Redirigiendo al acceso original de articulo en 15 segundos...
Inicio  /  Information  /  Vol: 11 Par: 2 (2020)  /  Artículo
ARTÍCULO
TITULO

Fastai: A Layered API for Deep Learning

Jeremy Howard and Sylvain Gugger    

Resumen

fastai is a deep learning library which provides practitioners with high-level components that can quickly and easily provide state-of-the-art results in standard deep learning domains, and provides researchers with low-level components that can be mixed and matched to build new approaches. It aims to do both things without substantial compromises in ease of use, flexibility, or performance. This is possible thanks to a carefully layered architecture, which expresses common underlying patterns of many deep learning and data processing techniques in terms of decoupled abstractions. These abstractions can be expressed concisely and clearly by leveraging the dynamism of the underlying Python language and the flexibility of the PyTorch library. fastai includes: a new type dispatch system for Python along with a semantic type hierarchy for tensors; a GPU-optimized computer vision library which can be extended in pure Python; an optimizer which refactors out the common functionality of modern optimizers into two basic pieces, allowing optimization algorithms to be implemented in 4?5 lines of code; a novel 2-way callback system that can access any part of the data, model, or optimizer and change it at any point during training; a new data block API; and much more. We used this library to successfully create a complete deep learning course, which we were able to write more quickly than using previous approaches, and the code was more clear. The library is already in wide use in research, industry, and teaching.

 Artículos similares

       
 
Tianlong Jia, Hui Qin, Dong Yan, Zhendong Zhang, Bin Liu, Chaoshun Li, Jinwen Wang and Jianzhong Zhou    
Traditional reservoir operation mainly focuses on economic benefits, while ignoring the impacts on navigation. Thus, the economic operation of reservoirs considering navigational demands is of great significance for improving benefits. A navigation capac... ver más
Revista: Water

 
Luis Miguel Silva-Novoa Sanchez, Jeltsje Sanne Kemerink-Seyoum and Margreet Zwarteveen    
Using the concept of sociotechnical tinkering, this paper provides detailed empirical observations about the everyday practices of design, construction, operation, maintenance and use of a piped water supply network in a small town in Mozambique. We use ... ver más
Revista: Water

 
Bruno Martins, Adélia Nunes, Luciano Lourenço and Fátima Velez-Castro    
São Vicente Island (Republic of Cape Verde) lies within the Sahelian zone and faces several natural hazards, one of which is flash flooding. With the purpose of understanding what factors determine flash flood risk perception, a questionnaire entitled Fl... ver más
Revista: Water

 
Husnain Haider, Mohammed Hammed Alkhowaiter, Md. Shafiquzzaman, Saleem S. AlSaleem, Meshal Almoshaogeh and Fawaz Alharbi    
Original Canadian Council of Minster of the Environment Water Quality Index (CCME WQI) is being used for assessing the water quality of surface water sources and distribution systems on a case by case basis. Its full potential as a management tool for co... ver más
Revista: Water

 
Zuhier Alakayleh, Xing Fang and T. Prabhakar Clement    
This study aims at furthering our understanding of the Modified Philip?Dunne Infiltrometer (MPDI), which is used to determine the saturated hydraulic conductivity Ks and the Green?Ampt suction head ? at the wetting front. We have developed a forward-mode... ver más
Revista: Water