NVIDIA NeMo Megatron Integrations

BioNeMo is a cloud service and framework for drug discovery that leverages AI, built on NVIDIA NeMo Megatron, which enables the training and deployment of large-scale biomolecular transformer models. This service features pre-trained large language models (LLMs) and offers comprehensive support for standard file formats related to proteins, DNA, RNA, and chemistry, including data loaders for SMILES molecular structures and FASTA sequences for amino acids and nucleotides. Additionally, users can download the BioNeMo framework for use on their own systems. Among the tools provided are ESM-1 and ProtT5, both transformer-based protein language models that facilitate the generation of learned embeddings for predicting protein structures and properties. Furthermore, the BioNeMo service will include OpenFold, an advanced deep learning model designed for predicting the 3D structures of novel protein sequences, enhancing its utility for researchers in the field. This comprehensive offering positions BioNeMo as a pivotal resource in modern drug discovery efforts.

Amazon SageMaker Model Training streamlines the process of training and fine-tuning machine learning (ML) models at scale, significantly cutting down both time and costs while eliminating the need for infrastructure management. Users can leverage top-tier ML compute infrastructure, benefiting from SageMaker’s capability to seamlessly scale from a single GPU to thousands, adapting to demand as necessary. The pay-as-you-go model enables more effective management of training expenses, making it easier to keep costs in check. To accelerate the training of deep learning models, SageMaker’s distributed training libraries can divide extensive models and datasets across multiple AWS GPU instances, while also supporting third-party libraries like DeepSpeed, Horovod, or Megatron for added flexibility. Additionally, you can efficiently allocate system resources by choosing from a diverse range of GPUs and CPUs, including the powerful P4d.24xl instances, which are currently the fastest cloud training options available. With just one click, you can specify data locations and the desired SageMaker instances, simplifying the entire setup process for users. This user-friendly approach makes it accessible for both newcomers and experienced data scientists to maximize their ML training capabilities.