paddlex 3.0.0rc0__py3-none-any.whl → 3.0.0rc1__py3-none-any.whl

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.
@@ -13,18 +13,17 @@
 # limitations under the License.
 
 
+import json
 import os
 import os.path as osp
+from collections import Counter, defaultdict
 from pathlib import Path
-from collections import defaultdict, Counter
 
 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 .dataset_src import check, convert, deep_analyse, split_dataset
 
 
 class TextRecDatasetChecker(BaseDatasetChecker):

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

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.
@@ -18,10 +18,8 @@ import os.path as osp
 from collections import defaultdict
 
 from PIL import Image, ImageOps
-import json
-import numpy as np
 
-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":

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.

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.

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.

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.
@@ -13,12 +13,10 @@
 # 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
 
 
@@ -103,4 +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.
@@ -13,18 +13,17 @@
 # limitations under the License.
 
 
+import json
 import os
 import os.path as osp
+from collections import Counter, defaultdict
 from pathlib import Path
-from collections import defaultdict, Counter
 
 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):

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
 
 
@@ -35,9 +33,9 @@ class TSADTrainer(BaseTrainer):
         self.update_config()
         self.dump_config()
         train_args = self.get_train_kwargs()
-        export_with_pir = self.global_config.get("export_with_pir", False) or os.getenv(
-            "FLAGS_json_format_model"
-        ) in ["1", "True"]
+        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(
@@ -110,4 +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.
@@ -13,18 +13,17 @@
 # limitations under the License.
 
 
+import json
 import os
 import os.path as osp
+from collections import Counter, defaultdict
 from pathlib import Path
-from collections import defaultdict, Counter
 
 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):

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

paddlex/modules/ts_classification/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.
@@ -14,12 +14,8 @@
 
 
 import os
-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_classification/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_classification/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.