evo2_dataset / README.md
Zhongning's picture
Upload folder using huggingface_hub
f02e2c9 verified
|
Raw
History Blame Contribute Delete
2.64 kB
metadata
license: other
tags:
  - chr
  - mini dataset
language:
  - en
  - zh

Evo2 Dataset

Dataset Description

Evo2 Dataset is an Evo2 mini genome dataset adapted for OneScience/evo2/. It contains FASTA, compressed FASTA, and merged FASTA files for human chr20, chr21, and chr22, as well as train, validation, and test .bin/.idx splits preprocessed with the Byte-Level tokenizer.

Supported Tasks

This dataset is not the complete OpenGenome2 dataset and is not intended to reproduce full-scale pretraining. It is intended for preflight checks of the standard Evo2 model repository, FASTA prediction input, and smoke tests for mini training or fine-tuning.

Dataset Format and Structure

Data Format Description
Raw sequences FASTA / gzip FASTA chr20, chr21, chr22, and merged FASTA
Preprocessed training split .bin + .idx chr20_21_22_uint8_distinct_byte-level_train
Preprocessed validation split .bin + .idx chr20_21_22_uint8_distinct_byte-level_val
Preprocessed test split .bin + .idx chr20_21_22_uint8_distinct_byte-level_test
Checksum manifest TSV / SHA256 File names, sizes, and hashes

The default location within the dataset package is data_mini/genome_data/.

How to Use the Dataset

Files and Download

Download the dataset:

hf download --dataset OneScience-Group/evo2_dataset

Model weights must be downloaded separately to validate the model:

hf download --model OneScience-Group/evo2/

Data Placement and Validation

Default model data path:

<MODEL_REPO_ROOT>/data/evo2_dataset/data_mini/genome_data/

Place data_mini in the model package:

mkdir -p <MODEL_REPO_ROOT>/data/evo2_dataset
cp -a data_mini <MODEL_REPO_ROOT>/data/evo2_dataset/data_mini

Validation command:

python scripts/validate_evo2_dataset.py --package-root . --dataset-root data_mini

For a quick structure-only check, add --skip-sha256.

Official OneScience Information

Limitations and License

Repository license: Apache-2.0 Data license: LicenseRef-UCSC-hg38-public-use Data source: UCSC hg38 / GRCh38, chromosomes 20, 21 and 22 Note: Derived bin/idx files follow the source genome data terms.