Data

BBBC021 Loader

data.pybbbc_loader.download_bbbc021(data_root: str = '/scratch/cv-course2025/group8') → None

Downloads the complete BBBC021 dataset.

Parameters:

data_root – Root directory where the dataset will be stored.

Note

We are using the pybbbc library to download the dataset. This function is simply a wrapper around the BBBC021.download method.

Note

If you are using RAMSES, you can use the default path, the data should be there already.

Dataset Structure:

After download, the directory structure should look like:

data_root/raw/
├── images/
│   ├── Week1_22123/
│   ├── Week1_22141/
│   ├── Week1_22161/
│   └── ... (all experimental weeks)
└── some_metadata_file.csv

Example

>>> # Download to default location
>>> download_bbbc021()

>>> # Download to custom location
>>> download_bbbc021("/path/to/my/data/bbbc021")

References

Dataset: https://bbbc.broadinstitute.org/BBBC021 pybbbc docs: giacomodeodato/pybbbc

data.pybbbc_loader.preprocess_bbbc021(data_root: str = '/scratch/cv-course2025/group8') → None

Preprocess the BBBC021 dataset. More information on the preprocessing can be found in the pybbbc documentation.

Parameters:

data_root – Root directory where the raw dataset is stored.