EXTREMUM_web/base/download_wildboar_datasets.py

29 lines
981 B
Python

import pandas as pd
from wildboar.datasets import load_dataset, list_datasets
# Get the list of all available datasets in the 'wildboar/ucr' repository
available_datasets = list_datasets(repository='wildboar/ucr')
available_datasets = ["FordB"]
for dataset_name in available_datasets:
try:
# Load the dataset
x, y = load_dataset(dataset_name, repository='wildboar/ucr')
# Convert x to a DataFrame
x_df = pd.DataFrame(x)
# Convert y to a Series with the name 'label'
y_df = pd.Series(y, name='label')
# Concatenate x and y along columns (axis=1)
merged_df = pd.concat([x_df, y_df], axis=1)
# Save the DataFrame to a CSV file
filename = f'{dataset_name}_merged.csv'
merged_df.to_csv(filename, index=False)
print(f"File saved as '{filename}'")
except Exception as e:
print(f"Failed to process dataset '{dataset_name}': {e}")