Changes in loading script
Browse files- README.md +10 -3
- court_view_generation.py +5 -14
README.md
CHANGED
@@ -17,21 +17,28 @@ size_categories:
|
|
17 |
## Introduction
|
18 |
|
19 |
This dataset contains court views for different languages and court chambers. It includes information such as decision id, language, chamber, file name, url, and the number of tokens in the facts and considerations sections.
|
20 |
-
|
21 |
|
22 |
## Size
|
23 |
-
###
|
24 |
- All: 385'840
|
25 |
- validation: 47'661
|
26 |
- test: 106'608
|
27 |
- train: 231'571
|
28 |
|
29 |
-
###
|
30 |
- All: 272
|
31 |
- validation: 9
|
32 |
- test: 236
|
33 |
- train: 27
|
34 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
35 |
## Columns
|
36 |
|
37 |
- decision_id: unique identifier for the decision
|
|
|
17 |
## Introduction
|
18 |
|
19 |
This dataset contains court views for different languages and court chambers. It includes information such as decision id, language, chamber, file name, url, and the number of tokens in the facts and considerations sections.
|
20 |
+
Full (L1) contains all the data, Origin (L2) contains only data with complete origin facts & origin considerations.
|
21 |
|
22 |
## Size
|
23 |
+
### Full (L1)
|
24 |
- All: 385'840
|
25 |
- validation: 47'661
|
26 |
- test: 106'608
|
27 |
- train: 231'571
|
28 |
|
29 |
+
### Origin (L2)
|
30 |
- All: 272
|
31 |
- validation: 9
|
32 |
- test: 236
|
33 |
- train: 27
|
34 |
|
35 |
+
## Load datasets
|
36 |
+
|
37 |
+
```python
|
38 |
+
dataset = load_dataset("rcds/court_view_generation")
|
39 |
+
dataset_origin = load_dataset("rcds/court_view_generation", "origin")
|
40 |
+
```
|
41 |
+
|
42 |
## Columns
|
43 |
|
44 |
- decision_id: unique identifier for the decision
|
court_view_generation.py
CHANGED
@@ -53,7 +53,7 @@ _LICENSE = ""
|
|
53 |
# This can be an arbitrary nested dict/list of URLs (see below in `_split_generators` method)
|
54 |
_URLS = {
|
55 |
"full": "https://huggingface.co/datasets/rcds/court_view_generation/resolve/main/L1/huggingface",
|
56 |
-
"
|
57 |
}
|
58 |
|
59 |
|
@@ -75,7 +75,7 @@ class CourtViewGeneration(datasets.GeneratorBasedBuilder):
|
|
75 |
# data = datasets.load_dataset('my_dataset', 'second_domain')
|
76 |
BUILDER_CONFIGS = [
|
77 |
datasets.BuilderConfig(name="full", version=VERSION, description="This part of my dataset covers the whole dataset"),
|
78 |
-
datasets.BuilderConfig(name="
|
79 |
]
|
80 |
|
81 |
DEFAULT_CONFIG_NAME = "full" # It's not mandatory to have a default configuration. Just use one if it make sense.
|
@@ -87,9 +87,7 @@ class CourtViewGeneration(datasets.GeneratorBasedBuilder):
|
|
87 |
"decision_id": datasets.Value("string"),
|
88 |
"facts": datasets.Value("string"),
|
89 |
"considerations": datasets.Value("string"),
|
90 |
-
"court": datasets.Value("string")
|
91 |
-
"origin_facts": datasets.Value("string"),
|
92 |
-
"origin_considerations": datasets.Value("string")
|
93 |
# These are the features of your dataset like images, labels ...
|
94 |
}
|
95 |
)
|
@@ -132,10 +130,6 @@ class CourtViewGeneration(datasets.GeneratorBasedBuilder):
|
|
132 |
filepath_validation = dl_manager.download(os.path.join(urls, "validation.jsonl.xz"))
|
133 |
filepath_test = dl_manager.download(os.path.join(urls, "test.jsonl.xz"))
|
134 |
|
135 |
-
print("filepath_train", filepath_train)
|
136 |
-
# show files in the downloaded and extracted dir
|
137 |
-
print(os.listdir(os.path.dirname(filepath_train)))
|
138 |
-
|
139 |
return [
|
140 |
datasets.SplitGenerator(
|
141 |
name=datasets.Split.TRAIN,
|
@@ -160,7 +154,7 @@ class CourtViewGeneration(datasets.GeneratorBasedBuilder):
|
|
160 |
"filepath": filepath_test,
|
161 |
"split": "test"
|
162 |
},
|
163 |
-
)
|
164 |
]
|
165 |
|
166 |
# method parameters are unpacked from `gen_kwargs` as given in `_split_generators`
|
@@ -171,17 +165,14 @@ class CourtViewGeneration(datasets.GeneratorBasedBuilder):
|
|
171 |
with xz.open(open(filepath, "rb"), "rt", encoding="utf-8") as f:
|
172 |
for id, line in enumerate(f):
|
173 |
line_counter += 1
|
174 |
-
print(id, len(line), line[:200])
|
175 |
if line:
|
176 |
data = json.loads(line)
|
177 |
-
if self.config.name == "
|
178 |
yield id, {
|
179 |
"decision_id": data["decision_id"],
|
180 |
"facts": data["facts"],
|
181 |
"considerations": data["considerations"],
|
182 |
"court": data["court"],
|
183 |
-
"origin_facts": data["origin_facts"],
|
184 |
-
"origin_considerations": data["origin_considerations"]
|
185 |
}
|
186 |
else:
|
187 |
yield id, {
|
|
|
53 |
# This can be an arbitrary nested dict/list of URLs (see below in `_split_generators` method)
|
54 |
_URLS = {
|
55 |
"full": "https://huggingface.co/datasets/rcds/court_view_generation/resolve/main/L1/huggingface",
|
56 |
+
"origin": "https://huggingface.co/datasets/rcds/court_view_generation/resolve/main/L2/huggingface"
|
57 |
}
|
58 |
|
59 |
|
|
|
75 |
# data = datasets.load_dataset('my_dataset', 'second_domain')
|
76 |
BUILDER_CONFIGS = [
|
77 |
datasets.BuilderConfig(name="full", version=VERSION, description="This part of my dataset covers the whole dataset"),
|
78 |
+
datasets.BuilderConfig(name="origin", version=VERSION, description="This part of my dataset covers a subset containing only cases with origin data")
|
79 |
]
|
80 |
|
81 |
DEFAULT_CONFIG_NAME = "full" # It's not mandatory to have a default configuration. Just use one if it make sense.
|
|
|
87 |
"decision_id": datasets.Value("string"),
|
88 |
"facts": datasets.Value("string"),
|
89 |
"considerations": datasets.Value("string"),
|
90 |
+
"court": datasets.Value("string")
|
|
|
|
|
91 |
# These are the features of your dataset like images, labels ...
|
92 |
}
|
93 |
)
|
|
|
130 |
filepath_validation = dl_manager.download(os.path.join(urls, "validation.jsonl.xz"))
|
131 |
filepath_test = dl_manager.download(os.path.join(urls, "test.jsonl.xz"))
|
132 |
|
|
|
|
|
|
|
|
|
133 |
return [
|
134 |
datasets.SplitGenerator(
|
135 |
name=datasets.Split.TRAIN,
|
|
|
154 |
"filepath": filepath_test,
|
155 |
"split": "test"
|
156 |
},
|
157 |
+
)
|
158 |
]
|
159 |
|
160 |
# method parameters are unpacked from `gen_kwargs` as given in `_split_generators`
|
|
|
165 |
with xz.open(open(filepath, "rb"), "rt", encoding="utf-8") as f:
|
166 |
for id, line in enumerate(f):
|
167 |
line_counter += 1
|
|
|
168 |
if line:
|
169 |
data = json.loads(line)
|
170 |
+
if self.config.name == "full":
|
171 |
yield id, {
|
172 |
"decision_id": data["decision_id"],
|
173 |
"facts": data["facts"],
|
174 |
"considerations": data["considerations"],
|
175 |
"court": data["court"],
|
|
|
|
|
176 |
}
|
177 |
else:
|
178 |
yield id, {
|