pierreguillou commited on
Commit
4d0cb9d
1 Parent(s): 048e3df

Update DocLayNet-large.py

Browse files
Files changed (1) hide show
  1. DocLayNet-large.py +23 -27
DocLayNet-large.py CHANGED
@@ -72,17 +72,17 @@ def load_image(image_path):
72
  logger = datasets.logging.get_logger(__name__)
73
 
74
 
75
- class DocLayNetConfig(datasets.BuilderConfig):
76
- """BuilderConfig for DocLayNet large"""
77
 
78
- def __init__(self, **kwargs):
79
  """BuilderConfig for DocLayNet large.
80
  Args:
81
  **kwargs: keyword arguments forwarded to super.
82
  """
83
- super(DocLayNetConfig, self).__init__(**kwargs)
84
-
85
 
 
86
  class DocLayNet(datasets.GeneratorBasedBuilder):
87
  """
88
  DocLayNet large is a about 99% of the dataset DocLayNet (more information at https://huggingface.co/datasets/pierreguillou/DocLayNet-large)
@@ -105,12 +105,14 @@ class DocLayNet(datasets.GeneratorBasedBuilder):
105
  # You will be able to load one or the other configurations in the following list with
106
  # data = datasets.load_dataset('my_dataset', 'first_domain')
107
  # data = datasets.load_dataset('my_dataset', 'second_domain')
 
 
108
  BUILDER_CONFIGS = [
109
- DocLayNetConfig(name="DocLayNet", version=datasets.Version("1.0.0"), description="DocLayNeT large dataset"),
110
  ]
111
 
112
- #DEFAULT_CONFIG_NAME = "DocLayNet" # It's not mandatory to have a default configuration. Just use one if it make sense.
113
-
114
  def _info(self):
115
 
116
  features = datasets.Features(
@@ -164,10 +166,9 @@ class DocLayNet(datasets.GeneratorBasedBuilder):
164
  # By default the archives will be extracted and a path to a cached folder where they are extracted is returned instead of the archive
165
 
166
  archive_path = dl_manager.download_and_extract(_URLs)
167
- splits = []
168
- for split in self.config.splits:
169
- if split == "train":
170
- dataset = datasets.SplitGenerator(
171
  name=datasets.Split.TRAIN,
172
  # These kwargs will be passed to _generate_examples
173
  gen_kwargs={
@@ -175,11 +176,10 @@ class DocLayNet(datasets.GeneratorBasedBuilder):
175
  "filepath_1": os.path.join(archive_path["part_dataset_1"], "part_dataset_1/train/"),
176
  "filepath_2": os.path.join(archive_path["part_dataset_2"], "part_dataset_2/train/"),
177
  "filepath_3": os.path.join(archive_path["part_dataset_3"], "part_dataset_3/train/"),
178
- "split": "train",
179
  },
180
- )
181
- elif split in ["val", "valid", "validation", "dev"]:
182
- dataset = datasets.SplitGenerator(
183
  name=datasets.Split.VALIDATION,
184
  # These kwargs will be passed to _generate_examples
185
  gen_kwargs={
@@ -187,11 +187,10 @@ class DocLayNet(datasets.GeneratorBasedBuilder):
187
  "filepath_1": os.path.join(archive_path["part_dataset_1"], "part_dataset_1/val/"),
188
  "filepath_2": os.path.join(archive_path["part_dataset_2"], "part_dataset_2/val/"),
189
  "filepath_3": os.path.join(archive_path["part_dataset_3"], "part_dataset_3/val/"),
190
- "split": "val",
191
  },
192
- )
193
- elif split in ["test"]:
194
- dataset = datasets.SplitGenerator(
195
  name=datasets.Split.TEST,
196
  # These kwargs will be passed to _generate_examples
197
  gen_kwargs={
@@ -199,16 +198,13 @@ class DocLayNet(datasets.GeneratorBasedBuilder):
199
  "filepath_1": os.path.join(archive_path["part_dataset_1"], "part_dataset_1/test/"),
200
  "filepath_2": os.path.join(archive_path["part_dataset_2"], "part_dataset_2/test/"),
201
  "filepath_3": os.path.join(archive_path["part_dataset_3"], "part_dataset_3/test/"),
202
- "split": "test",
203
  },
204
- )
205
- else:
206
- continue
207
 
208
- splits.append(dataset)
209
- return splits
210
 
211
- def _generate_examples(self, filepath_0, filepath_1, filepath_2, filepath_3, split):
212
  filepath = (filepath_0, filepath_1, filepath_2, filepath_3)
213
  logger.info("⏳ Generating examples from = %s", filepath)
214
  ann_dirs = [os.path.join(filepath_0, "annotations"), os.path.join(filepath_1, "annotations"), os.path.join(filepath_2, "annotations"), os.path.join(filepath_3, "annotations")]
 
72
  logger = datasets.logging.get_logger(__name__)
73
 
74
 
75
+ class DocLayNetBuilderConfig(datasets.BuilderConfig):
76
+ """BuilderConfig for DocLayNet base"""
77
 
78
+ def __init__(self, name, **kwargs):
79
  """BuilderConfig for DocLayNet large.
80
  Args:
81
  **kwargs: keyword arguments forwarded to super.
82
  """
83
+ super().__init__(name, **kwargs)
 
84
 
85
+
86
  class DocLayNet(datasets.GeneratorBasedBuilder):
87
  """
88
  DocLayNet large is a about 99% of the dataset DocLayNet (more information at https://huggingface.co/datasets/pierreguillou/DocLayNet-large)
 
105
  # You will be able to load one or the other configurations in the following list with
106
  # data = datasets.load_dataset('my_dataset', 'first_domain')
107
  # data = datasets.load_dataset('my_dataset', 'second_domain')
108
+ DEFAULT_CONFIG_NAME = "DocLayNet_2022.08_processed_on_2023.01" # It's not mandatory to have a default configuration. Just use one if it make sense.
109
+
110
  BUILDER_CONFIGS = [
111
+ DocLayNetConfig(name=DEFAULT_CONFIG_NAME, version=VERSION, description="DocLayNet large dataset"),
112
  ]
113
 
114
+ BUILDER_CONFIG_CLASS = DocLayNetBuilderConfig
115
+
116
  def _info(self):
117
 
118
  features = datasets.Features(
 
166
  # By default the archives will be extracted and a path to a cached folder where they are extracted is returned instead of the archive
167
 
168
  archive_path = dl_manager.download_and_extract(_URLs)
169
+
170
+ return [
171
+ datasets.SplitGenerator(
 
172
  name=datasets.Split.TRAIN,
173
  # These kwargs will be passed to _generate_examples
174
  gen_kwargs={
 
176
  "filepath_1": os.path.join(archive_path["part_dataset_1"], "part_dataset_1/train/"),
177
  "filepath_2": os.path.join(archive_path["part_dataset_2"], "part_dataset_2/train/"),
178
  "filepath_3": os.path.join(archive_path["part_dataset_3"], "part_dataset_3/train/"),
179
+ # "split_key": "train",
180
  },
181
+ ),
182
+ datasets.SplitGenerator(
 
183
  name=datasets.Split.VALIDATION,
184
  # These kwargs will be passed to _generate_examples
185
  gen_kwargs={
 
187
  "filepath_1": os.path.join(archive_path["part_dataset_1"], "part_dataset_1/val/"),
188
  "filepath_2": os.path.join(archive_path["part_dataset_2"], "part_dataset_2/val/"),
189
  "filepath_3": os.path.join(archive_path["part_dataset_3"], "part_dataset_3/val/"),
190
+ # "split_key": "validation",
191
  },
192
+ ),
193
+ datasets.SplitGenerator(
 
194
  name=datasets.Split.TEST,
195
  # These kwargs will be passed to _generate_examples
196
  gen_kwargs={
 
198
  "filepath_1": os.path.join(archive_path["part_dataset_1"], "part_dataset_1/test/"),
199
  "filepath_2": os.path.join(archive_path["part_dataset_2"], "part_dataset_2/test/"),
200
  "filepath_3": os.path.join(archive_path["part_dataset_3"], "part_dataset_3/test/"),
201
+ # "split_key": "test"
202
  },
203
+ ),
204
+ ]
 
205
 
 
 
206
 
207
+ def _generate_examples(self, filepath_0, filepath_1, filepath_2, filepath_3):
208
  filepath = (filepath_0, filepath_1, filepath_2, filepath_3)
209
  logger.info("⏳ Generating examples from = %s", filepath)
210
  ann_dirs = [os.path.join(filepath_0, "annotations"), os.path.join(filepath_1, "annotations"), os.path.join(filepath_2, "annotations"), os.path.join(filepath_3, "annotations")]