主题策略-【专题报告】AI+HI系列(2):PatchTST、TSMixer、ModernTCN时序深度网络构建量价因子-华创证券[秦玄晋,王小川]-20240311【34页】.docx,证监会审核华创证券投资咨询业务资格批文号:证监许可(2009)1210 号 未经许可,禁止转载 金融工程 金融工程 证券研究报
python -u run.py --task_name classification --is_training 1 --root_path ./all_datasets/Heartbeat/ --data UEA --model_id Heartbeat --model ModernTCN --ffn_ratio 1 --patch_size 8 --patch_stride 4 --num_blocks 1 --large_size 31 --small_size 5 --dims 256 --head_dropout 0.0 ...
Recurrent neural networks, containing CNN-LSTM [130], CNN-GRU [131], and TCN-LSTM [132] layers, require advanced Adam-type algorithms. Second-order optimization methods have a higher rate of convergence in the neighborhood of global minimum, but their time consumption is higher. Even quasi-...