On Monday, Google brain team released its Open source system called Tensor2Tensor(T2T) for orientation of deep learning models.
These T2T libraries assist researchers to simulate results from recent papers, pushing the boundaries with a new synthesis of models, datasets, hyperparameters, and so on.It is created with Tensor Flow tools and empowers the best practices for AI deep learning models.
It enforces a standard interface in various useful technologies like object detection, machine translation and speech recognition in the popular ML applications.
“T2T facilitates the creation of state-of-the-art models for a wide variety of ML applications, such as translation, parsing, image captioning and more, enabling the exploration of various ideas much faster than previously possible,” said Lukasz Kaiser, senior research scientist on the Google Brain team.
Not only does Google aim to make research more reproducible outside the lab, but also one Facebook open-sourced tool, ParlAI, enables dialog research that comes with prepackaged datasets more often.In a similar fashion, Google’s Tensor2Tensor function on the models from Google research projects such as “Attention Is All You Need” and “One Model To Learn Them All”.
This tool is open source and can be availed here.