Commit
•
9954eea
1
Parent(s):
b5b3d35
Delete loading script
Browse files- GTZAN_genre_classification.py +0 -116
GTZAN_genre_classification.py
DELETED
|
@@ -1,116 +0,0 @@
|
|
| 1 |
-
import os
|
| 2 |
-
import csv
|
| 3 |
-
import datasets
|
| 4 |
-
|
| 5 |
-
logger = datasets.logging.get_logger(__name__)
|
| 6 |
-
|
| 7 |
-
_CITATION = """\
|
| 8 |
-
@misc{gtzan2023,
|
| 9 |
-
title={GTZAN Music Genre Classification Dataset},
|
| 10 |
-
author={Your Name},
|
| 11 |
-
year={2023},
|
| 12 |
-
url={https://example.com},
|
| 13 |
-
}
|
| 14 |
-
"""
|
| 15 |
-
|
| 16 |
-
_DESCRIPTION = """\
|
| 17 |
-
The GTZAN dataset is a music genre classification dataset.
|
| 18 |
-
It consists of 10 genres, each represented by 100 tracks.
|
| 19 |
-
"""
|
| 20 |
-
|
| 21 |
-
_HOMEPAGE_URL = "https://example.com"
|
| 22 |
-
_DATA_URL = "data/gtzan.zip"
|
| 23 |
-
|
| 24 |
-
class GTZANConfig(datasets.BuilderConfig):
|
| 25 |
-
"""BuilderConfig for GTZAN"""
|
| 26 |
-
|
| 27 |
-
def __init__(self, **kwargs):
|
| 28 |
-
super(GTZANConfig, self).__init__(**kwargs)
|
| 29 |
-
|
| 30 |
-
|
| 31 |
-
class GTZAN(datasets.GeneratorBasedBuilder):
|
| 32 |
-
BUILDER_CONFIGS = [
|
| 33 |
-
GTZANConfig(name="gtzan", version=datasets.Version("1.0.0"), description="GTZAN Music Genre Classification Dataset")
|
| 34 |
-
]
|
| 35 |
-
|
| 36 |
-
def _info(self):
|
| 37 |
-
return datasets.DatasetInfo(
|
| 38 |
-
description=_DESCRIPTION,
|
| 39 |
-
features=datasets.Features(
|
| 40 |
-
{
|
| 41 |
-
"id": datasets.Value("int32"),
|
| 42 |
-
"audio": datasets.Audio(sampling_rate=44100),
|
| 43 |
-
"genre": datasets.Value("string"),
|
| 44 |
-
"title": datasets.Value("string"),
|
| 45 |
-
"artist": datasets.Value("string"),
|
| 46 |
-
"tempo": datasets.Value("float"),
|
| 47 |
-
"keys": datasets.Value("string"),
|
| 48 |
-
"loudness": datasets.Value("float"),
|
| 49 |
-
"sorted_pred_genres": datasets.Value("string"),
|
| 50 |
-
"x_tsne": datasets.Value("float"),
|
| 51 |
-
"y_tsne": datasets.Value("float"),
|
| 52 |
-
"z_tsne": datasets.Value("float"),
|
| 53 |
-
"x_umap": datasets.Value("float"),
|
| 54 |
-
"y_umap": datasets.Value("float"),
|
| 55 |
-
"z_umap": datasets.Value("float"),
|
| 56 |
-
# "album_cover_path": datasets.Value("string"),
|
| 57 |
-
# "key": datasets.Value("string"),
|
| 58 |
-
# "filepath": datasets.Value("string"),
|
| 59 |
-
}
|
| 60 |
-
),
|
| 61 |
-
supervised_keys=("genre", "title"),
|
| 62 |
-
homepage=_HOMEPAGE_URL,
|
| 63 |
-
citation=_CITATION,
|
| 64 |
-
)
|
| 65 |
-
|
| 66 |
-
def _split_generators(self, dl_manager):
|
| 67 |
-
archive_path = dl_manager.download_and_extract(_DATA_URL)
|
| 68 |
-
data_dir = os.path.join(archive_path, "gtzan")
|
| 69 |
-
|
| 70 |
-
# Debugging the paths
|
| 71 |
-
logger.info(f"Archive path: {archive_path}")
|
| 72 |
-
logger.info(f"Data directory: {data_dir}")
|
| 73 |
-
|
| 74 |
-
return [
|
| 75 |
-
datasets.SplitGenerator(
|
| 76 |
-
name=datasets.Split.TRAIN,
|
| 77 |
-
gen_kwargs={
|
| 78 |
-
"metadata_file": os.path.join(data_dir, "metadata.csv"),
|
| 79 |
-
"data_dir": data_dir,
|
| 80 |
-
},
|
| 81 |
-
),
|
| 82 |
-
]
|
| 83 |
-
|
| 84 |
-
def _generate_examples(self, metadata_file, data_dir):
|
| 85 |
-
with open(metadata_file, "r", encoding="utf-8") as f:
|
| 86 |
-
reader = csv.DictReader(f)
|
| 87 |
-
for id_, row in enumerate(reader):
|
| 88 |
-
# Correct the file paths by removing the 'data/gtzan/' prefix
|
| 89 |
-
file_path = os.path.join(data_dir, row["filepath"].replace("data/gtzan/", ""))
|
| 90 |
-
album_cover_path = os.path.join(data_dir, row["album_cover_path"].replace("data/gtzan/", ""))
|
| 91 |
-
|
| 92 |
-
# Debugging the file paths
|
| 93 |
-
logger.info(f"Processing file: {file_path}")
|
| 94 |
-
if not os.path.exists(file_path):
|
| 95 |
-
logger.error(f"File not found: {file_path}")
|
| 96 |
-
|
| 97 |
-
yield id_, {
|
| 98 |
-
"id": int(row["id"]),
|
| 99 |
-
"audio": file_path, # Include the audio path next to the ID
|
| 100 |
-
"genre": row["genre"],
|
| 101 |
-
"title": row["title"],
|
| 102 |
-
"artist": row["artist"],
|
| 103 |
-
"tempo": float(row["tempo"]),
|
| 104 |
-
"keys": row["keys"],
|
| 105 |
-
"loudness": float(row["loudness"]),
|
| 106 |
-
"sorted_pred_genres": row["sorted_pred_genres"],
|
| 107 |
-
"x_tsne": float(row["x_tsne"]),
|
| 108 |
-
"y_tsne": float(row["y_tsne"]),
|
| 109 |
-
"z_tsne": float(row["z_tsne"]),
|
| 110 |
-
"x_umap": float(row["x_umap"]),
|
| 111 |
-
"y_umap": float(row["y_umap"]),
|
| 112 |
-
"z_umap": float(row["z_umap"]),
|
| 113 |
-
# "album_cover_path": album_cover_path,
|
| 114 |
-
# "key": row["key"],
|
| 115 |
-
# "filepath": file_path,
|
| 116 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|