paddlex 3.0.0b2__py3-none-any.whl → 3.0.0rc1__py3-none-any.whl

paddlex/modules/text_recognition/dataset_checker/dataset_src/analyse_dataset.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -13,23 +13,24 @@
 # limitations under the License.
 
 
-import os
-import json
 import math
+import os
 import platform
-from pathlib import Path
-
 from collections import defaultdict
-from PIL import Image
-import cv2
+
 import numpy as np
-import matplotlib.pyplot as plt
-from matplotlib.backends.backend_agg import FigureCanvasAgg
-from matplotlib import font_manager
 
+from .....utils.deps import function_requires_deps, is_dep_available
 from .....utils.file_interface import custom_open
-from .....utils.logging import warning
 from .....utils.fonts import PINGFANG_FONT_FILE_PATH
+from .....utils.logging import warning
+
+if is_dep_available("opencv-contrib-python"):
+    import cv2
+if is_dep_available("matplotlib"):
+    import matplotlib.pyplot as plt
+    from matplotlib import font_manager
+    from matplotlib.backends.backend_agg import FigureCanvasAgg
 
 
 def simple_analyse(dataset_path, images_dict):
@@ -77,10 +78,10 @@ def simple_analyse(dataset_path, images_dict):
     )
 
 
+@function_requires_deps("matplotlib", "opencv-contrib-python")
 def deep_analyse(dataset_path, output, datatype="MSTextRecDataset"):
     """class analysis for dataset"""
     tags = ["train", "val"]
-    all_instances = 0
     labels_cnt = {}
     x_max = []
     classes_max = []

paddlex/modules/text_recognition/dataset_checker/dataset_src/check_dataset.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -17,11 +17,9 @@ import os
 import os.path as osp
 from collections import defaultdict
 
-from PIL import Image
-import json
-import numpy as np
+from PIL import Image, ImageOps
 
-from .....utils.errors import DatasetFileNotFoundError, CheckFailedError
+from .....utils.errors import CheckFailedError, DatasetFileNotFoundError
 
 
 def check(
@@ -35,7 +33,6 @@ def check(
         tags = ["train", "val"]
         delim = "\t"
         valid_num_parts = 2
-        max_recorded_sample_cnts = 50
         sample_cnts = dict()
         sample_paths = defaultdict(list)
         if dataset_type == "LaTeXOCRDataset":
@@ -79,11 +76,21 @@ def check(
                             )
                         file_name = substr[0]
                         img_path = osp.join(dataset_dir, file_name)
-                        if len(sample_paths[tag]) < max_recorded_sample_cnts:
-                            sample_paths[tag].append(os.path.relpath(img_path, output))
 
                         if not os.path.exists(img_path):
                             raise DatasetFileNotFoundError(file_path=img_path)
+                        vis_save_dir = osp.join(output, "demo_img")
+                        if not osp.exists(vis_save_dir):
+                            os.makedirs(vis_save_dir)
+                        if len(sample_paths[tag]) < sample_num:
+                            img = Image.open(img_path)
+                            img = ImageOps.exif_transpose(img)
+                            vis_path = osp.join(vis_save_dir, osp.basename(file_name))
+                            img.save(vis_path)
+                            sample_path = osp.join(
+                                "check_dataset", os.path.relpath(vis_path, output)
+                            )
+                            sample_paths[tag].append(sample_path)
 
         meta = {}
         meta["train_samples"] = sample_cnts["train"]

paddlex/modules/text_recognition/dataset_checker/dataset_src/convert_dataset.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -13,23 +13,24 @@
 # limitations under the License.
 
 
-import os
-import shutil
-import json
-import random
 import math
+import os
 import pickle
-from tqdm import tqdm
 from collections import defaultdict
-import imagesize
+
+from .....utils.deps import function_requires_deps, is_dep_available
 from .....utils.errors import ConvertFailedError
-from .....utils.logging import info, warning
+
+if is_dep_available("imagesize"):
+    import imagesize
+if is_dep_available("tqdm"):
+    from tqdm import tqdm
 
 
 def check_src_dataset(root_dir, dataset_type):
     """check src dataset format validity"""
     if dataset_type in ("MSTextRecDataset"):
-        anno_suffix = ".txt"
+        pass
     else:
         raise ConvertFailedError(
             message=f"数据格式转换失败！不支持{dataset_type}格式数据集。当前仅支持 MSTextRecDataset 格式。"
@@ -65,12 +66,12 @@ def convert_pkl_dataset(root_dir):
         txt2pickle(src_img_dir, src_anno_path, root_dir)
 
 
+@function_requires_deps("tqdm", "imagesize")
 def txt2pickle(images, equations, save_dir):
     phase = os.path.basename(equations).replace(".txt", "")
     save_p = os.path.join(save_dir, "latexocr_{}.pkl".format(phase))
     min_dimensions = (32, 32)
     max_dimensions = (672, 192)
-    max_length = 512
     data = defaultdict(lambda: [])
     pic_num = 0
     if images is not None and equations is not None:

paddlex/modules/text_recognition/dataset_checker/dataset_src/split_dataset.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -14,7 +14,6 @@
 
 
 import os
-import shutil
 from random import shuffle
 
 from .....utils.file_interface import custom_open

paddlex/modules/text_recognition/evaluator.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -17,9 +17,6 @@ from pathlib import Path
 
 from ..base import BaseEvaluator
 from .model_list import MODELS
-from ..formula_recognition.model_list import MODELS as MODELS_LaTeX
-
-MODELS = MODELS + MODELS_LaTeX
 
 
 class TextRecEvaluator(BaseEvaluator):
@@ -35,6 +32,10 @@ class TextRecEvaluator(BaseEvaluator):
             self.pdx_config.update_dataset(
                 self.global_config.dataset_dir, "LaTeXOCRDataSet"
             )
+        elif "PP-OCRv3" in self.global_config["model"]:
+            self.pdx_config.update_dataset(
+                self.global_config.dataset_dir, "SimpleDataSet"
+            )
         else:
             self.pdx_config.update_dataset(
                 self.global_config.dataset_dir, "MSTextRecDataset"

paddlex/modules/text_recognition/exportor.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -14,9 +14,6 @@
 
 from ..base import BaseExportor
 from .model_list import MODELS
-from ..formula_recognition.model_list import MODELS as MODELS_LaTeX
-
-MODELS = MODELS + MODELS_LaTeX
 
 
 class TextRecExportor(BaseExportor):

paddlex/modules/text_recognition/model_list.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -13,8 +13,22 @@
 # limitations under the License.
 
 MODELS = [
+    "PP-OCRv3_mobile_rec",
+    "en_PP-OCRv3_mobile_rec",
+    "korean_PP-OCRv3_mobile_rec",
+    "japan_PP-OCRv3_mobile_rec",
+    "chinese_cht_PP-OCRv3_mobile_rec",
+    "te_PP-OCRv3_mobile_rec",
+    "ka_PP-OCRv3_mobile_rec",
+    "ta_PP-OCRv3_mobile_rec",
+    "latin_PP-OCRv3_mobile_rec",
+    "arabic_PP-OCRv3_mobile_rec",
+    "cyrillic_PP-OCRv3_mobile_rec",
+    "devanagari_PP-OCRv3_mobile_rec",
     "PP-OCRv4_mobile_rec",
     "PP-OCRv4_server_rec",
+    "en_PP-OCRv4_mobile_rec",
+    "PP-OCRv4_server_rec_doc",
     "ch_SVTRv2_rec",
     "ch_RepSVTR_rec",
 ]

paddlex/modules/text_recognition/trainer.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -13,16 +13,11 @@
 # limitations under the License.
 
 
-import os
 import shutil
 from pathlib import Path
 
 from ..base import BaseTrainer
-from ...utils.config import AttrDict
 from .model_list import MODELS
-from ..formula_recognition.model_list import MODELS as MODELS_LaTeX
-
-MODELS = MODELS + MODELS_LaTeX
 
 
 class TextRecTrainer(BaseTrainer):
@@ -54,6 +49,10 @@ class TextRecTrainer(BaseTrainer):
             self.pdx_config.update_dataset(
                 self.global_config.dataset_dir, "LaTeXOCRDataSet"
             )
+        elif "PP-OCRv3" in self.global_config["model"]:
+            self.pdx_config.update_dataset(
+                self.global_config.dataset_dir, "SimpleDataSet"
+            )
         else:
             self.pdx_config.update_dataset(
                 self.global_config.dataset_dir, "MSTextRecDataset"
@@ -102,4 +101,5 @@ class TextRecTrainer(BaseTrainer):
         return {
             "device": self.get_device(),
             "dy2st": self.train_config.get("dy2st", False),
+            "amp": self.train_config.get("amp", "OFF"),  # amp support 'O1', 'O2', 'OFF'
         }

paddlex/modules/ts_anomaly_detection/__init__.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -14,6 +14,6 @@
 
 
 from .dataset_checker import TSADDatasetChecker
-from .trainer import TSADTrainer
 from .evaluator import TSADEvaluator
 from .exportor import TSADExportor
+from .trainer import TSADTrainer

paddlex/modules/ts_anomaly_detection/dataset_checker/__init__.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -13,17 +13,17 @@
 # limitations under the License.
 
 
+import json
 import os
 import os.path as osp
-from collections import defaultdict, Counter
+from collections import Counter, defaultdict
+from pathlib import Path
 
 from PIL import Image
-import json
 
 from ...base import BaseDatasetChecker
-from .dataset_src import check, convert, split_dataset, deep_analyse
-
 from ..model_list import MODELS
+from .dataset_src import check, convert, deep_analyse, split_dataset
 
 
 class TSADDatasetChecker(BaseDatasetChecker):
@@ -32,6 +32,20 @@ class TSADDatasetChecker(BaseDatasetChecker):
     entities = MODELS
     sample_num = 10
 
+    def get_dataset_root(self, dataset_dir: str) -> str:
+        """find the dataset root dir
+
+        Args:
+            dataset_dir (str): the directory that contain dataset.
+
+        Returns:
+            str: the root directory of dataset.
+        """
+        anno_dirs = list(Path(dataset_dir).glob("**/train.csv"))
+        assert len(anno_dirs) == 1
+        dataset_dir = anno_dirs[0].parent.as_posix()
+        return dataset_dir
+
     def convert_dataset(self, src_dataset_dir: str) -> str:
         """convert the dataset from other type to specified type
 

paddlex/modules/ts_anomaly_detection/dataset_checker/dataset_src/__init__.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -13,7 +13,7 @@
 # limitations under the License.
 
 
+from .analyse_dataset import deep_analyse
 from .check_dataset import check
 from .convert_dataset import convert
 from .split_dataset import split_dataset
-from .analyse_dataset import deep_analyse

paddlex/modules/ts_anomaly_detection/dataset_checker/dataset_src/analyse_dataset.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -13,14 +13,6 @@
 # limitations under the License.
 
 
-import os
-import platform
-from pathlib import Path
-
-import pandas as pd
-import numpy as np
-
-
 def deep_analyse(dataset_dir, output):
     """class analysis for dataset"""
 

paddlex/modules/ts_anomaly_detection/dataset_checker/dataset_src/check_dataset.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -13,10 +13,10 @@
 # limitations under the License.
 
 
-import os
 import os.path as osp
 from collections import defaultdict
 from pathlib import Path
+
 import pandas as pd
 
 from .....utils.errors import DatasetFileNotFoundError

paddlex/modules/ts_anomaly_detection/dataset_checker/dataset_src/convert_dataset.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -15,12 +15,8 @@
 
 import os
 import os.path as osp
-import shutil
-import json
-import random
-import numpy as np
+
 import pandas as pd
-from tqdm import tqdm
 
 from .....utils.errors import ConvertFailedError
 

paddlex/modules/ts_anomaly_detection/dataset_checker/dataset_src/split_dataset.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -13,13 +13,13 @@
 # limitations under the License.
 
 
+import math
 import os
 import os.path as osp
 import shutil
-import random
-import math
+
 import pandas as pd
-from tqdm import tqdm
+
 from .....utils.logging import info
 
 

paddlex/modules/ts_anomaly_detection/evaluator.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

paddlex/modules/ts_anomaly_detection/exportor.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -12,12 +12,11 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-import tarfile
 from pathlib import Path
 
+from ...utils import logging
 from ..base import BaseExportor
 from .model_list import MODELS
-from ...utils import logging
 
 
 class TSADExportor(BaseExportor):

paddlex/modules/ts_anomaly_detection/model_list.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.

paddlex/modules/ts_anomaly_detection/trainer.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -13,13 +13,11 @@
 # limitations under the License.
 
 import os
-import json
-import time
-from pathlib import Path
 import tarfile
+from pathlib import Path
 
+from ...utils.flags import FLAGS_json_format_model
 from ..base import BaseTrainer
-from ...utils.config import AttrDict
 from .model_list import MODELS
 
 
@@ -34,7 +32,21 @@ class TSADTrainer(BaseTrainer):
         os.makedirs(self.global_config.output, exist_ok=True)
         self.update_config()
         self.dump_config()
-        train_result = self.pdx_model.train(**self.get_train_kwargs())
+        train_args = self.get_train_kwargs()
+        export_with_pir = (
+            self.global_config.get("export_with_pir", False) or FLAGS_json_format_model
+        )
+        train_args.update(
+            {
+                "uniform_output_enabled": self.train_config.get(
+                    "uniform_output_enabled", True
+                ),
+                "export_with_pir": export_with_pir,
+            }
+        )
+        if self.benchmark_config is not None:
+            train_args.update({"benchmark": self.benchmark_config})
+        train_result = self.pdx_model.train(**train_args)
         assert (
             train_result.returncode == 0
         ), f"Encountered an unexpected error({train_result.returncode}) in \
@@ -80,6 +92,8 @@ training!"
             self.pdx_config.update_learning_rate(self.train_config.learning_rate)
         if self.train_config.epochs_iters is not None:
             self.pdx_config.update_epochs(self.train_config.epochs_iters)
+        if self.train_config.get("dy2st", False):
+            self.pdx_config.update_to_static(self.train_config.dy2st)
         if self.train_config.log_interval is not None:
             self.pdx_config.update_log_interval(self.train_config.log_interval)
         if self.global_config.output is not None:
@@ -94,4 +108,6 @@ training!"
         train_args = {"device": self.get_device(using_device_number=1)}
         if self.global_config.output is not None:
             train_args["save_dir"] = self.global_config.output
+        # amp support 'O1', 'O2', 'OFF'
+        train_args["amp"] = self.train_config.get("amp", "OFF")
         return train_args

paddlex/modules/ts_classification/__init__.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -14,6 +14,6 @@
 
 
 from .dataset_checker import TSCLSDatasetChecker
-from .trainer import TSCLSTrainer
 from .evaluator import TSCLSEvaluator
 from .exportor import TSCLSExportor
+from .trainer import TSCLSTrainer

paddlex/modules/ts_classification/dataset_checker/__init__.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -13,17 +13,17 @@
 # limitations under the License.
 
 
+import json
 import os
 import os.path as osp
-from collections import defaultdict, Counter
+from collections import Counter, defaultdict
+from pathlib import Path
 
 from PIL import Image
-import json
 
 from ...base import BaseDatasetChecker
-from .dataset_src import check, convert, split_dataset, deep_analyse
-
 from ..model_list import MODELS
+from .dataset_src import check, convert, deep_analyse, split_dataset
 
 
 class TSCLSDatasetChecker(BaseDatasetChecker):
@@ -32,6 +32,20 @@ class TSCLSDatasetChecker(BaseDatasetChecker):
     entities = MODELS
     sample_num = 10
 
+    def get_dataset_root(self, dataset_dir: str) -> str:
+        """find the dataset root dir
+
+        Args:
+            dataset_dir (str): the directory that contain dataset.
+
+        Returns:
+            str: the root directory of dataset.
+        """
+        anno_dirs = list(Path(dataset_dir).glob("**/train.csv"))
+        assert len(anno_dirs) == 1
+        dataset_dir = anno_dirs[0].parent.as_posix()
+        return dataset_dir
+
     def convert_dataset(self, src_dataset_dir: str) -> str:
         """convert the dataset from other type to specified type
 

paddlex/modules/ts_classification/dataset_checker/dataset_src/__init__.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -13,7 +13,7 @@
 # limitations under the License.
 
 
+from .analyse_dataset import deep_analyse
 from .check_dataset import check
 from .convert_dataset import convert
 from .split_dataset import split_dataset
-from .analyse_dataset import deep_analyse

paddlex/modules/ts_classification/dataset_checker/dataset_src/analyse_dataset.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -15,16 +15,19 @@
 
 import os
 import platform
-from pathlib import Path
 
-import pandas as pd
 import numpy as np
-import matplotlib.pyplot as plt
-from matplotlib import font_manager
+import pandas as pd
 
+from .....utils.deps import function_requires_deps, is_dep_available
 from .....utils.fonts import PINGFANG_FONT_FILE_PATH
 
+if is_dep_available("matplotlib"):
+    import matplotlib.pyplot as plt
+    from matplotlib import font_manager
+
 
+@function_requires_deps("matplotlib")
 def deep_analyse(dataset_dir, output, label_col="label"):
     """class analysis for dataset"""
     tags = ["train", "val"]

paddlex/modules/ts_classification/dataset_checker/dataset_src/check_dataset.py

@@ -1,4 +1,4 @@
-# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve.
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -13,10 +13,10 @@
 # limitations under the License.
 
 
-import os
 import os.path as osp
 from collections import defaultdict
 from pathlib import Path
+
 import pandas as pd
 
 from .....utils.errors import DatasetFileNotFoundError