search

ChenQian0618 / TFN

this is the open code of paper entitled "TFN: An Interpretable Neural Network With Time Frequency Transform Embedded for Intelligent Fault Diagnosis".
MIT License
74 stars 12 forks source link

channelwise amplitude–frequency response (C-FR) #3

Closed liguge closed 7 months ago

liguge commented 7 months ago

恭喜恭喜，论文终于在线发表了。请问 channelwise amplitude–frequency response (C-FR)这个函数代码开源了吗？我想要使用一下这个方法，不胜感激。

ChenQian0618 commented 7 months ago

Sorry, this is beyond our workload. While, this is not very difficult to record the convolutional kernel weight during the training process, and then use FFT in channel-wise to calculate the frequency spectrum, which is the C-FR exactly. Literature [1] have shown the equivalence between the convolutional kernel and the filter process, and literature [2] already did frequency response analysis upon the convolutional kernel weight in 2018.

[1] V. Andrearczyk and P. F. Whelan, “Using filter banks in Convolutional Neural Networks for texture classification,” Pattern Recognition Letters, vol. 84, pp. 63–69, Dec. 2016, doi: 10.1016/j.patrec.2016.08.016. [2] M. Ravanelli and Y. Bengio, “Speaker Recognition from Raw Waveform with SincNet,” in 2018 IEEE Spoken Language Technology Workshop (SLT), Athens, Greece: IEEE, Dec. 2018, pp. 1021–1028. doi: 10.1109/SLT.2018.8639585.

liguge commented 7 months ago

thanks a lot.

At 2023-11-29 11:39:00, "ChenQian0618" @.***> wrote:

Sorry, this is beyond our workload. While, this is not very difficult to record the convolutional kernel weight during the training process, and then use FFT in channel-wise to calculate the frequency spectrum, which is the C-FR exactly. Literature [1] have shown the equivalence between the convolutional kernel and the filter process, and literature [2] already did frequency response analysis upon the convolutional kernel weight in 2018.

[1] V. Andrearczyk and P. F. Whelan, “Using filter banks in Convolutional Neural Networks for texture classification,” Pattern Recognition Letters, vol. 84, pp. 63–69, Dec. 2016, doi: 10.1016/j.patrec.2016.08.016. [2] M. Ravanelli and Y. Bengio, “Speaker Recognition from Raw Waveform with SincNet,” in 2018 IEEE Spoken Language Technology Workshop (SLT), Athens, Greece: IEEE, Dec. 2018, pp. 1021–1028. doi: 10.1109/SLT.2018.8639585.

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>