eong's picture
Update README.md
882afd0 verified
---
dataset_info:
features:
- name: image
dtype: image
- name: label
dtype:
class_label:
names:
'0': Blues
'1': Classical
'2': Country
'3': DeathMetal
'4': DoomMetal
'5': DrumNBass
'6': Electronic
'7': Folk
'8': Grime
'9': HeavyMetal
'10': HipHop
'11': Jazz
'12': LoFi
'13': Pop
'14': PsychedelicRock
'15': Punk
'16': Reggae
'17': Rock
'18': Soul
'19': Techno
splits:
- name: train
num_bytes: 237337295.0
num_examples: 20000
download_size: 236720781
dataset_size: 237337295.0
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
---
# Dataset Card for "20k-Album-Covers-within-20-Genres"
The dataset contains common album covers for 20 music genres.
Each genre has 1000 album covers.
The dataset was got frame [Kaggle](https://www.kaggle.com/datasets/michaeljkerr/20k-album-covers-within-20-genres)