论文信息 - DeepPy: Pythonic deep learning

DeepPy: Pythonic deep learning

This technical report introduces DeepPy – a deep learning framework built on top of NumPy with GPU acceleration. DeepPy bridges the gap between highperformance neural networks and the ease of development from Python/NumPy. Users with a background in scientific computing in Python will quickly be able to understand and change the DeepPy codebase as it is mainly implemented using high-level NumPy primitives. Moreover, DeepPy supports complex network architectures by letting the user compose mathematical expressions as directed graphs. The latest version is available at http://github.com/andersbll/deeppy under the MIT license.

Anders Boesen Lindbo Larsen

[1] Sergey Ioffe,et al. Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[2] John Tran,et al. cuDNN: Efficient Primitives for Deep Learning , 2014, ArXiv.

[3] Jimmy Ba,et al. Adam: A Method for Stochastic Optimization , 2014, ICLR.

[4] Nitish Srivastava,et al. Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[5] Diederik P. Kingma,et al. Stochastic Gradient VB and the Variational Auto-Encoder , 2013 .

[6] Kenta Oono,et al. Chainer : a Next-Generation Open Source Framework for Deep Learning , 2015 .

[7] Travis E. Oliphant,et al. Python for Scientific Computing , 2007, Computing in Science & Engineering.

[8] Martín Abadi,et al. TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems , 2016, ArXiv.

[9] Daan Wierstra,et al. Stochastic Backpropagation and Approximate Inference in Deep Generative Models , 2014, ICML.

[10] Clément Farabet,et al. Torch7: A Matlab-like Environment for Machine Learning , 2011, NIPS 2011.

[11] Yoshua Bengio,et al. Generative Adversarial Nets , 2014, NIPS.

[12] Anders Boesen Lindbo Larsen. CUDArray: CUDA-based NumPy , 2014 .