Datasets:
Tasks:
Tabular Classification
Modalities:
Tabular
Formats:
csv
Languages:
English
Size:
10K - 100K
License:
Upload page_blocks.py
Browse files- page_blocks.py +0 -8
page_blocks.py
CHANGED
|
@@ -91,7 +91,6 @@ class PageBlocks(datasets.GeneratorBasedBuilder):
|
|
| 91 |
|
| 92 |
def _generate_examples(self, filepath: str):
|
| 93 |
data = pandas.read_csv(filepath)
|
| 94 |
-
print("preprocessing...\n\n\n\n\n")
|
| 95 |
data = self.preprocess(data)
|
| 96 |
|
| 97 |
for row_id, row in data.iterrows():
|
|
@@ -102,21 +101,14 @@ class PageBlocks(datasets.GeneratorBasedBuilder):
|
|
| 102 |
def preprocess(self, data: pandas.DataFrame) -> pandas.DataFrame:
|
| 103 |
if self.config.name == "page_blocks_binary":
|
| 104 |
data["number_of_transitions"] = data["number_of_transitions"].apply(lambda x: 1 if x > 1 else 0)
|
| 105 |
-
print("mapped\n\n\n\n")
|
| 106 |
data = data.rename(columns={"number_of_transitions": "has_multiple_transitions"})
|
| 107 |
-
print("renamed\n\n\n\n")
|
| 108 |
|
| 109 |
-
print("encoding\n\n\n\n")
|
| 110 |
for feature in _ENCODING_DICS:
|
| 111 |
encoding_function = partial(self.encode, feature)
|
| 112 |
-
print("\tencoding...\n\n\n\n")
|
| 113 |
data.loc[:, feature] = data[feature].apply(encoding_function)
|
| 114 |
|
| 115 |
data = data.reset_index()
|
| 116 |
data.drop("index", axis="columns", inplace=True)
|
| 117 |
-
print("encoded\n\n\n\n")
|
| 118 |
-
print(list(features_types_per_config[self.config.name].keys()))
|
| 119 |
-
print(data.head())
|
| 120 |
|
| 121 |
return data[list(features_types_per_config[self.config.name].keys())]
|
| 122 |
|
|
|
|
| 91 |
|
| 92 |
def _generate_examples(self, filepath: str):
|
| 93 |
data = pandas.read_csv(filepath)
|
|
|
|
| 94 |
data = self.preprocess(data)
|
| 95 |
|
| 96 |
for row_id, row in data.iterrows():
|
|
|
|
| 101 |
def preprocess(self, data: pandas.DataFrame) -> pandas.DataFrame:
|
| 102 |
if self.config.name == "page_blocks_binary":
|
| 103 |
data["number_of_transitions"] = data["number_of_transitions"].apply(lambda x: 1 if x > 1 else 0)
|
|
|
|
| 104 |
data = data.rename(columns={"number_of_transitions": "has_multiple_transitions"})
|
|
|
|
| 105 |
|
|
|
|
| 106 |
for feature in _ENCODING_DICS:
|
| 107 |
encoding_function = partial(self.encode, feature)
|
|
|
|
| 108 |
data.loc[:, feature] = data[feature].apply(encoding_function)
|
| 109 |
|
| 110 |
data = data.reset_index()
|
| 111 |
data.drop("index", axis="columns", inplace=True)
|
|
|
|
|
|
|
|
|
|
| 112 |
|
| 113 |
return data[list(features_types_per_config[self.config.name].keys())]
|
| 114 |
|