正在解决//stay with you

代码修改的位置

Autoformer.py

Autoformer_EncDec.py

tsmoothie/smoother.py at master · cerlymarco/tsmoothie · GitHub

tsmoothie/Basic Smoothing.ipynb at master · cerlymarco/tsmoothie · GitHub

Question on KalmanSmoother usage · Issue #7 · cerlymarco/tsmoothie · GitHub

Fedformer的MOE模块替换Autoformer的Decomp模块，效果确实不错，开始研究

再研究研究tsmoothie模块，还有一些github上面的开源时间序列处理的代码

ICML、ICLR、WWWL、AAAI，时序人整理的代码

吴现在在研究denoisng diffusion模块,看看理论，研究研究

另辟蹊径—Denoising Diffusion Probabilistic 一种从噪音中剥离出图像/音频的模型 - 知乎 (zhihu.com)

【读论文】117 Denoising Diffusion Probabilistic Models_哔哩哔哩_bilibili

刘硕论文里面的SDAE模块可以考虑加入Autoformer. 

GitHub - Zero-coder/MEDIUM_NoteBook: Repository containing notebooks of my posts on Medium

Time Series generation with VAE LSTM [post]|[code]

mixture of experts的论文思想和理论可以看看，MOE-layer

稀疏性在机器学习中的发展趋势：MoE、稀疏注意力机制_zenRRan的博客-CSDN博客

OUTRAGEOUSLY LARGE NEURAL NETWORKS_Buyi_Shizi的博客-CSDN博客

原文地址：

Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer

GitHub - davidmrau/mixture-of-experts: PyTorch Re-Implementation of "The Sparsely-Gated Mixture-of-Experts Layer" by Noam Shazeer et al. https://arxiv.org/abs/1701.06538