zoukankan      html  css  js  c++  java
  • Deep Learning for Time Series Classification (InceptionTime)

    Deep Learning for Time Series Classification (InceptionTime)

    2020-02-03 17:02:30

     

    Sourcehttps://towardsdatascience.com/deep-learning-for-time-series-classification-inceptiontime-245703f422db 

    Index

    1. Motivation

    2. Machine Learning for Time Series Classification

     
    Fig. 2: The Fourier series expansion of a square wave (red line). Here I present only the first three modes (blue dashed lines) and their addition (green line). Hopefully, it is not hard to see that by adding the next modes the series quickly converges to a square wave.

    3. Best DL practices for TSC: InceptionTime

     
    Fig. 3: The inception module of InceptionTime. The first number in the boxes indicates the kernel size while the second indicates the size of the stride. “(S)” specifies the type of padding, i.e. “same”.
     
     
    Fig. 4: The Inception network for time series classification.
     
    Fig. 5: Residual Connections in the Inception network.
     

    4. Understanding InceptionTime

     
    Fig. 6: The receptive field of neurons in filters.

    5. Conclusion

    References

  • 相关阅读:
    直播平台的相关技术(转载)
    各种排序算法分析总结(待整理))
    算法:60.第k个排列
    三种随机化算法:舍伍德算法 拉斯维加斯算法 蒙特卡洛算法
    随机化算法之随机数
    caffe调试
    Euclideanloss_layer层解析
    布线问题(分支限界法)
    最大堆和最小堆
    机器学习知识框架
  • 原文地址:https://www.cnblogs.com/wangxiaocvpr/p/12256355.html
Copyright © 2011-2022 走看看