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

paddlex/modules/semantic_segmentation/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.
@@ -14,11 +14,8 @@
 
 
 import os
-import glob
-from pathlib import Path
 
 from ..base import BaseTrainer
-from ...utils.config import AttrDict
 from .model_list import MODELS
 
 
@@ -68,4 +65,8 @@ class SegTrainer(BaseTrainer):
             train_args["do_eval"] = True
             train_args["save_interval"] = self.train_config.eval_interval
         train_args["dy2st"] = self.train_config.get("dy2st", False)
+        if self.train_config.get("input_shape") is not None:
+            train_args["input_shape"] = self.train_config.input_shape
+        # amp support 'O1', 'O2', 'OFF'
+        train_args["amp"] = self.train_config.get("amp", "OFF")
         return train_args

paddlex/modules/table_recognition/__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.
@@ -13,6 +13,6 @@
 # limitations under the License.
 
 from .dataset_checker import TableRecDatasetChecker
-from .trainer import TableRecTrainer
 from .evaluator import TableRecEvaluator
 from .exportor import TableRecExportor
+from .trainer import TableRecTrainer

paddlex/modules/table_recognition/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.
@@ -13,16 +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, split_dataset, deep_analyse
-
 from ..model_list import MODELS
+from .dataset_src import check, deep_analyse, split_dataset
 
 
 class TableRecDatasetChecker(BaseDatasetChecker):
@@ -30,6 +31,20 @@ class TableRecDatasetChecker(BaseDatasetChecker):
 
     entities = MODELS
 
+    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.txt"))
+        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
 
@@ -64,7 +79,7 @@ class TableRecDatasetChecker(BaseDatasetChecker):
         Returns:
             dict: dataset summary.
         """
-        return check(dataset_dir, self.global_config.output, sample_num=10)
+        return check(dataset_dir, self.output, sample_num=10)
 
     def get_show_type(self) -> str:
         """get the show type of dataset

paddlex/modules/table_recognition/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.
@@ -13,6 +13,6 @@
 # limitations under the License.
 
 
+from .analyse_dataset import deep_analyse
 from .check_dataset import check
 from .split_dataset import split_dataset
-from .analyse_dataset import deep_analyse

paddlex/modules/table_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.
@@ -15,6 +15,7 @@
 
 import os
 from collections import defaultdict
+
 from .....utils.file_interface import custom_open
 
 
@@ -31,7 +32,7 @@ def simple_analyse(dataset_path):
     """
     tags = ["train", "val", "test"]
     sample_cnts = defaultdict(int)
-    img_paths = defaultdict(list)
+    defaultdict(list)
     res = [None] * 6
 
     for tag in tags:

paddlex/modules/table_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.
@@ -13,11 +13,14 @@
 # limitations under the License.
 
 
-import os
 import json
+import os
 import os.path as osp
 from collections import defaultdict
-from .....utils.errors import DatasetFileNotFoundError, CheckFailedError
+
+from PIL import Image, ImageOps
+
+from .....utils.errors import DatasetFileNotFoundError
 
 
 def check(dataset_dir, output, dataset_type="PubTabTableRecDataset", sample_num=10):
@@ -31,7 +34,6 @@ def check(dataset_dir, output, dataset_type="PubTabTableRecDataset", sample_num=
             raise DatasetFileNotFoundError(file_path=dataset_dir)
 
         tags = ["train", "val"]
-        max_recorded_sample_cnts = 50
         sample_cnts = dict()
         sample_paths = defaultdict(list)
 
@@ -54,28 +56,26 @@ def check(dataset_dir, output, dataset_type="PubTabTableRecDataset", sample_num=
                     for line in all_lines:
                         info = json.loads(line.strip("\n"))
                         file_name = info["filename"]
-                        cells = info["html"]["cells"].copy()
-                        structure = info["html"]["structure"]["tokens"].copy()
+                        info["html"]["cells"].copy()
+                        info["html"]["structure"]["tokens"].copy()
 
                         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)
-
-                        boxes_num = len(cells)
-                        tokens_num = sum(
-                            [
-                                structure.count(x)
-                                for x in ["<td>", "<td", "<eb></eb>", "<td></td>"]
-                            ]
-                        )
-                        if boxes_num != tokens_num:
-                            raise CheckFailedError(
-                                f"The number of cells needs to be consistent with the number of tokens "
-                                "but the number of cells is {boxes_num}, and the number of tokens is {tokens_num}."
+                        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/table_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.
@@ -15,6 +15,7 @@
 
 import os
 from random import shuffle
+
 from .....utils.file_interface import custom_open
 
 

paddlex/modules/table_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.

paddlex/modules/table_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.

paddlex/modules/table_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.
@@ -16,4 +16,6 @@
 MODELS = [
     "SLANet",
     "SLANet_plus",
+    "SLANeXt_wired",
+    "SLANeXt_wireless",
 ]

paddlex/modules/table_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.
@@ -13,11 +13,7 @@
 # limitations under the License.
 
 
-import os
-from pathlib import Path
-
 from ..base import BaseTrainer
-from ...utils.config import AttrDict
 from .model_list import MODELS
 
 
@@ -67,4 +63,5 @@ class TableRecTrainer(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/text_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.
@@ -13,6 +13,6 @@
 # limitations under the License.
 
 from .dataset_checker import TextDetDatasetChecker
-from .trainer import TextDetTrainer
 from .evaluator import TextDetEvaluator
 from .exportor import TextDetExportor
+from .trainer import TextDetTrainer

paddlex/modules/text_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.
@@ -13,16 +13,15 @@
 # limitations under the License.
 
 
+import json
 import os
 import os.path as osp
-from collections import defaultdict, Counter
-
-import json
+from collections import Counter, defaultdict
+from pathlib import Path
 
 from ...base import BaseDatasetChecker
-from .dataset_src import check, split_dataset, deep_analyse
-
 from ..model_list import MODELS
+from .dataset_src import check, deep_analyse, split_dataset
 
 
 class TextDetDatasetChecker(BaseDatasetChecker):
@@ -30,6 +29,20 @@ class TextDetDatasetChecker(BaseDatasetChecker):
 
     entities = MODELS
 
+    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("**/images"))
+        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
 
@@ -64,7 +77,7 @@ class TextDetDatasetChecker(BaseDatasetChecker):
         Returns:
             dict: dataset summary.
         """
-        return check(dataset_dir, self.global_config.output, sample_num=10)
+        return check(dataset_dir, self.output, sample_num=10)
 
     def analyse(self, dataset_dir: str) -> dict:
         """deep analyse dataset

paddlex/modules/text_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.
@@ -13,6 +13,6 @@
 # limitations under the License.
 
 
+from .analyse_dataset import deep_analyse
 from .check_dataset import check
 from .split_dataset import split_dataset
-from .analyse_dataset import deep_analyse

paddlex/modules/text_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.
@@ -13,20 +13,22 @@
 # limitations under the License.
 
 
-from cProfile import label
+import json
 import os
-
 from collections import defaultdict
 
-import matplotlib.pyplot as plt
-from matplotlib.backends.backend_agg import FigureCanvasAgg
 import numpy as np
 from PIL import Image, ImageOps
-import cv2
-import json
 
+from .....utils.deps import function_requires_deps, is_dep_available
 from .....utils.file_interface import custom_open
 
+if is_dep_available("opencv-contrib-python"):
+    import cv2
+if is_dep_available("matplotlib"):
+    import matplotlib.pyplot as plt
+    from matplotlib.backends.backend_agg import FigureCanvasAgg
+
 
 # show data samples
 def simple_analyse(dataset_path, max_recorded_sample_cnts=20, show_label=True):
@@ -49,7 +51,6 @@ def simple_analyse(dataset_path, max_recorded_sample_cnts=20, show_label=True):
     lab_infos = defaultdict(list)
     res = [None] * 9
     delim = "\t"
-    valid_num_parts = 2
 
     for tag in tags:
         file_list = os.path.join(dataset_path, f"{tag}.txt")
@@ -142,6 +143,7 @@ def simple_analyse(dataset_path, max_recorded_sample_cnts=20, show_label=True):
         )
 
 
+@function_requires_deps("opencv-contrib-python")
 def show_label_img(img_path, dt_boxes):
     """draw ocr detection label"""
     img = cv2.imread(img_path)
@@ -151,13 +153,14 @@ def show_label_img(img_path, dt_boxes):
     return img[:, :, ::-1]
 
 
+@function_requires_deps("matplotlib", "opencv-contrib-python")
 def deep_analyse(dataset_path, output):
     """class analysis for dataset"""
     sample_results = simple_analyse(
         dataset_path, max_recorded_sample_cnts=float("inf"), show_label=False
     )
     lab_infos = sample_results[-3] + sample_results[-2] + sample_results[-1]
-    labels_cnt = defaultdict(int)
+    defaultdict(int)
     img_shapes = []  # w, h
     ratios_w = []
     ratios_h = []

paddlex/modules/text_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.
@@ -13,13 +13,13 @@
 # limitations under the License.
 
 
+import json
 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
 
@@ -66,10 +66,20 @@ def check(dataset_dir, output, sample_num=10):
                     file_name = substr[0]
                     label = substr[1]
                     img_path = osp.join(dataset_dir, file_name)
-                    if len(sample_paths[tag]) < sample_num:
-                        sample_paths[tag].append(os.path.relpath(img_path, output))
                     if not osp.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)
 
                     # check det label
                     label = json.loads(label)

paddlex/modules/text_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.
@@ -15,9 +15,10 @@
 
 import glob
 import os.path
-import numpy as np
 import shutil
 
+import numpy as np
+
 from .....utils.file_interface import custom_open
 from .....utils.logging import info
 
@@ -71,7 +72,6 @@ def split_dataset(dataset_root, train_rate, val_rate):
     start = 0
     rate_list = [train_rate, val_rate]
     name_list = ["train", "val"]
-    separator = " "
     for i, name in enumerate(name_list):
         info("Creating {}.txt...".format(name))
 

paddlex/modules/text_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/text_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.

paddlex/modules/text_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.
@@ -17,6 +17,8 @@ MODELS = [
     "PP-OCRv4_server_det",
     "PP-OCRv4_mobile_seal_det",
     "PP-OCRv4_server_seal_det",
+    "PP-OCRv3_mobile_det",
+    "PP-OCRv3_server_det",
 ]
 
 CURVE_MODELS = ["PP-OCRv4_mobile_seal_det", "PP-OCRv4_server_seal_det"]

paddlex/modules/text_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.
@@ -13,11 +13,7 @@
 # limitations under the License.
 
 
-import os
-from pathlib import Path
-
 from ..base import BaseTrainer
-from ...utils.config import AttrDict
 from .model_list import MODELS
 
 
@@ -65,4 +61,5 @@ class TextDetTrainer(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/text_recognition/__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.
@@ -13,6 +13,6 @@
 # limitations under the License.
 
 from .dataset_checker import TextRecDatasetChecker
-from .trainer import TextRecTrainer
 from .evaluator import TextRecEvaluator
 from .exportor import TextRecExportor
+from .trainer import TextRecTrainer

paddlex/modules/text_recognition/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.
@@ -13,20 +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, split_dataset, deep_analyse, convert
-
 from ..model_list import MODELS
-from ...formula_recognition.model_list import MODELS as MODELS_LaTeX
-
-MODELS = MODELS + MODELS_LaTeX
+from .dataset_src import check, convert, deep_analyse, split_dataset
 
 
 class TextRecDatasetChecker(BaseDatasetChecker):
@@ -35,6 +32,20 @@ class TextRecDatasetChecker(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.txt"))
+        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
 
@@ -74,7 +85,7 @@ class TextRecDatasetChecker(BaseDatasetChecker):
         """
         return check(
             dataset_dir,
-            self.global_config.output,
+            self.output,
             sample_num=10,
             dataset_type=self.get_dataset_type(),
         )

paddlex/modules/text_recognition/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