[bot] Conversion to Parquet
#1
by
parquet-converter
- opened
The parquet-converter bot has created a version of this dataset in the Parquet format. You can learn more about the advantages associated with this format in the documentation.
The Parquet files are published in the refs/convert/parquet branch.
FYI: Some of the files seems to be missing from the git repo, making it hard to load the data. However, the dataset can be read remotely as follows (based on Claude).
from datasets import load_dataset
Load the dataset
dataset = load_dataset("RobinWZQ/improved_aesthetics_6.5plus")
Convert to pandas DataFrame
df = dataset['train'].to_pandas()
Save as CSV
df.to_csv('improved_aesthetics_6.5plus.csv', index=False)