metadata

license: mit
task_categories:
  - image-classification
  - zero-shot-image-classification

Extended label set of UC Merced for Hierarchical Multi-Label Classification

This dataset contains an extended version of the original label set of BigEarthNet-43 for Hierarchical Multi-Label Classification.

Dataset creation

It was created based on the CORINE Land Cover database of the year 2018 (CLC 2018), which provides detailed information about the land cover classes at multiple levels of the hierarchy

Loading the Dataset

To load the dataset into your project, you can use the following code snippet:

import pandas as pd
from datasets import load_dataset

# Load the dataset from Hugging Face
dataset = load_dataset(
    "marjandl/UCM-HMLC", 
    split='train'
)

data = [line.split('\t')[1:] for line in dataset['text']]
df = pd.DataFrame(data)

# Set the first row as the header and drop it from the DataFrame
df.columns = df.iloc[0]
df = df.drop(0).reset_index(drop=True)