The Receptive Field (RF) size has been one of the most important factors for One Dimensional Convolutional Neural Networks (1D-CNNs) on time series classification tasks. Large efforts have been taken to choose the appropriate size because it has a huge influence on the performance and differs significantly for each dataset. In this paper, we propose an Omni-Scale block (OS-block) for 1D-CNNs, where the kernel sizes are decided by a simple and universal rule. Particularly, it is a set of kernel sizes that can efficiently cover the best RF size across different datasets via consisting of multiple prime numbers according to the length of the time series. The experiment result shows that models with the OS-block can achieve a similar performance as models with the searched optimal RF size and due to the strong optimal RF size capture ability, simple 1D-CNN models with OS-block achieves the state-of-the-art performance on four time series benchmarks, including both univariate and multivariate data from multiple domains. Comprehensive analysis and discussions shed light on why the OS-block can capture optimal RF sizes across different datasets. Code available [https://github.com/Wensi-Tang/OS-CNN]
翻译:在时间序列分类任务中,一个多维进化神经网络(1D-CNNs)的内核大小一直是一个最重要的因素。已经作出巨大努力,选择适当的大小,因为它对性能有巨大影响,而且每个数据集差异很大。在本文件中,我们提议为1D-CNNs提供一个Omni-Sergy区块(OS-struction),其中内核大小由简单和普遍的规则决定。特别是,它是一套内核大小,能够有效覆盖不同数据集之间最佳的RF规模(1D-CNNs),根据时间序列的长度由多个质数组成。实验结果表明,使用OS-Slock的模型可以取得类似的性能,作为搜索最佳RF尺寸的模型,而使用OS-S-NCs的简单1D-CNNNml模型在四个时间序列基准上达到状态-最先进的性能,包括单级和多域的多级CNCSFS/多级数据。全面分析并讨论为什么现有的OS-NFS-CS-C-CS-CFS-CS-CS-S-COS-Cstestalstable supregmissional