4_window7_224_infer.tar) |
-| SwinTransformer_base_patch4_window12_384<sup>[1]</sup>    | 0.8642 | 0.9807 | 19.65 | 64.72 | 123.42 | 44.45 | 87.70   | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_base_patch4_window12_384_22kto1k_pretrained.pdparams) | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_base_patch4_window12_384_infer.tar) |
-| SwinTransformer_large_patch4_window7_224<sup>[1]</sup>    | 0.8596 | 0.9783 | 15.74 | 38.57 | 71.49 | 34.02 | 196.43  | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_large_patch4_window7_224_22kto1k_pretrained.pdparams) | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_large_patch4_window7_224_22kto1k_infer.tar) |
-| SwinTransformer_large_patch4_window12_384<sup>[1]</sup>   | 0.8719 | 0.9823 | 32.61 | 116.59 | 223.23 | 99.97 | 196.43 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_large_patch4_window12_384_22kto1k_pretrained.pdparams) | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_large_patch4_window12_384_22kto1k_infer.tar) |
+| SwinTransformer_tiny_patch4_window7_224    | 0.8069 | 0.9534 | 6.59 | 9.68 | 16.32 | 4.35  | 28.26   | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_tiny_patch4_window7_224_pretrained.pdparams) | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_tiny_patch4_window7_224_infer.tar) |
+| SwinTransformer_small_patch4_window7_224   | 0.8275 | 0.9613 | 12.54 | 17.07 | 28.08 | 8.51  | 49.56   | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_small_patch4_window7_224_pretrained.pdparams) | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_small_patch4_window7_224_infer.tar) |
+| SwinTransformer_base_patch4_window7_224    | 0.8300 | 0.9626 | 13.37 | 23.53 | 39.11 | 15.13 | 87.70   | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_base_patch4_window7_224_pretrained.pdparams) | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_base_patch4_window7_224_infer.tar) |
+| SwinTransformer_base_patch4_window12_384   | 0.8439 | 0.9693 | 19.52 | 64.56 | 123.30 | 44.45 | 87.70   | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_base_patch4_window12_384_pretrained.pdparams) | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_base_patch4_window12_384_infer.tar) |
+| SwinTransformer_base_patch4_window7_224<sup>[1]</sup>     | 0.8487 | 0.9746 | 13.53 | 23.46 | 39.13 | 15.13 | 87.70   | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_base_patch4_window7_224_22kto1k_pretrained.pdparams) | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_base_patch4_window7_224_infer.tar) |
+| SwinTransformer_base_patch4_window12_384<sup>[1]</sup>    | 0.8642 | 0.9807 | 19.65 | 64.72 | 123.42 | 44.45 | 87.70   | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_base_patch4_window12_384_22kto1k_pretrained.pdparams) | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_base_patch4_window12_384_infer.tar) |
+| SwinTransformer_large_patch4_window7_224<sup>[1]</sup>    | 0.8596 | 0.9783 | 15.74 | 38.57 | 71.49 | 34.02 | 196.43  | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_large_patch4_window7_224_22kto1k_pretrained.pdparams) | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_large_patch4_window7_224_22kto1k_infer.tar) |
+| SwinTransformer_large_patch4_window12_384<sup>[1]</sup>   | 0.8719 | 0.9823 | 32.61 | 116.59 | 223.23 | 99.97 | 196.43 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_large_patch4_window12_384_22kto1k_pretrained.pdparams) | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/SwinTransformer_large_patch4_window12_384_22kto1k_infer.tar) |
 
 [1]：基于 ImageNet22k 数据集预训练，然后在 ImageNet1k 数据集迁移学习得到。
 

docs/zh_CN/inference_deployment/whl_deploy.md

@@ -92,7 +92,6 @@ Predict complete!
 
 * 命令行中
 ```bash
-from paddleclas import PaddleClas, get_default_confg
 paddleclas --model_name=ViT_base_patch16_384 --infer_imgs='docs/images/inference_deployment/whl_demo.jpg' --resize_short=384 --crop_size=384
 ```
 

docs/zh_CN/installation/install_paddleclas.md

@@ -26,7 +26,7 @@
 
 #### 1.1.1 使用Paddle官方镜像
 
-* 切换到工作目录下，例如工作目录为`/home/Projects`，则运行命令: 
+* 切换到工作目录下，例如工作目录为`/home/Projects`，则运行命令:
 
 ```shell
 cd /home/Projects
@@ -96,7 +96,13 @@ git clone https://gitee.com/paddlepaddle/PaddleClas.git -b release/2.4
 ```
 <a name='1.3'></a>
 
-### 1.3 安装 Python 依赖库
+### 1.3 安装 PaddleClas 及其 Python 依赖库
+
+建议直接从 PyPI 安装 PaddleClas：
+
+```shell
+pip install paddleclas
+```
 
 PaddleClas 的 Python 依赖库在 `requirements.txt` 中给出，可通过如下命令安装：
 
@@ -114,5 +120,3 @@ pip install --upgrade -r requirements.txt -i https://mirror.baidu.com/pypi/simpl
 如果您对自动化制作docker镜像感兴趣，或有自定义需求，请访问[PaddlePaddle/PaddleCloud](https://github.com/PaddlePaddle/PaddleCloud/tree/main/tekton)做进一步了解。
 
 **备注**：当前的镜像中的 PaddleClas 代码默认使用最新的 release/2.4 分支。
-
-

paddleclas.py

@@ -13,11 +13,6 @@
 # limitations under the License.
 
 import os
-import sys
-__dir__ = os.path.dirname(__file__)
-sys.path.append(os.path.join(__dir__, ""))
-sys.path.append(os.path.join(__dir__, "deploy"))
-
 from typing import Union, Generator
 import argparse
 import shutil
@@ -33,12 +28,16 @@ from tqdm import tqdm
 from prettytable import PrettyTable
 import paddle
 
-from deploy.python.predict_cls import ClsPredictor
-from deploy.utils.get_image_list import get_image_list
-from deploy.utils import config
+from .ppcls.arch import backbone
+from .ppcls.utils import logger
+
+from .deploy.python.predict_cls import ClsPredictor
+from .deploy.utils.get_image_list import get_image_list
+from .deploy.utils import config
 
-import ppcls.arch.backbone as backbone
-from ppcls.utils import logger
+# for the PaddleClas Project
+from . import deploy
+from . import ppcls
 
 # for building model with loading pretrained weights from backbone
 logger.init_logger()
@@ -51,6 +50,11 @@ BASE_IMAGES_DIR = os.path.join(BASE_DIR, "images")
 IMN_MODEL_BASE_DOWNLOAD_URL = "https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/{}_infer.tar"
 IMN_MODEL_SERIES = {
     "AlexNet": ["AlexNet"],
+    "CSWinTransformer": [
+        "CSWinTransformer_tiny_224", "CSWinTransformer_small_224",
+        "CSWinTransformer_base_224", "CSWinTransformer_base_384",
+        "CSWinTransformer_large_224", "CSWinTransformer_large_384"
+    ],
     "DarkNet": ["DarkNet53"],
     "DeiT": [
         "DeiT_base_distilled_patch16_224", "DeiT_base_distilled_patch16_384",
@@ -82,6 +86,8 @@ IMN_MODEL_SERIES = {
         "HRNet_W48_C_ssld"
     ],
     "Inception": ["GoogLeNet", "InceptionV3", "InceptionV4"],
+    "LeViT":
+    ["LeViT_128S", "LeViT_128", "LeViT_192", "LeViT_256", "LeViT_384"],
     "MixNet": ["MixNet_S", "MixNet_M", "MixNet_L"],
     "MobileNetV1": [
         "MobileNetV1_x0_25", "MobileNetV1_x0_5", "MobileNetV1_x0_75",
@@ -100,6 +106,7 @@ IMN_MODEL_SERIES = {
         "MobileNetV3_large_x1_0", "MobileNetV3_large_x1_25",
         "MobileNetV3_small_x1_0_ssld", "MobileNetV3_large_x1_0_ssld"
     ],
+    "MobileViT": ["MobileViT_XXS", "MobileViT_XS", "MobileViT_S"],
     "PPHGNet": [
         "PPHGNet_tiny",
         "PPHGNet_small",
@@ -111,6 +118,10 @@ IMN_MODEL_SERIES = {
         "PPLCNet_x1_0", "PPLCNet_x1_5", "PPLCNet_x2_0", "PPLCNet_x2_5"
     ],
     "PPLCNetV2": ["PPLCNetV2_base"],
+    "PVTV2": [
+        "PVT_V2_B0", "PVT_V2_B1", "PVT_V2_B2", "PVT_V2_B2_Linear", "PVT_V2_B3",
+        "PVT_V2_B4", "PVT_V2_B5"
+    ],
     "RedNet": ["RedNet26", "RedNet38", "RedNet50", "RedNet101", "RedNet152"],
     "RegNet": ["RegNetX_4GF"],
     "Res2Net": [
@@ -163,6 +174,7 @@ IMN_MODEL_SERIES = {
         "pcpvt_small", "pcpvt_base", "pcpvt_large", "alt_gvt_small",
         "alt_gvt_base", "alt_gvt_large"
     ],
+    "TNT": ["TNT_small"],
     "VGG": ["VGG11", "VGG13", "VGG16", "VGG19"],
     "VisionTransformer": [
         "ViT_base_patch16_224", "ViT_base_patch16_384", "ViT_base_patch32_384",
@@ -202,6 +214,7 @@ class InputModelError(Exception):
 def init_config(model_type, model_name, inference_model_dir, **kwargs):
 
     cfg_path = f"deploy/configs/PULC/{model_name}/inference_{model_name}.yaml" if model_type == "pulc" else "deploy/configs/inference_cls.yaml"
+    __dir__ = os.path.dirname(__file__)
     cfg_path = os.path.join(__dir__, cfg_path)
     cfg = config.get_config(cfg_path, show=False)
 
@@ -453,10 +466,6 @@ class PaddleClas(object):
     """PaddleClas.
     """
 
-    if not os.environ.get('ppcls', False):
-        os.environ.setdefault('ppcls', 'True')
-        print_info()
-
     def __init__(self,
                  model_name: str=None,
                  inference_model_dir: str=None,
@@ -471,6 +480,7 @@ class PaddleClas(object):
             topk (int, optional): Return the top k prediction results with the highest score. Defaults to 5.
         """
         super().__init__()
+
         self.model_type, inference_model_dir = self._check_input_model(
             model_name, inference_model_dir)
         self._config = init_config(self.model_type, model_name,
@@ -595,6 +605,7 @@ class PaddleClas(object):
 def main():
     """Function API used for commad line.
     """
+    print_info()
     cfg = args_cfg()
     clas_engine = PaddleClas(**cfg)
     res = clas_engine.predict(cfg["infer_imgs"], print_pred=True)

ppcls/arch/__init__.py

@@ -23,11 +23,11 @@ from . import backbone, gears
 from .backbone import *
 from .gears import build_gear
 from .utils import *
-from ppcls.arch.backbone.base.theseus_layer import TheseusLayer
-from ppcls.utils import logger
-from ppcls.utils.save_load import load_dygraph_pretrain
-from ppcls.arch.slim import prune_model, quantize_model
-from ppcls.arch.distill.afd_attention import LinearTransformStudent, LinearTransformTeacher
+from .backbone.base.theseus_layer import TheseusLayer
+from ..utils import logger
+from ..utils.save_load import load_dygraph_pretrain
+from .slim import prune_model, quantize_model
+from .distill.afd_attention import LinearTransformStudent, LinearTransformTeacher
 
 __all__ = ["build_model", "RecModel", "DistillationModel", "AttentionModel"]
 

ppcls/arch/backbone/__init__.py

@@ -15,65 +15,65 @@
 import sys
 import inspect
 
-from ppcls.arch.backbone.legendary_models.mobilenet_v1 import MobileNetV1_x0_25, MobileNetV1_x0_5, MobileNetV1_x0_75, MobileNetV1
-from ppcls.arch.backbone.legendary_models.mobilenet_v3 import MobileNetV3_small_x0_35, MobileNetV3_small_x0_5, MobileNetV3_small_x0_75, MobileNetV3_small_x1_0, MobileNetV3_small_x1_25, MobileNetV3_large_x0_35, MobileNetV3_large_x0_5, MobileNetV3_large_x0_75, MobileNetV3_large_x1_0, MobileNetV3_large_x1_25
-from ppcls.arch.backbone.legendary_models.resnet import ResNet18, ResNet18_vd, ResNet34, ResNet34_vd, ResNet50, ResNet50_vd, ResNet101, ResNet101_vd, ResNet152, ResNet152_vd, ResNet200_vd
-from ppcls.arch.backbone.legendary_models.vgg import VGG11, VGG13, VGG16, VGG19
-from ppcls.arch.backbone.legendary_models.inception_v3 import InceptionV3
-from ppcls.arch.backbone.legendary_models.hrnet import HRNet_W18_C, HRNet_W30_C, HRNet_W32_C, HRNet_W40_C, HRNet_W44_C, HRNet_W48_C, HRNet_W60_C, HRNet_W64_C, SE_HRNet_W64_C
-from ppcls.arch.backbone.legendary_models.pp_lcnet import PPLCNet_x0_25, PPLCNet_x0_35, PPLCNet_x0_5, PPLCNet_x0_75, PPLCNet_x1_0, PPLCNet_x1_5, PPLCNet_x2_0, PPLCNet_x2_5
-from ppcls.arch.backbone.legendary_models.pp_lcnet_v2 import PPLCNetV2_base
-from ppcls.arch.backbone.legendary_models.esnet import ESNet_x0_25, ESNet_x0_5, ESNet_x0_75, ESNet_x1_0
-from ppcls.arch.backbone.legendary_models.pp_hgnet import PPHGNet_tiny, PPHGNet_small, PPHGNet_base
+from .legendary_models.mobilenet_v1 import MobileNetV1_x0_25, MobileNetV1_x0_5, MobileNetV1_x0_75, MobileNetV1
+from .legendary_models.mobilenet_v3 import MobileNetV3_small_x0_35, MobileNetV3_small_x0_5, MobileNetV3_small_x0_75, MobileNetV3_small_x1_0, MobileNetV3_small_x1_25, MobileNetV3_large_x0_35, MobileNetV3_large_x0_5, MobileNetV3_large_x0_75, MobileNetV3_large_x1_0, MobileNetV3_large_x1_25
+from .legendary_models.resnet import ResNet18, ResNet18_vd, ResNet34, ResNet34_vd, ResNet50, ResNet50_vd, ResNet101, ResNet101_vd, ResNet152, ResNet152_vd, ResNet200_vd
+from .legendary_models.vgg import VGG11, VGG13, VGG16, VGG19
+from .legendary_models.inception_v3 import InceptionV3
+from .legendary_models.hrnet import HRNet_W18_C, HRNet_W30_C, HRNet_W32_C, HRNet_W40_C, HRNet_W44_C, HRNet_W48_C, HRNet_W60_C, HRNet_W64_C, SE_HRNet_W64_C
+from .legendary_models.pp_lcnet import PPLCNet_x0_25, PPLCNet_x0_35, PPLCNet_x0_5, PPLCNet_x0_75, PPLCNet_x1_0, PPLCNet_x1_5, PPLCNet_x2_0, PPLCNet_x2_5
+from .legendary_models.pp_lcnet_v2 import PPLCNetV2_base
+from .legendary_models.esnet import ESNet_x0_25, ESNet_x0_5, ESNet_x0_75, ESNet_x1_0
+from .legendary_models.pp_hgnet import PPHGNet_tiny, PPHGNet_small, PPHGNet_base
 
-from ppcls.arch.backbone.model_zoo.resnet_vc import ResNet50_vc
-from ppcls.arch.backbone.model_zoo.resnext import ResNeXt50_32x4d, ResNeXt50_64x4d, ResNeXt101_32x4d, ResNeXt101_64x4d, ResNeXt152_32x4d, ResNeXt152_64x4d
-from ppcls.arch.backbone.model_zoo.resnext_vd import ResNeXt50_vd_32x4d, ResNeXt50_vd_64x4d, ResNeXt101_vd_32x4d, ResNeXt101_vd_64x4d, ResNeXt152_vd_32x4d, ResNeXt152_vd_64x4d
-from ppcls.arch.backbone.model_zoo.res2net import Res2Net50_26w_4s, Res2Net50_14w_8s
-from ppcls.arch.backbone.model_zoo.res2net_vd import Res2Net50_vd_26w_4s, Res2Net101_vd_26w_4s, Res2Net200_vd_26w_4s
-from ppcls.arch.backbone.model_zoo.se_resnet_vd import SE_ResNet18_vd, SE_ResNet34_vd, SE_ResNet50_vd
-from ppcls.arch.backbone.model_zoo.se_resnext_vd import SE_ResNeXt50_vd_32x4d, SE_ResNeXt50_vd_32x4d, SENet154_vd
-from ppcls.arch.backbone.model_zoo.se_resnext import SE_ResNeXt50_32x4d, SE_ResNeXt101_32x4d, SE_ResNeXt152_64x4d
-from ppcls.arch.backbone.model_zoo.dpn import DPN68, DPN92, DPN98, DPN107, DPN131
-from ppcls.arch.backbone.model_zoo.densenet import DenseNet121, DenseNet161, DenseNet169, DenseNet201, DenseNet264
-from ppcls.arch.backbone.model_zoo.efficientnet import EfficientNetB0, EfficientNetB1, EfficientNetB2, EfficientNetB3, EfficientNetB4, EfficientNetB5, EfficientNetB6, EfficientNetB7, EfficientNetB0_small
-from ppcls.arch.backbone.model_zoo.resnest import ResNeSt50_fast_1s1x64d, ResNeSt50, ResNeSt101
-from ppcls.arch.backbone.model_zoo.googlenet import GoogLeNet
-from ppcls.arch.backbone.model_zoo.mobilenet_v2 import MobileNetV2_x0_25, MobileNetV2_x0_5, MobileNetV2_x0_75, MobileNetV2, MobileNetV2_x1_5, MobileNetV2_x2_0
-from ppcls.arch.backbone.model_zoo.shufflenet_v2 import ShuffleNetV2_x0_25, ShuffleNetV2_x0_33, ShuffleNetV2_x0_5, ShuffleNetV2_x1_0, ShuffleNetV2_x1_5, ShuffleNetV2_x2_0, ShuffleNetV2_swish
-from ppcls.arch.backbone.model_zoo.ghostnet import GhostNet_x0_5, GhostNet_x1_0, GhostNet_x1_3
-from ppcls.arch.backbone.model_zoo.alexnet import AlexNet
-from ppcls.arch.backbone.model_zoo.inception_v4 import InceptionV4
-from ppcls.arch.backbone.model_zoo.xception import Xception41, Xception65, Xception71
-from ppcls.arch.backbone.model_zoo.xception_deeplab import Xception41_deeplab, Xception65_deeplab
-from ppcls.arch.backbone.model_zoo.resnext101_wsl import ResNeXt101_32x8d_wsl, ResNeXt101_32x16d_wsl, ResNeXt101_32x32d_wsl, ResNeXt101_32x48d_wsl
-from ppcls.arch.backbone.model_zoo.squeezenet import SqueezeNet1_0, SqueezeNet1_1
-from ppcls.arch.backbone.model_zoo.darknet import DarkNet53
-from ppcls.arch.backbone.model_zoo.regnet import RegNetX_200MF, RegNetX_4GF, RegNetX_32GF, RegNetY_200MF, RegNetY_4GF, RegNetY_32GF
-from ppcls.arch.backbone.model_zoo.vision_transformer import ViT_small_patch16_224, ViT_base_patch16_224, ViT_base_patch16_384, ViT_base_patch32_384, ViT_large_patch16_224, ViT_large_patch16_384, ViT_large_patch32_384
-from ppcls.arch.backbone.model_zoo.distilled_vision_transformer import DeiT_tiny_patch16_224, DeiT_small_patch16_224, DeiT_base_patch16_224, DeiT_tiny_distilled_patch16_224, DeiT_small_distilled_patch16_224, DeiT_base_distilled_patch16_224, DeiT_base_patch16_384, DeiT_base_distilled_patch16_384
-from ppcls.arch.backbone.legendary_models.swin_transformer import SwinTransformer_tiny_patch4_window7_224, SwinTransformer_small_patch4_window7_224, SwinTransformer_base_patch4_window7_224, SwinTransformer_base_patch4_window12_384, SwinTransformer_large_patch4_window7_224, SwinTransformer_large_patch4_window12_384
-from ppcls.arch.backbone.model_zoo.cswin_transformer import CSWinTransformer_tiny_224, CSWinTransformer_small_224, CSWinTransformer_base_224, CSWinTransformer_large_224, CSWinTransformer_base_384, CSWinTransformer_large_384
-from ppcls.arch.backbone.model_zoo.mixnet import MixNet_S, MixNet_M, MixNet_L
-from ppcls.arch.backbone.model_zoo.rexnet import ReXNet_1_0, ReXNet_1_3, ReXNet_1_5, ReXNet_2_0, ReXNet_3_0
-from ppcls.arch.backbone.model_zoo.gvt import pcpvt_small, pcpvt_base, pcpvt_large, alt_gvt_small, alt_gvt_base, alt_gvt_large
-from ppcls.arch.backbone.model_zoo.levit import LeViT_128S, LeViT_128, LeViT_192, LeViT_256, LeViT_384
-from ppcls.arch.backbone.model_zoo.dla import DLA34, DLA46_c, DLA46x_c, DLA60, DLA60x, DLA60x_c, DLA102, DLA102x, DLA102x2, DLA169
-from ppcls.arch.backbone.model_zoo.rednet import RedNet26, RedNet38, RedNet50, RedNet101, RedNet152
-from ppcls.arch.backbone.model_zoo.tnt import TNT_small
-from ppcls.arch.backbone.model_zoo.hardnet import HarDNet68, HarDNet85, HarDNet39_ds, HarDNet68_ds
-from ppcls.arch.backbone.model_zoo.cspnet import CSPDarkNet53
-from ppcls.arch.backbone.model_zoo.pvt_v2 import PVT_V2_B0, PVT_V2_B1, PVT_V2_B2_Linear, PVT_V2_B2, PVT_V2_B3, PVT_V2_B4, PVT_V2_B5
-from ppcls.arch.backbone.model_zoo.mobilevit import MobileViT_XXS, MobileViT_XS, MobileViT_S
-from ppcls.arch.backbone.model_zoo.repvgg import RepVGG_A0, RepVGG_A1, RepVGG_A2, RepVGG_B0, RepVGG_B1, RepVGG_B2, RepVGG_B1g2, RepVGG_B1g4, RepVGG_B2g4, RepVGG_B3g4
-from ppcls.arch.backbone.model_zoo.van import VAN_tiny
-from ppcls.arch.backbone.model_zoo.peleenet import PeleeNet
-from ppcls.arch.backbone.model_zoo.convnext import ConvNeXt_tiny
+from .model_zoo.resnet_vc import ResNet50_vc
+from .model_zoo.resnext import ResNeXt50_32x4d, ResNeXt50_64x4d, ResNeXt101_32x4d, ResNeXt101_64x4d, ResNeXt152_32x4d, ResNeXt152_64x4d
+from .model_zoo.resnext_vd import ResNeXt50_vd_32x4d, ResNeXt50_vd_64x4d, ResNeXt101_vd_32x4d, ResNeXt101_vd_64x4d, ResNeXt152_vd_32x4d, ResNeXt152_vd_64x4d
+from .model_zoo.res2net import Res2Net50_26w_4s, Res2Net50_14w_8s
+from .model_zoo.res2net_vd import Res2Net50_vd_26w_4s, Res2Net101_vd_26w_4s, Res2Net200_vd_26w_4s
+from .model_zoo.se_resnet_vd import SE_ResNet18_vd, SE_ResNet34_vd, SE_ResNet50_vd
+from .model_zoo.se_resnext_vd import SE_ResNeXt50_vd_32x4d, SE_ResNeXt50_vd_32x4d, SENet154_vd
+from .model_zoo.se_resnext import SE_ResNeXt50_32x4d, SE_ResNeXt101_32x4d, SE_ResNeXt152_64x4d
+from .model_zoo.dpn import DPN68, DPN92, DPN98, DPN107, DPN131
+from .model_zoo.densenet import DenseNet121, DenseNet161, DenseNet169, DenseNet201, DenseNet264
+from .model_zoo.efficientnet import EfficientNetB0, EfficientNetB1, EfficientNetB2, EfficientNetB3, EfficientNetB4, EfficientNetB5, EfficientNetB6, EfficientNetB7, EfficientNetB0_small
+from .model_zoo.resnest import ResNeSt50_fast_1s1x64d, ResNeSt50, ResNeSt101
+from .model_zoo.googlenet import GoogLeNet
+from .model_zoo.mobilenet_v2 import MobileNetV2_x0_25, MobileNetV2_x0_5, MobileNetV2_x0_75, MobileNetV2, MobileNetV2_x1_5, MobileNetV2_x2_0
+from .model_zoo.shufflenet_v2 import ShuffleNetV2_x0_25, ShuffleNetV2_x0_33, ShuffleNetV2_x0_5, ShuffleNetV2_x1_0, ShuffleNetV2_x1_5, ShuffleNetV2_x2_0, ShuffleNetV2_swish
+from .model_zoo.ghostnet import GhostNet_x0_5, GhostNet_x1_0, GhostNet_x1_3
+from .model_zoo.alexnet import AlexNet
+from .model_zoo.inception_v4 import InceptionV4
+from .model_zoo.xception import Xception41, Xception65, Xception71
+from .model_zoo.xception_deeplab import Xception41_deeplab, Xception65_deeplab
+from .model_zoo.resnext101_wsl import ResNeXt101_32x8d_wsl, ResNeXt101_32x16d_wsl, ResNeXt101_32x32d_wsl, ResNeXt101_32x48d_wsl
+from .model_zoo.squeezenet import SqueezeNet1_0, SqueezeNet1_1
+from .model_zoo.darknet import DarkNet53
+from .model_zoo.regnet import RegNetX_200MF, RegNetX_4GF, RegNetX_32GF, RegNetY_200MF, RegNetY_4GF, RegNetY_32GF
+from .model_zoo.vision_transformer import ViT_small_patch16_224, ViT_base_patch16_224, ViT_base_patch16_384, ViT_base_patch32_384, ViT_large_patch16_224, ViT_large_patch16_384, ViT_large_patch32_384
+from .model_zoo.distilled_vision_transformer import DeiT_tiny_patch16_224, DeiT_small_patch16_224, DeiT_base_patch16_224, DeiT_tiny_distilled_patch16_224, DeiT_small_distilled_patch16_224, DeiT_base_distilled_patch16_224, DeiT_base_patch16_384, DeiT_base_distilled_patch16_384
+from .legendary_models.swin_transformer import SwinTransformer_tiny_patch4_window7_224, SwinTransformer_small_patch4_window7_224, SwinTransformer_base_patch4_window7_224, SwinTransformer_base_patch4_window12_384, SwinTransformer_large_patch4_window7_224, SwinTransformer_large_patch4_window12_384
+from .model_zoo.cswin_transformer import CSWinTransformer_tiny_224, CSWinTransformer_small_224, CSWinTransformer_base_224, CSWinTransformer_large_224, CSWinTransformer_base_384, CSWinTransformer_large_384
+from .model_zoo.mixnet import MixNet_S, MixNet_M, MixNet_L
+from .model_zoo.rexnet import ReXNet_1_0, ReXNet_1_3, ReXNet_1_5, ReXNet_2_0, ReXNet_3_0
+from .model_zoo.gvt import pcpvt_small, pcpvt_base, pcpvt_large, alt_gvt_small, alt_gvt_base, alt_gvt_large
+from .model_zoo.levit import LeViT_128S, LeViT_128, LeViT_192, LeViT_256, LeViT_384
+from .model_zoo.dla import DLA34, DLA46_c, DLA46x_c, DLA60, DLA60x, DLA60x_c, DLA102, DLA102x, DLA102x2, DLA169
+from .model_zoo.rednet import RedNet26, RedNet38, RedNet50, RedNet101, RedNet152
+from .model_zoo.tnt import TNT_small
+from .model_zoo.hardnet import HarDNet68, HarDNet85, HarDNet39_ds, HarDNet68_ds
+from .model_zoo.cspnet import CSPDarkNet53
+from .model_zoo.pvt_v2 import PVT_V2_B0, PVT_V2_B1, PVT_V2_B2_Linear, PVT_V2_B2, PVT_V2_B3, PVT_V2_B4, PVT_V2_B5
+from .model_zoo.mobilevit import MobileViT_XXS, MobileViT_XS, MobileViT_S
+from .model_zoo.repvgg import RepVGG_A0, RepVGG_A1, RepVGG_A2, RepVGG_B0, RepVGG_B1, RepVGG_B2, RepVGG_B1g2, RepVGG_B1g4, RepVGG_B2g4, RepVGG_B3g4
+from .model_zoo.van import VAN_tiny
+from .model_zoo.peleenet import PeleeNet
+from .model_zoo.convnext import ConvNeXt_tiny
 
-from ppcls.arch.backbone.variant_models.resnet_variant import ResNet50_last_stage_stride1
-from ppcls.arch.backbone.variant_models.vgg_variant import VGG19Sigmoid
-from ppcls.arch.backbone.variant_models.pp_lcnet_variant import PPLCNet_x2_5_Tanh
-from ppcls.arch.backbone.model_zoo.adaface_ir_net import AdaFace_IR_18, AdaFace_IR_34, AdaFace_IR_50, AdaFace_IR_101, AdaFace_IR_152, AdaFace_IR_SE_50, AdaFace_IR_SE_101, AdaFace_IR_SE_152, AdaFace_IR_SE_200
+from .variant_models.resnet_variant import ResNet50_last_stage_stride1
+from .variant_models.vgg_variant import VGG19Sigmoid
+from .variant_models.pp_lcnet_variant import PPLCNet_x2_5_Tanh
+from .model_zoo.adaface_ir_net import AdaFace_IR_18, AdaFace_IR_34, AdaFace_IR_50, AdaFace_IR_101, AdaFace_IR_152, AdaFace_IR_SE_50, AdaFace_IR_SE_101, AdaFace_IR_SE_152, AdaFace_IR_SE_200
 
 
 # help whl get all the models' api (class type) and components' api (func type)

ppcls/arch/backbone/base/theseus_layer.py

@@ -15,7 +15,7 @@
 from typing import Tuple, List, Dict, Union, Callable, Any
 
 from paddle import nn
-from ppcls.utils import logger
+from ....utils import logger
 
 
 class Identity(nn.Layer):

ppcls/arch/backbone/legendary_models/esnet.py

@@ -22,8 +22,8 @@ from paddle.nn import AdaptiveAvgPool2D, MaxPool2D
 from paddle.nn.initializer import KaimingNormal
 from paddle.regularizer import L2Decay
 
-from ppcls.arch.backbone.base.theseus_layer import TheseusLayer
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ..base.theseus_layer import TheseusLayer
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "ESNet_x0_25":

ppcls/arch/backbone/legendary_models/hrnet.py

@@ -25,8 +25,8 @@ from paddle import ParamAttr
 from paddle.nn.functional import upsample
 from paddle.nn.initializer import Uniform
 
-from ppcls.arch.backbone.base.theseus_layer import TheseusLayer, Identity
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ..base.theseus_layer import TheseusLayer, Identity
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "HRNet_W18_C":

ppcls/arch/backbone/legendary_models/inception_v3.py

@@ -23,8 +23,8 @@ from paddle.nn import Conv2D, BatchNorm, Linear, Dropout
 from paddle.nn import AdaptiveAvgPool2D, MaxPool2D, AvgPool2D
 from paddle.nn.initializer import Uniform
 
-from ppcls.arch.backbone.base.theseus_layer import TheseusLayer
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ..base.theseus_layer import TheseusLayer
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "InceptionV3":

ppcls/arch/backbone/legendary_models/mobilenet_v1.py

@@ -22,8 +22,8 @@ from paddle.nn import Conv2D, BatchNorm, Linear, ReLU, Flatten
 from paddle.nn import AdaptiveAvgPool2D
 from paddle.nn.initializer import KaimingNormal
 
-from ppcls.arch.backbone.base.theseus_layer import TheseusLayer
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ..base.theseus_layer import TheseusLayer
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "MobileNetV1_x0_25":

ppcls/arch/backbone/legendary_models/mobilenet_v3.py

@@ -21,8 +21,9 @@ import paddle.nn as nn
 from paddle import ParamAttr
 from paddle.nn import AdaptiveAvgPool2D, BatchNorm, Conv2D, Dropout, Linear
 from paddle.regularizer import L2Decay
-from ppcls.arch.backbone.base.theseus_layer import TheseusLayer
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+
+from ..base.theseus_layer import TheseusLayer
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "MobileNetV3_small_x0_35":

ppcls/arch/backbone/legendary_models/pp_hgnet.py

@@ -20,8 +20,8 @@ from paddle.nn import Conv2D, BatchNorm2D, ReLU, AdaptiveAvgPool2D, MaxPool2D
 from paddle.regularizer import L2Decay
 from paddle import ParamAttr
 
-from ppcls.arch.backbone.base.theseus_layer import TheseusLayer
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ..base.theseus_layer import TheseusLayer
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "PPHGNet_tiny":
@@ -199,6 +199,7 @@ class PPHGNet(TheseusLayer):
     Returns:
         model: nn.Layer. Specific PPHGNet model depends on args.
     """
+
     def __init__(self,
                  stem_channels,
                  stage_config,
@@ -230,7 +231,7 @@ class PPHGNet(TheseusLayer):
                 k]
             self.stages.append(
                 HG_Stage(in_channels, mid_channels, out_channels, block_num,
-                          layer_num, downsample))
+                         layer_num, downsample))
 
         self.avg_pool = AdaptiveAvgPool2D(1)
         if self.use_last_conv:

ppcls/arch/backbone/legendary_models/pp_lcnet.py

@@ -20,8 +20,9 @@ from paddle import ParamAttr
 from paddle.nn import AdaptiveAvgPool2D, BatchNorm2D, Conv2D, Dropout, Linear
 from paddle.regularizer import L2Decay
 from paddle.nn.initializer import KaimingNormal
-from ppcls.arch.backbone.base.theseus_layer import TheseusLayer
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+
+from ..base.theseus_layer import TheseusLayer
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "PPLCNet_x0_25":
@@ -229,64 +230,59 @@ class PPLCNet(TheseusLayer):
             stride=stride_list[0],
             lr_mult=self.lr_mult_list[0])
 
-        self.blocks2 = nn.Sequential(*[
+        self.blocks2 = nn.Sequential(* [
             DepthwiseSeparable(
                 num_channels=make_divisible(in_c * scale),
                 num_filters=make_divisible(out_c * scale),
                 dw_size=k,
                 stride=s,
                 use_se=se,
-                lr_mult=self.lr_mult_list[1])
-            for i, (k, in_c, out_c, s, se
-                    ) in enumerate(self.net_config["blocks2"])
+                lr_mult=self.lr_mult_list[1]) for i, (k, in_c, out_c, s, se) in
+            enumerate(self.net_config["blocks2"])
         ])
 
-        self.blocks3 = nn.Sequential(*[
+        self.blocks3 = nn.Sequential(* [
             DepthwiseSeparable(
                 num_channels=make_divisible(in_c * scale),
                 num_filters=make_divisible(out_c * scale),
                 dw_size=k,
                 stride=s,
                 use_se=se,
-                lr_mult=self.lr_mult_list[2])
-            for i, (k, in_c, out_c, s, se
-                    ) in enumerate(self.net_config["blocks3"])
+                lr_mult=self.lr_mult_list[2]) for i, (k, in_c, out_c, s, se) in
+            enumerate(self.net_config["blocks3"])
         ])
 
-        self.blocks4 = nn.Sequential(*[
+        self.blocks4 = nn.Sequential(* [
             DepthwiseSeparable(
                 num_channels=make_divisible(in_c * scale),
                 num_filters=make_divisible(out_c * scale),
                 dw_size=k,
                 stride=s,
                 use_se=se,
-                lr_mult=self.lr_mult_list[3])
-            for i, (k, in_c, out_c, s, se
-                    ) in enumerate(self.net_config["blocks4"])
+                lr_mult=self.lr_mult_list[3]) for i, (k, in_c, out_c, s, se) in
+            enumerate(self.net_config["blocks4"])
         ])
 
-        self.blocks5 = nn.Sequential(*[
+        self.blocks5 = nn.Sequential(* [
             DepthwiseSeparable(
                 num_channels=make_divisible(in_c * scale),
                 num_filters=make_divisible(out_c * scale),
                 dw_size=k,
                 stride=s,
                 use_se=se,
-                lr_mult=self.lr_mult_list[4])
-            for i, (k, in_c, out_c, s, se
-                    ) in enumerate(self.net_config["blocks5"])
+                lr_mult=self.lr_mult_list[4]) for i, (k, in_c, out_c, s, se) in
+            enumerate(self.net_config["blocks5"])
         ])
 
-        self.blocks6 = nn.Sequential(*[
+        self.blocks6 = nn.Sequential(* [
             DepthwiseSeparable(
                 num_channels=make_divisible(in_c * scale),
                 num_filters=make_divisible(out_c * scale),
                 dw_size=k,
                 stride=s,
                 use_se=se,
-                lr_mult=self.lr_mult_list[5])
-            for i, (k, in_c, out_c, s, se
-                    ) in enumerate(self.net_config["blocks6"])
+                lr_mult=self.lr_mult_list[5]) for i, (k, in_c, out_c, s, se) in
+            enumerate(self.net_config["blocks6"])
         ])
 
         self.avg_pool = AdaptiveAvgPool2D(1)

ppcls/arch/backbone/legendary_models/pp_lcnet_v2.py

@@ -21,8 +21,9 @@ from paddle import ParamAttr
 from paddle.nn import AdaptiveAvgPool2D, BatchNorm2D, Conv2D, Dropout, Linear
 from paddle.regularizer import L2Decay
 from paddle.nn.initializer import KaimingNormal
-from ppcls.arch.backbone.base.theseus_layer import TheseusLayer
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+
+from ..base.theseus_layer import TheseusLayer
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "PPLCNetV2_base":

ppcls/arch/backbone/legendary_models/resnet.py

@@ -26,9 +26,9 @@ from paddle.nn.initializer import Uniform
 from paddle.regularizer import L2Decay
 import math
 
-from ppcls.utils import logger
-from ppcls.arch.backbone.base.theseus_layer import TheseusLayer
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils import logger
+from ..base.theseus_layer import TheseusLayer
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "ResNet18":
@@ -328,7 +328,7 @@ class ResNet(TheseusLayer):
                    [32, 32, 3, 1], [32, 64, 3, 1]]
         }
 
-        self.stem = nn.Sequential(*[
+        self.stem = nn.Sequential(* [
             ConvBNLayer(
                 num_channels=in_c,
                 num_filters=out_c,

ppcls/arch/backbone/legendary_models/swin_transformer.py

@@ -21,23 +21,23 @@ import paddle.nn as nn
 import paddle.nn.functional as F
 from paddle.nn.initializer import TruncatedNormal, Constant
 
-from ppcls.arch.backbone.base.theseus_layer import TheseusLayer
-from ppcls.arch.backbone.model_zoo.vision_transformer import trunc_normal_, zeros_, ones_, to_2tuple, DropPath, Identity
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ..model_zoo.vision_transformer import trunc_normal_, zeros_, ones_, to_2tuple, DropPath, Identity
+from ..base.theseus_layer import TheseusLayer
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "SwinTransformer_tiny_patch4_window7_224":
-    "https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_tiny_patch4_window7_224_pretrained.pdparams",
+    "https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_tiny_patch4_window7_224_pretrained.pdparams",
     "SwinTransformer_small_patch4_window7_224":
-    "https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_small_patch4_window7_224_pretrained.pdparams",
+    "https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_small_patch4_window7_224_pretrained.pdparams",
     "SwinTransformer_base_patch4_window7_224":
-    "https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_base_patch4_window7_224_pretrained.pdparams",
+    "https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_base_patch4_window7_224_pretrained.pdparams",
     "SwinTransformer_base_patch4_window12_384":
-    "https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_base_patch4_window12_384_pretrained.pdparams",
+    "https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_base_patch4_window12_384_pretrained.pdparams",
     "SwinTransformer_large_patch4_window7_224":
-    "https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_large_patch4_window7_224_22kto1k_pretrained.pdparams",
+    "https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_large_patch4_window7_224_22kto1k_pretrained.pdparams",
     "SwinTransformer_large_patch4_window12_384":
-    "https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SwinTransformer_large_patch4_window12_384_22kto1k_pretrained.pdparams",
+    "https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/legendary_models/SwinTransformer_large_patch4_window12_384_22kto1k_pretrained.pdparams",
 }
 
 __all__ = list(MODEL_URLS.keys())

ppcls/arch/backbone/legendary_models/vgg.py

@@ -20,8 +20,8 @@ import paddle.nn as nn
 from paddle.nn import Conv2D, BatchNorm, Linear, Dropout
 from paddle.nn import MaxPool2D
 
-from ppcls.arch.backbone.base.theseus_layer import TheseusLayer
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ..base.theseus_layer import TheseusLayer
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "VGG11":

ppcls/arch/backbone/model_zoo/alexnet.py

@@ -23,7 +23,7 @@ from paddle.nn import AdaptiveAvgPool2D, MaxPool2D, AvgPool2D
 from paddle.nn.initializer import Uniform
 import math
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "AlexNet":

ppcls/arch/backbone/model_zoo/convnext.py

@@ -18,7 +18,7 @@ import paddle
 import paddle.nn as nn
 from paddle.nn.initializer import TruncatedNormal, Constant
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "ConvNeXt_tiny": "",  # TODO
@@ -176,7 +176,7 @@ class ConvNeXt(nn.Layer):
         ]
         cur = 0
         for i in range(4):
-            stage = nn.Sequential(*[
+            stage = nn.Sequential(* [
                 Block(
                     dim=dims[i],
                     drop_path=dp_rates[cur + j],

ppcls/arch/backbone/model_zoo/cspnet.py

@@ -20,7 +20,7 @@ import paddle.nn as nn
 import paddle.nn.functional as F
 from paddle import ParamAttr
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "CSPDarkNet53":

ppcls/arch/backbone/model_zoo/cswin_transformer.py

@@ -21,7 +21,7 @@ import paddle
 import paddle.nn as nn
 from .vision_transformer import trunc_normal_, zeros_, ones_, to_2tuple, DropPath, Identity
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "CSWinTransformer_tiny_224":

ppcls/arch/backbone/model_zoo/darknet.py

@@ -23,7 +23,7 @@ from paddle.nn import AdaptiveAvgPool2D, MaxPool2D, AvgPool2D
 from paddle.nn.initializer import Uniform
 import math
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "DarkNet53":

ppcls/arch/backbone/model_zoo/densenet.py

@@ -28,7 +28,7 @@ from paddle.nn.initializer import Uniform
 
 import math
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "DenseNet121":

ppcls/arch/backbone/model_zoo/distilled_vision_transformer.py

@@ -19,7 +19,7 @@ import paddle
 import paddle.nn as nn
 from .vision_transformer import VisionTransformer, Identity, trunc_normal_, zeros_
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "DeiT_tiny_patch16_224":

ppcls/arch/backbone/model_zoo/dla.py

@@ -23,8 +23,8 @@ import paddle.nn.functional as F
 
 from paddle.nn.initializer import Normal, Constant
 
-from ppcls.arch.backbone.base.theseus_layer import Identity
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ..base.theseus_layer import Identity
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "DLA34":

ppcls/arch/backbone/model_zoo/dpn.py

@@ -29,7 +29,7 @@ from paddle.nn.initializer import Uniform
 
 import math
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "DPN68":

ppcls/arch/backbone/model_zoo/efficientnet.py

@@ -26,7 +26,7 @@ import collections
 import re
 import copy
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "EfficientNetB0_small":

ppcls/arch/backbone/model_zoo/ghostnet.py

@@ -24,7 +24,7 @@ from paddle.nn import Conv2D, BatchNorm, AdaptiveAvgPool2D, Linear
 from paddle.regularizer import L2Decay
 from paddle.nn.initializer import Uniform, KaimingNormal
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "GhostNet_x0_5":

ppcls/arch/backbone/model_zoo/googlenet.py

@@ -24,7 +24,7 @@ from paddle.nn.initializer import Uniform
 
 import math
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "GoogLeNet":

ppcls/arch/backbone/model_zoo/gvt.py

@@ -25,7 +25,7 @@ from paddle.regularizer import L2Decay
 from .vision_transformer import trunc_normal_, normal_, zeros_, ones_, to_2tuple, DropPath, Identity, Mlp
 from .vision_transformer import Block as ViTBlock
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "pcpvt_small":

ppcls/arch/backbone/model_zoo/hardnet.py

@@ -18,7 +18,7 @@
 import paddle
 import paddle.nn as nn
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     'HarDNet39_ds':

ppcls/arch/backbone/model_zoo/inception_v4.py

@@ -23,7 +23,7 @@ from paddle.nn import AdaptiveAvgPool2D, MaxPool2D, AvgPool2D
 from paddle.nn.initializer import Uniform
 import math
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "InceptionV4":

ppcls/arch/backbone/model_zoo/levit.py

@@ -27,7 +27,7 @@ from paddle.regularizer import L2Decay
 
 from .vision_transformer import trunc_normal_, zeros_, ones_, Identity
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "LeViT_128S":

ppcls/arch/backbone/model_zoo/mixnet.py

@@ -20,7 +20,7 @@ from functools import reduce
 import paddle
 import paddle.nn as nn
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "MixNet_S":

ppcls/arch/backbone/model_zoo/mobilenet_v2.py

@@ -28,7 +28,7 @@ from paddle.nn import AdaptiveAvgPool2D, MaxPool2D, AvgPool2D
 
 import math
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "MobileNetV2_x0_25":

ppcls/arch/backbone/model_zoo/mobilevit.py

@@ -23,7 +23,7 @@ import paddle.nn.functional as F
 from paddle.nn.initializer import KaimingUniform, TruncatedNormal, Constant
 import math
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "MobileViT_XXS":

ppcls/arch/backbone/model_zoo/peleenet.py

@@ -22,7 +22,7 @@ import paddle.nn as nn
 import paddle.nn.functional as F
 from paddle.nn.initializer import Normal, Constant
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "PeleeNet": ""  # TODO
@@ -37,7 +37,8 @@ ones_ = Constant(value=1.)
 
 
 class _DenseLayer(nn.Layer):
-    def __init__(self, num_input_features, growth_rate, bottleneck_width, drop_rate):
+    def __init__(self, num_input_features, growth_rate, bottleneck_width,
+                 drop_rate):
         super(_DenseLayer, self).__init__()
 
         growth_rate = int(growth_rate / 2)
@@ -71,11 +72,12 @@ class _DenseLayer(nn.Layer):
 
 
 class _DenseBlock(nn.Sequential):
-    def __init__(self, num_layers, num_input_features, bn_size, growth_rate, drop_rate):
+    def __init__(self, num_layers, num_input_features, bn_size, growth_rate,
+                 drop_rate):
         super(_DenseBlock, self).__init__()
         for i in range(num_layers):
-            layer = _DenseLayer(num_input_features + i *
-                                growth_rate, growth_rate, bn_size, drop_rate)
+            layer = _DenseLayer(num_input_features + i * growth_rate,
+                                growth_rate, bn_size, drop_rate)
             setattr(self, 'denselayer%d' % (i + 1), layer)
 
 
@@ -83,16 +85,32 @@ class _StemBlock(nn.Layer):
     def __init__(self, num_input_channels, num_init_features):
         super(_StemBlock, self).__init__()
 
-        num_stem_features = int(num_init_features/2)
+        num_stem_features = int(num_init_features / 2)
 
         self.stem1 = BasicConv2D(
-            num_input_channels, num_init_features, kernel_size=3, stride=2, padding=1)
+            num_input_channels,
+            num_init_features,
+            kernel_size=3,
+            stride=2,
+            padding=1)
         self.stem2a = BasicConv2D(
-            num_init_features, num_stem_features, kernel_size=1, stride=1, padding=0)
+            num_init_features,
+            num_stem_features,
+            kernel_size=1,
+            stride=1,
+            padding=0)
         self.stem2b = BasicConv2D(
-            num_stem_features, num_init_features, kernel_size=3, stride=2, padding=1)
+            num_stem_features,
+            num_init_features,
+            kernel_size=3,
+            stride=2,
+            padding=1)
         self.stem3 = BasicConv2D(
-            2*num_init_features, num_init_features, kernel_size=1, stride=1, padding=0)
+            2 * num_init_features,
+            num_init_features,
+            kernel_size=1,
+            stride=1,
+            padding=0)
         self.pool = nn.MaxPool2D(kernel_size=2, stride=2)
 
     def forward(self, x):
@@ -109,11 +127,10 @@ class _StemBlock(nn.Layer):
 
 
 class BasicConv2D(nn.Layer):
-
     def __init__(self, in_channels, out_channels, activation=True, **kwargs):
         super(BasicConv2D, self).__init__()
-        self.conv = nn.Conv2D(in_channels, out_channels,
-                              bias_attr=False, **kwargs)
+        self.conv = nn.Conv2D(
+            in_channels, out_channels, bias_attr=False, **kwargs)
         self.norm = nn.BatchNorm2D(out_channels)
         self.activation = activation
 
@@ -141,15 +158,18 @@ class PeleeNetDY(nn.Layer):
         class_num (int) - number of classification classes
     """
 
-    def __init__(self, growth_rate=32, block_config=[3, 4, 8, 6],
-                 num_init_features=32, bottleneck_width=[1, 2, 4, 4],
-                 drop_rate=0.05, class_num=1000):
+    def __init__(self,
+                 growth_rate=32,
+                 block_config=[3, 4, 8, 6],
+                 num_init_features=32,
+                 bottleneck_width=[1, 2, 4, 4],
+                 drop_rate=0.05,
+                 class_num=1000):
 
         super(PeleeNetDY, self).__init__()
 
-        self.features = nn.Sequential(*[
-            ('stemblock', _StemBlock(3, num_init_features)),
-        ])
+        self.features = nn.Sequential(* [('stemblock', _StemBlock(
+            3, num_init_features)), ])
 
         if type(growth_rate) is list:
             growth_rates = growth_rate
@@ -168,20 +188,31 @@ class PeleeNetDY(nn.Layer):
         # Each denseblock
         num_features = num_init_features
         for i, num_layers in enumerate(block_config):
-            block = _DenseBlock(num_layers=num_layers,
-                                num_input_features=num_features,
-                                bn_size=bottleneck_widths[i],
-                                growth_rate=growth_rates[i],
-                                drop_rate=drop_rate)
+            block = _DenseBlock(
+                num_layers=num_layers,
+                num_input_features=num_features,
+                bn_size=bottleneck_widths[i],
+                growth_rate=growth_rates[i],
+                drop_rate=drop_rate)
             setattr(self.features, 'denseblock%d' % (i + 1), block)
             num_features = num_features + num_layers * growth_rates[i]
 
-            setattr(self.features, 'transition%d' % (i + 1), BasicConv2D(
-                num_features, num_features, kernel_size=1, stride=1, padding=0))
+            setattr(
+                self.features,
+                'transition%d' % (i + 1),
+                BasicConv2D(
+                    num_features,
+                    num_features,
+                    kernel_size=1,
+                    stride=1,
+                    padding=0))
 
             if i != len(block_config) - 1:
-                setattr(self.features, 'transition%d_pool' %
-                        (i + 1), nn.AvgPool2D(kernel_size=2, stride=2))
+                setattr(
+                    self.features,
+                    'transition%d_pool' % (i + 1),
+                    nn.AvgPool2D(
+                        kernel_size=2, stride=2))
                 num_features = num_features
 
         # Linear layer
@@ -192,7 +223,8 @@ class PeleeNetDY(nn.Layer):
 
     def forward(self, x):
         features = self.features(x)
-        out = F.avg_pool2d(features, kernel_size=features.shape[2:4]).flatten(1)
+        out = F.avg_pool2d(
+            features, kernel_size=features.shape[2:4]).flatten(1)
         if self.drop_rate > 0:
             out = F.dropout(out, p=self.drop_rate, training=self.training)
         out = self.classifier(out)

ppcls/arch/backbone/model_zoo/pvt_v2.py

@@ -24,7 +24,7 @@ from paddle.nn.initializer import TruncatedNormal, Constant
 
 from .vision_transformer import trunc_normal_, zeros_, ones_, to_2tuple, DropPath, Identity, drop_path
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "PVT_V2_B0":

ppcls/arch/backbone/model_zoo/rednet.py

@@ -20,7 +20,7 @@ import paddle.nn as nn
 
 from paddle.vision.models import resnet
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "RedNet26":

ppcls/arch/backbone/model_zoo/regnet.py

@@ -29,7 +29,7 @@ from paddle.nn import AdaptiveAvgPool2D, MaxPool2D, AvgPool2D
 from paddle.nn.initializer import Uniform
 import math
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "RegNetX_200MF":

ppcls/arch/backbone/model_zoo/repvgg.py

@@ -19,7 +19,7 @@ import paddle.nn as nn
 import paddle
 import numpy as np
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "RepVGG_A0":

ppcls/arch/backbone/model_zoo/res2net.py

@@ -29,7 +29,7 @@ from paddle.nn.initializer import Uniform
 
 import math
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "Res2Net50_26w_4s":

ppcls/arch/backbone/model_zoo/res2net_vd.py

@@ -29,7 +29,7 @@ from paddle.nn.initializer import Uniform
 
 import math
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "Res2Net50_vd_26w_4s":

ppcls/arch/backbone/model_zoo/resnest.py

@@ -30,7 +30,7 @@ from paddle.nn import Conv2D, BatchNorm, Linear, Dropout
 from paddle.nn import AdaptiveAvgPool2D, MaxPool2D, AvgPool2D
 from paddle.regularizer import L2Decay
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "ResNeSt50_fast_1s1x64d":

ppcls/arch/backbone/model_zoo/resnet_vc.py

@@ -29,7 +29,7 @@ from paddle.nn.initializer import Uniform
 
 import math
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "ResNet50_vc":

ppcls/arch/backbone/model_zoo/resnext.py

@@ -29,7 +29,7 @@ from paddle.nn.initializer import Uniform
 
 import math
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "ResNeXt50_32x4d":

ppcls/arch/backbone/model_zoo/resnext101_wsl.py

@@ -22,7 +22,7 @@ from paddle.nn import Conv2D, BatchNorm, Linear, Dropout
 from paddle.nn import AdaptiveAvgPool2D, MaxPool2D, AvgPool2D
 from paddle.nn.initializer import Uniform
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "ResNeXt101_32x8d_wsl":

ppcls/arch/backbone/model_zoo/resnext_vd.py

@@ -29,7 +29,7 @@ from paddle.nn.initializer import Uniform
 
 import math
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "ResNeXt50_vd_32x4d":

ppcls/arch/backbone/model_zoo/rexnet.py

@@ -24,7 +24,7 @@ from paddle import ParamAttr
 import paddle.nn as nn
 from math import ceil
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "ReXNet_1_0":

ppcls/arch/backbone/model_zoo/se_resnet_vd.py

@@ -28,7 +28,7 @@ from paddle.nn.initializer import Uniform
 
 import math
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "SE_ResNet18_vd":

ppcls/arch/backbone/model_zoo/se_resnext.py

@@ -29,7 +29,7 @@ from paddle.nn.initializer import Uniform
 
 import math
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "SE_ResNeXt50_32x4d":

ppcls/arch/backbone/model_zoo/se_resnext_vd.py

@@ -29,7 +29,7 @@ from paddle.nn.initializer import Uniform
 
 import math
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "SE_ResNeXt50_vd_32x4d":

ppcls/arch/backbone/model_zoo/shufflenet_v2.py

@@ -24,7 +24,7 @@ from paddle.nn import Layer, Conv2D, MaxPool2D, AdaptiveAvgPool2D, BatchNorm, Li
 from paddle.nn.initializer import KaimingNormal
 from paddle.nn.functional import swish
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "ShuffleNetV2_x0_25":

ppcls/arch/backbone/model_zoo/squeezenet.py

@@ -21,7 +21,7 @@ import paddle.nn.functional as F
 from paddle.nn import Conv2D, BatchNorm, Linear, Dropout
 from paddle.nn import AdaptiveAvgPool2D, MaxPool2D, AvgPool2D
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "SqueezeNet1_0":

ppcls/arch/backbone/model_zoo/tnt.py

@@ -23,8 +23,8 @@ import paddle.nn as nn
 
 from paddle.nn.initializer import TruncatedNormal, Constant
 
-from ppcls.arch.backbone.base.theseus_layer import Identity
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ..base.theseus_layer import Identity
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "TNT_small":

ppcls/arch/backbone/model_zoo/van.py

@@ -21,7 +21,7 @@ import paddle
 import paddle.nn as nn
 from paddle.nn.initializer import TruncatedNormal, Constant
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "VAN_tiny": "",  # TODO

ppcls/arch/backbone/model_zoo/vision_transformer.py

@@ -22,7 +22,7 @@ import paddle
 import paddle.nn as nn
 from paddle.nn.initializer import TruncatedNormal, Constant, Normal
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "ViT_small_patch16_224":

ppcls/arch/backbone/model_zoo/xception.py

@@ -24,7 +24,7 @@ from paddle.nn.initializer import Uniform
 import math
 import sys
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "Xception41":

ppcls/arch/backbone/model_zoo/xception_deeplab.py

@@ -21,7 +21,7 @@ import paddle.nn.functional as F
 from paddle.nn import Conv2D, BatchNorm, Linear, Dropout
 from paddle.nn import AdaptiveAvgPool2D, MaxPool2D, AvgPool2D
 
-from ppcls.utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
+from ....utils.save_load import load_dygraph_pretrain, load_dygraph_pretrain_from_url
 
 MODEL_URLS = {
     "Xception41_deeplab":

ppcls/arch/backbone/variant_models/pp_lcnet_variant.py

 import paddle
 from paddle.nn import Sigmoid
 from paddle.nn import Tanh
-from ppcls.arch.backbone.legendary_models.pp_lcnet import PPLCNet_x2_5
+from ..legendary_models.pp_lcnet import PPLCNet_x2_5
 
 __all__ = ["PPLCNet_x2_5_Tanh"]
 

ppcls/arch/backbone/variant_models/resnet_variant.py

 from paddle.nn import Conv2D
-from ppcls.arch.backbone.legendary_models.resnet import ResNet50, MODEL_URLS, _load_pretrained
+from ..legendary_models.resnet import ResNet50, MODEL_URLS, _load_pretrained
 
 __all__ = ["ResNet50_last_stage_stride1"]
 

ppcls/arch/backbone/variant_models/vgg_variant.py

 import paddle
 from paddle.nn import Sigmoid
-from ppcls.arch.backbone.legendary_models.vgg import VGG19
+from ..legendary_models.vgg import VGG19
 
 __all__ = ["VGG19Sigmoid"]
 

ppcls/arch/gears/bnneck.py

@@ -17,7 +17,7 @@ from __future__ import absolute_import, division, print_function
 import paddle
 import paddle.nn as nn
 
-from ppcls.arch.utils import get_param_attr_dict
+from ..utils import get_param_attr_dict
 
 
 class BNNeck(nn.Layer):

ppcls/arch/gears/fc.py

@@ -19,7 +19,7 @@ from __future__ import print_function
 import paddle
 import paddle.nn as nn
 
-from ppcls.arch.utils import get_param_attr_dict
+from ..utils import get_param_attr_dict
 
 
 class FC(nn.Layer):

ppcls/arch/slim/__init__.py

@@ -12,5 +12,5 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-from ppcls.arch.slim.prune import prune_model
-from ppcls.arch.slim.quant import quantize_model
+from .prune import prune_model
+from .quant import quantize_model

ppcls/arch/slim/prune.py

@@ -14,7 +14,7 @@
 
 from __future__ import absolute_import, division, print_function
 import paddle
-from ppcls.utils import logger
+from ...utils import logger
 
 
 def prune_model(config, model):
@@ -37,7 +37,6 @@ def prune_model(config, model):
         model.pruner = None
 
 
-
 def _prune_model(config, model):
     from paddleslim.analysis import dygraph_flops as flops
     logger.info("FLOPs before pruning: {}GFLOPs".format(

ppcls/arch/slim/quant.py

@@ -14,7 +14,7 @@
 
 from __future__ import absolute_import, division, print_function
 import paddle
-from ppcls.utils import logger
+from ...utils import logger
 
 QUANT_CONFIG = {
     # weight preprocess type, default is None and no preprocessing is performed.

ppcls/configs/ImageNet/Distillation/mv3_large_x1_0_distill_mv3_small_x1_0.yaml

@@ -140,8 +140,7 @@ Infer:
           order: ''
       - ToCHWImage:
   PostProcess:
-    name: DistillationPostProcess
-    func: Topk
+    name: Topk
     topk: 5
     class_id_map_file: "ppcls/utils/imagenet1k_label_list.txt"
 

ppcls/configs/ImageNet/Distillation/res2net200_vd_distill_pphgnet_base.yaml

@@ -153,8 +153,7 @@ Infer:
           order: ''
       - ToCHWImage:
   PostProcess:
-    name: DistillationPostProcess
-    func: Topk
+    name: Topk
     topk: 5
     class_id_map_file: "ppcls/utils/imagenet1k_label_list.txt"
 

ppcls/configs/ImageNet/Distillation/resnet34_distill_resnet18_afd.yaml

@@ -184,8 +184,7 @@ Infer:
           order: ''
       - ToCHWImage:
   PostProcess:
-    name: DistillationPostProcess
-    func: Topk
+    name: Topk
     topk: 5
     class_id_map_file: "ppcls/utils/imagenet1k_label_list.txt"
 

ppcls/configs/ImageNet/Distillation/resnet34_distill_resnet18_dkd.yaml

@@ -139,8 +139,7 @@ Infer:
           order: ''
       - ToCHWImage:
   PostProcess:
-    name: DistillationPostProcess
-    func: Topk
+    name: Topk
     topk: 5
     class_id_map_file: "ppcls/utils/imagenet1k_label_list.txt"
 

ppcls/data/dataloader/DistributedRandomIdentitySampler.py

@@ -14,24 +14,27 @@
 
 from __future__ import absolute_import
 from __future__ import division
-from collections import defaultdict
-import numpy as np
+
 import copy
 import random
+from collections import defaultdict
+
+import numpy as np
 from paddle.io import DistributedBatchSampler, Sampler
 
 
 class DistributedRandomIdentitySampler(DistributedBatchSampler):
-    """
-    Randomly sample N identities, then for each identity,
-    randomly sample K instances, therefore batch size is N*K.
+    """Randomly sample N identities, then for each identity,
+       randomly sample K instances, therefore batch size equals to N * K.
     Args:
-    - data_source (list): list of (img_path, pid, camid).
-    - num_instances (int): number of instances per identity in a batch.
-    - batch_size (int): number of examples in a batch.
+        dataset(Dataset): Dataset which contains list of (img_path, pid, camid))
+        batch_size (int): batch size
+        num_instances (int): number of instance(s) within an class
+        drop_last (bool): whether to discard the data at the end
     """
-
     def __init__(self, dataset, batch_size, num_instances, drop_last, **args):
+        assert batch_size % num_instances == 0, \
+            f"batch_size({batch_size}) must be divisible by num_instances({num_instances}) when using DistributedRandomIdentitySampler"
         self.dataset = dataset
         self.batch_size = batch_size
         self.num_instances = num_instances

ppcls/data/dataloader/pk_sampler.py

@@ -14,27 +14,27 @@
 
 from __future__ import absolute_import
 from __future__ import division
+
 from collections import defaultdict
+
 import numpy as np
-import random
 from paddle.io import DistributedBatchSampler
-
 from ppcls.utils import logger
 
 
 class PKSampler(DistributedBatchSampler):
-    """
-    First, randomly sample P identities.
-    Then for each identity randomly sample K instances.
-    Therefore batch size is P*K, and the sampler called PKSampler.
+    """First, randomly sample P identities.
+        Then for each identity randomly sample K instances.
+        Therefore batch size equals to P * K, and the sampler called PKSampler.
+
     Args:
-        dataset (paddle.io.Dataset): list of (img_path, pid, cam_id).
-        sample_per_id(int): number of instances per identity in a batch.
-        batch_size (int): number of examples in a batch.
-        shuffle(bool): whether to shuffle indices order before generating
-            batch indices. Default False.
+        dataset (Dataset): Dataset which contains list of (img_path, pid, camid))
+        batch_size (int): batch size
+        sample_per_id (int): number of instance(s) within an class
+        shuffle (bool, optional): _description_. Defaults to True.
+        drop_last (bool, optional): whether to discard the data at the end. Defaults to True.
+        sample_method (str, optional): sample method when generating prob_list. Defaults to "sample_avg_prob".
     """
-
     def __init__(self,
                  dataset,
                  batch_size,
@@ -42,10 +42,9 @@ class PKSampler(DistributedBatchSampler):
                  shuffle=True,
                  drop_last=True,
                  sample_method="sample_avg_prob"):
-        super().__init__(
-            dataset, batch_size, shuffle=shuffle, drop_last=drop_last)
+        super().__init__(dataset, batch_size, shuffle=shuffle, drop_last=drop_last)
         assert batch_size % sample_per_id == 0, \
-            "PKSampler configs error, Sample_per_id must be a divisor of batch_size."
+            f"PKSampler configs error, sample_per_id({sample_per_id}) must be a divisor of batch_size({batch_size})."
         assert hasattr(self.dataset,
                        "labels"), "Dataset must have labels attribute."
         self.sample_per_label = sample_per_id

ppcls/utils/check.py

@@ -22,10 +22,8 @@ import sys
 import paddle
 from paddle import is_compiled_with_cuda
 
-from ppcls.arch import get_architectures
-from ppcls.arch import similar_architectures
-from ppcls.arch import get_blacklist_model_in_static_mode
-from ppcls.utils import logger
+from ..arch.utils import get_architectures, similar_architectures, get_blacklist_model_in_static_mode
+from . import logger
 
 
 def check_version():

ppcls/utils/config.py

@@ -16,8 +16,9 @@ import os
 import copy
 import argparse
 import yaml
-from ppcls.utils import logger
-from ppcls.utils import check
+from . import logger
+from . import check
+
 __all__ = ['get_config']
 
 

ppcls/utils/download.py

@@ -28,7 +28,7 @@ import time
 from collections import OrderedDict
 from tqdm import tqdm
 
-from ppcls.utils import logger
+from . import logger
 
 __all__ = ['get_weights_path_from_url']
 

ppcls/utils/gallery2fc.py

@@ -20,12 +20,12 @@ import sys
 __dir__ = os.path.dirname(os.path.abspath(__file__))
 sys.path.append(os.path.abspath(os.path.join(__dir__, '../../')))
 
-from ppcls.arch import build_model
-from ppcls.utils.config import parse_config, parse_args
-from ppcls.utils.save_load import load_dygraph_pretrain
-from ppcls.utils.logger import init_logger
-from ppcls.data import create_operators
-from ppcls.arch.slim import quantize_model
+from ..arch import build_model
+from .config import parse_config, parse_args
+from .save_load import load_dygraph_pretrain
+from .logger import init_logger
+from ..data import create_operators
+from ..arch.slim import quantize_model
 
 
 class GalleryLayer(paddle.nn.Layer):

ppcls/utils/model_zoo.py

@@ -23,8 +23,8 @@ import tarfile
 import tqdm
 import zipfile
 
-from ppcls.arch import similar_architectures
-from ppcls.utils import logger
+from ..arch.utils import similar_architectures
+from . import logger
 
 __all__ = ['get']
 

ppcls/utils/save_load.py

@@ -20,7 +20,7 @@ import errno
 import os
 
 import paddle
-from ppcls.utils import logger
+from . import logger
 from .download import get_weights_path_from_url
 
 __all__ = ['init_model', 'save_model', 'load_dygraph_pretrain']

setup.py

@@ -33,7 +33,7 @@ setup(
     package_dir={'paddleclas': ''},
     include_package_data=True,
     entry_points={
-        "console_scripts": ["paddleclas= paddleclas.paddleclas:main"]
+        "console_scripts": ["paddleclas=paddleclas.paddleclas:main"]
     },
     version='0.0.0',
     install_requires=requirements,