NVIDIA Deep Learning AMI vs Tensor2Tensor

Many state of the art and baseline models are built-in and new models can be added easily; Many datasets across modalities - text, audio, image - available for generation and use, and new ones can be added easily; Models can be used with any dataset and input mode (or even multiple); all modality-specific processing (e.g. embedding lookups for text tokens) is done with bottom and top transformations, which are specified per-feature in the model; Support for multi-GPU machines and synchronous (1 master, many workers) and asynchronous (independent workers synchronizing through a parameter server) distributed training; Easily swap amongst datasets and models by command-line flag with the data generation script t2t-datagen and the training script t2t-trainer; Train on Google Cloud ML and Cloud TPUs

Provides AI researchers with fast and easy access to NVIDIA A100, V100 and T4 GPUs in the cloud, with performance-engineered deep learning framework containers that are fully integrated, optimized, and certified by NVIDIA; Optimized for highest performance across a wide range of workloads on NVIDIA GPUs; NVIDIA accelerates innovation by eliminating the complex do-it-yourself task of building and optimizing a complete deep learning software stack tuned specifically for GPUs