ZhuQue: A Neural Network Programming Model Based on Labeled Data Layout
Explore this paper's citation graph
Summary
A neural network development kit based on labeled data layout is implemented, and a neural network programming model is abstracted that establishes a Neural network computing graph at “creating time”, “compiling time“ sets the data label and “runtime” uses the label to control the data transfer.
- Published
- 2019-08-15
- Cited by
- 2
- References
- 13
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:199571890
References
- CNP: An FPGA-based processor for Convolutional Networks
- PuDianNao: A Polyvalent Machine Learning Accelerator
- DaDianNao: A Machine-Learning Supercomputer
- Supporting Differentiated Services in Computers via Programmable Architecture for Resourcing-on-Demand (PARD)
- DL: A data layout transformation system for heterogeneous computing
- Augmenting Loop Tiling with Data Alignment for Improved Cache Performance
- DianNao: a small-footprint high-throughput accelerator for ubiquitous machine-learning
- Caffe: Convolutional Architecture for Fast Feature Embedding
- Tile size selection using cache organization and data layout
- MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems
- DLPlib: A Library for Deep Learning Processor
- Cambricon-X: An accelerator for sparse neural networks
- This Paper Is Included in the Proceedings of the 12th Usenix Symposium on Operating Systems Design and Implementation (osdi '16). Tensorflow: a System for Large-scale Machine Learning Tensorflow: a System for Large-scale Machine Learning
Cited by
Related papers
No related papers recorded.