Hugging Face's logo

AIRI-Institute
/
gena-lm-bigbird-base-sparse

Transformers
PyTorch
bert
pretraining
dna
human_genome
custom_code
Inference Endpoints
Model card Files Community
gena-lm-bigbird-base-sparse / README.md
yurakuratov's picture
yurakuratov
update readme
764cdde
preview code
|
raw
history blame
3.75 kB
metadata
tags:
  - dna
  - human_genome

GENA-LM (gena-lm-bigbird-base-sparse)

GENA-LM is a Family of Open-Source Foundational Models for Long DNA Sequences.

GENA-LM models are transformer masked language models trained on human DNA sequence.

gena-lm-bigbird-base-sparse follows the BigBird architecture and uses sparse attention from DeepSpeed.

Differences between GENA-LM (gena-lm-bigbird-base-sparse) and DNABERT:

  • BPE tokenization instead of k-mers;
  • input sequence size is about 36000 nucleotides (4096 BPE tokens) compared to 512 nucleotides of DNABERT;
  • pre-training on T2T vs. GRCh38.p13 human genome assembly.

Source code and data: https://github.com/AIRI-Institute/GENA_LM

Paper: https://www.biorxiv.org/content/10.1101/2023.06.12.544594v1

Installation

gena-lm-bigbird-base-sparse sparse ops require DeepSpeed.

DeepSpeed

DeepSpeed installation is needed to work with SparseAttention versions of language models. DeepSpeed Sparse attention supports only GPUs with compute compatibility >= 7 (V100, T4, A100).

pip install triton==1.0.0
DS_BUILD_SPARSE_ATTN=1 pip install deepspeed==0.6.0 --global-option="build_ext" --global-option="-j8" --no-cache

and check installation with

ds_report

APEX for FP16

Install APEX https://github.com/NVIDIA/apex#quick-start

git clone https://github.com/NVIDIA/apex
cd apex
pip install -v --disable-pip-version-check --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./

Examples

Load pre-trained model 

from transformers import AutoTokenizer, BigBirdForMaskedLM

tokenizer = AutoTokenizer.from_pretrained('AIRI-Institute/gena-lm-bigbird-base-sparse')
model = BigBirdForMaskedLM.from_pretrained('AIRI-Institute/gena-lm-bigbird-base-sparse')

How to load the model to fine-tune it on classification task 

from transformers import AutoTokenizer, BigBirdForSequenceClassification

tokenizer = AutoTokenizer.from_pretrained('AIRI-Institute/gena-lm-bigbird-base-sparse')
model = BigBirdForSequenceClassification.from_pretrained('AIRI-Institute/gena-lm-bigbird-base-sparse')

Model description

GENA-LM (gena-lm-bigbird-base-sparse) model is trained in a masked language model (MLM) fashion, following the methods proposed in the BigBird paper by masking 15% of tokens. Model config for gena-lm-bigbird-base-sparse is similar to the google/bigbird-roberta-base:

  • 4096 Maximum sequence length
  • 12 Layers, 12 Attention heads
  • 768 Hidden size
  • sparse config:
    • block size: 64
    • random blocks: 3
    • global blocks: 2
    • sliding window blocks: 3
  • Rotary positional embeddings
  • 32k Vocabulary size, tokenizer trained on DNA data.

We pre-trained gena-lm-bigbird-base-sparse using the latest T2T human genome assembly (https://www.ncbi.nlm.nih.gov/assembly/GCA_009914755.3/). Pre-training was performed for 810,000 iterations with batch size 256. We modified Transformer with Pre-Layer normalization.

Evaluation

For evaluation results, see our paper: https://www.biorxiv.org/content/10.1101/2023.06.12.544594v1

Citation 

@article{GENA_LM,
    author = {Veniamin Fishman and Yuri Kuratov and Maxim Petrov and Aleksei Shmelev and Denis Shepelin and Nikolay Chekanov and Olga Kardymon and Mikhail Burtsev},
    title = {GENA-LM: A Family of Open-Source Foundational Models for Long DNA Sequences},
    elocation-id = {2023.06.12.544594},
    year = {2023},
    doi = {10.1101/2023.06.12.544594},
    publisher = {Cold Spring Harbor Laboratory},
    URL = {https://www.biorxiv.org/content/early/2023/06/13/2023.06.12.544594},
    eprint = {https://www.biorxiv.org/content/early/2023/06/13/2023.06.12.544594.full.pdf},
    journal = {bioRxiv}
}