Datasets:
Delete sonar.py
Browse files
sonar.py
DELETED
@@ -1,72 +0,0 @@
|
|
1 |
-
from typing import List
|
2 |
-
|
3 |
-
import datasets
|
4 |
-
|
5 |
-
import pandas
|
6 |
-
import gzip
|
7 |
-
|
8 |
-
|
9 |
-
VERSION = datasets.Version("1.0.0")
|
10 |
-
|
11 |
-
|
12 |
-
DESCRIPTION = "Sonar dataset from the UCI ML repository."
|
13 |
-
_HOMEPAGE = "https://archive-beta.ics.uci.edu/dataset/31/sonar"
|
14 |
-
_URLS = ("https://archive-beta.ics.uci.edu/dataset/31/sonar")
|
15 |
-
_CITATION = """"""
|
16 |
-
|
17 |
-
# Dataset info
|
18 |
-
urls_per_split = {
|
19 |
-
"train": "https://huggingface.co/datasets/mstz/sonar/raw/main/sonar.all-data"
|
20 |
-
}
|
21 |
-
features_types_per_config = {
|
22 |
-
"sonar": {str(i): datasets.Value("float32") for i in range(60)}
|
23 |
-
}
|
24 |
-
features_types_per_config["sonar"]["is_rock"] = datasets.ClassLabel(num_classes=2)
|
25 |
-
features_per_config = {k: datasets.Features(features_types_per_config[k]) for k in features_types_per_config}
|
26 |
-
|
27 |
-
|
28 |
-
class SonarConfig(datasets.BuilderConfig):
|
29 |
-
def __init__(self, **kwargs):
|
30 |
-
super(SonarConfig, self).__init__(version=VERSION, **kwargs)
|
31 |
-
self.features = features_per_config[kwargs["name"]]
|
32 |
-
|
33 |
-
|
34 |
-
class Sonar(datasets.GeneratorBasedBuilder):
|
35 |
-
# dataset versions
|
36 |
-
DEFAULT_CONFIG = "sonar"
|
37 |
-
BUILDER_CONFIGS = [
|
38 |
-
SonarConfig(name="sonar",
|
39 |
-
description="Sonar for binary classification.")
|
40 |
-
]
|
41 |
-
|
42 |
-
|
43 |
-
def _info(self):
|
44 |
-
if self.config.name not in features_per_config:
|
45 |
-
raise ValueError(f"Unknown configuration: {self.config.name}")
|
46 |
-
|
47 |
-
info = datasets.DatasetInfo(description=DESCRIPTION, citation=_CITATION, homepage=_HOMEPAGE,
|
48 |
-
features=features_per_config[self.config.name])
|
49 |
-
|
50 |
-
return info
|
51 |
-
|
52 |
-
def _split_generators(self, dl_manager: datasets.DownloadManager) -> List[datasets.SplitGenerator]:
|
53 |
-
downloads = dl_manager.download_and_extract(urls_per_split)
|
54 |
-
|
55 |
-
return [
|
56 |
-
datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloads["train"]})
|
57 |
-
]
|
58 |
-
|
59 |
-
def _generate_examples(self, filepath: str):
|
60 |
-
data = pandas.read_csv(filepath, header=None)
|
61 |
-
data.columns = [str(i) for i in range(60)] + ["is_rock"]
|
62 |
-
data = self.preprocess(data, config=self.config.name)
|
63 |
-
|
64 |
-
for row_id, row in data.iterrows():
|
65 |
-
data_row = dict(row)
|
66 |
-
|
67 |
-
yield row_id, data_row
|
68 |
-
|
69 |
-
def preprocess(self, data: pandas.DataFrame, config: str = DEFAULT_CONFIG) -> pandas.DataFrame:
|
70 |
-
data.loc[:, "is_rock"] = data["is_rock"].apply(lambda x: 1 if x == "R" else 0)
|
71 |
-
|
72 |
-
return data
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|