plantcad/Angiosperm_65_genomes_8192bp

Revision Notice!

This dataset was updated on 2025-12-010 to correct the duplicated sequences, see issues here. The update replaces the previous version in this repository.

Old version: b5b4eadfd32c6baea27a3e6fb7275ea34f064e87

🌱 Angiosperm_65_genomes_8192bp: 8,192 bp Genomic Windows from 65 Angiosperm Genomes

This dataset provides tokenized 8,192 base-pair (bp) genomic windows across 65 plant genomes from the Phytozome database. It is formatted using the Hugging Face datasets library and supports efficient loading, training, and evaluation of DNA language models.

📦 Dataset Structure

The dataset is stored in Hugging Face's Apache Arrow format and includes three standard splits:

• train/ – main training data
• validation/ – held-out validation set
• test/ – final evaluation set

Each split consists of .arrow shards and is registered in dataset_dict.json.

Example Usage

from datasets import load_dataset

dataset = load_dataset("plantcad/Angiosperm_65_genomes_8192bp")

print(dataset["train"])