Upload odinsynth_dataset.py
Browse files- odinsynth_dataset.py +8 -5
odinsynth_dataset.py
CHANGED
@@ -91,14 +91,17 @@ class OdinsynthDatasetBuilder(datasets.GeneratorBasedBuilder):
|
|
91 |
# If several configurations are possible (listed in BUILDER_CONFIGS), the configuration selected by the user is in self.config.name
|
92 |
|
93 |
JSON_PATH = dl_manager.download_and_extract('merged_train_split_train.jsonl.gz')
|
94 |
-
|
|
|
|
|
|
|
95 |
specs = self._build_specs(JSON_PATH)
|
96 |
return [
|
97 |
datasets.SplitGenerator(
|
98 |
name=datasets.Split.TRAIN,
|
99 |
# These kwargs will be passed to _generate_examples
|
100 |
gen_kwargs={
|
101 |
-
"archive_iter": dl_manager.iter_archive(
|
102 |
"specs": specs,
|
103 |
"split": "train",
|
104 |
},
|
@@ -107,7 +110,7 @@ class OdinsynthDatasetBuilder(datasets.GeneratorBasedBuilder):
|
|
107 |
name=datasets.Split.TEST,
|
108 |
# These kwargs will be passed to _generate_examples
|
109 |
gen_kwargs={
|
110 |
-
"archive_iter": dl_manager.iter_archive(
|
111 |
"specs": specs,
|
112 |
"split": "test",
|
113 |
},
|
@@ -116,7 +119,7 @@ class OdinsynthDatasetBuilder(datasets.GeneratorBasedBuilder):
|
|
116 |
name=datasets.Split.VALIDATION,
|
117 |
# These kwargs will be passed to _generate_examples
|
118 |
gen_kwargs={
|
119 |
-
"archive_iter": dl_manager.iter_archive(
|
120 |
"specs": specs,
|
121 |
"split": "val",
|
122 |
},
|
@@ -129,7 +132,7 @@ class OdinsynthDatasetBuilder(datasets.GeneratorBasedBuilder):
|
|
129 |
|
130 |
key = 0
|
131 |
for tsv_path, file in archive_iter:
|
132 |
-
if tsv_path.
|
133 |
# Read the lines
|
134 |
reader = csv.reader((l.decode() for l in file), delimiter='\t')
|
135 |
for row in reader:
|
|
|
91 |
# If several configurations are possible (listed in BUILDER_CONFIGS), the configuration selected by the user is in self.config.name
|
92 |
|
93 |
JSON_PATH = dl_manager.download_and_extract('merged_train_split_train.jsonl.gz')
|
94 |
+
TRAIN_ARCHIVE_PATH = dl_manager.download('train.tar.bz2')
|
95 |
+
VAL_ARCHIVE_PATH = dl_manager.download('val.tar.bz2')
|
96 |
+
TEST_ARCHIVE_PATH = dl_manager.download('test.tar.bz2')
|
97 |
+
|
98 |
specs = self._build_specs(JSON_PATH)
|
99 |
return [
|
100 |
datasets.SplitGenerator(
|
101 |
name=datasets.Split.TRAIN,
|
102 |
# These kwargs will be passed to _generate_examples
|
103 |
gen_kwargs={
|
104 |
+
"archive_iter": dl_manager.iter_archive(TRAIN_ARCHIVE_PATH),
|
105 |
"specs": specs,
|
106 |
"split": "train",
|
107 |
},
|
|
|
110 |
name=datasets.Split.TEST,
|
111 |
# These kwargs will be passed to _generate_examples
|
112 |
gen_kwargs={
|
113 |
+
"archive_iter": dl_manager.iter_archive(TEST_ARCHIVE_PATH),
|
114 |
"specs": specs,
|
115 |
"split": "test",
|
116 |
},
|
|
|
119 |
name=datasets.Split.VALIDATION,
|
120 |
# These kwargs will be passed to _generate_examples
|
121 |
gen_kwargs={
|
122 |
+
"archive_iter": dl_manager.iter_archive(VAL_ARCHIVE_PATH),
|
123 |
"specs": specs,
|
124 |
"split": "val",
|
125 |
},
|
|
|
132 |
|
133 |
key = 0
|
134 |
for tsv_path, file in archive_iter:
|
135 |
+
if tsv_path.endswith(".tsv"):
|
136 |
# Read the lines
|
137 |
reader = csv.reader((l.decode() for l in file), delimiter='\t')
|
138 |
for row in reader:
|