Update SciGraph.py
Browse files- SciGraph.py +18 -5
SciGraph.py
CHANGED
@@ -33,6 +33,12 @@ _HOMEPAGE = ""
|
|
33 |
# The license information was obtained from https://github.com/boudinfl/ake-datasets as the dataset shared over here is taken from here
|
34 |
_LICENSE = ""
|
35 |
|
|
|
|
|
|
|
|
|
|
|
|
|
36 |
# TODO: Add link to the official dataset URLs here
|
37 |
|
38 |
|
@@ -84,6 +90,7 @@ class SciGraph(datasets.GeneratorBasedBuilder):
|
|
84 |
)
|
85 |
|
86 |
def _split_generators(self, dl_manager):
|
|
|
87 |
|
88 |
return [
|
89 |
datasets.SplitGenerator(
|
@@ -91,23 +98,29 @@ class SciGraph(datasets.GeneratorBasedBuilder):
|
|
91 |
# These kwargs will be passed to _generate_examples
|
92 |
gen_kwargs={
|
93 |
"split": "train",
|
|
|
|
|
|
|
94 |
},
|
95 |
),
|
96 |
datasets.SplitGenerator(
|
97 |
name=datasets.Split.TEST,
|
98 |
# These kwargs will be passed to _generate_examples
|
99 |
gen_kwargs={
|
100 |
-
"split": "test"
|
|
|
|
|
|
|
101 |
},
|
102 |
)
|
103 |
]
|
104 |
|
105 |
# method parameters are unpacked from `gen_kwargs` as given in `_split_generators`
|
106 |
-
def _generate_examples(self, split):
|
107 |
if self.config.name == 'function':
|
108 |
-
with open(
|
109 |
functions = list(json.load(f).keys())
|
110 |
-
data = pd.read_json(
|
111 |
|
112 |
if split == 'train':
|
113 |
data = data.loc[data[functions].sum(axis=1) == 1]
|
@@ -130,7 +143,7 @@ class SciGraph(datasets.GeneratorBasedBuilder):
|
|
130 |
}
|
131 |
|
132 |
if self.config.name == 'topic':
|
133 |
-
data = pd.read_json(
|
134 |
data = data.replace(to_replace=r'^\s*$', value=np.nan, regex=True).dropna(subset=['keywords'], axis=0)
|
135 |
|
136 |
train_data, test_data = train_test_split(data, test_size=0.1, random_state=42)
|
|
|
33 |
# The license information was obtained from https://github.com/boudinfl/ake-datasets as the dataset shared over here is taken from here
|
34 |
_LICENSE = ""
|
35 |
|
36 |
+
_URLS = {
|
37 |
+
'classes': 'class.json',
|
38 |
+
'function': 'assign.jsonl',
|
39 |
+
'topic': 'paper_new.jsonl'
|
40 |
+
}
|
41 |
+
|
42 |
# TODO: Add link to the official dataset URLs here
|
43 |
|
44 |
|
|
|
90 |
)
|
91 |
|
92 |
def _split_generators(self, dl_manager):
|
93 |
+
data_dir = dl_manager.download_and_extract(_URLS)
|
94 |
|
95 |
return [
|
96 |
datasets.SplitGenerator(
|
|
|
98 |
# These kwargs will be passed to _generate_examples
|
99 |
gen_kwargs={
|
100 |
"split": "train",
|
101 |
+
"classes": data_dir['classes'],
|
102 |
+
"function": data_dir['function'],
|
103 |
+
"topic": data_dir['topic']
|
104 |
},
|
105 |
),
|
106 |
datasets.SplitGenerator(
|
107 |
name=datasets.Split.TEST,
|
108 |
# These kwargs will be passed to _generate_examples
|
109 |
gen_kwargs={
|
110 |
+
"split": "test",
|
111 |
+
"classes": data_dir['classes'],
|
112 |
+
"function": data_dir['function'],
|
113 |
+
"topic": data_dir['topic']
|
114 |
},
|
115 |
)
|
116 |
]
|
117 |
|
118 |
# method parameters are unpacked from `gen_kwargs` as given in `_split_generators`
|
119 |
+
def _generate_examples(self, split, classes, function, topic):
|
120 |
if self.config.name == 'function':
|
121 |
+
with open(classes, 'r') as f:
|
122 |
functions = list(json.load(f).keys())
|
123 |
+
data = pd.read_json(function)
|
124 |
|
125 |
if split == 'train':
|
126 |
data = data.loc[data[functions].sum(axis=1) == 1]
|
|
|
143 |
}
|
144 |
|
145 |
if self.config.name == 'topic':
|
146 |
+
data = pd.read_json(topic)
|
147 |
data = data.replace(to_replace=r'^\s*$', value=np.nan, regex=True).dropna(subset=['keywords'], axis=0)
|
148 |
|
149 |
train_data, test_data = train_test_split(data, test_size=0.1, random_state=42)
|