VEWOXIC
commented
4 months ago Maybe, you can try updating the torch version to 2.1 or above first? We will upload the requirement file later.
Closed 1353918285 closed 4 months ago
VEWOXIC
commented
4 months ago Maybe, you can try updating the torch version to 2.1 or above first? We will upload the requirement file later.
请问能否提供一下得到最好结果的各个包版本？ 另外我在复现的时候出现了一些错误，执行etth1_best.sh的时候出现了以下错误： exp = Exp(args) # set experimentsile "run_longExp_F.py", line 152, in
File "E:\YD\code\FITS-main\exp\exp_main_F.py", line 25, in init
super(Exp_Main, self).init(args) File "E:\YD\code\FITS-main\exp\exp_basic.py", line 10, in init self.model = self._build_model().to(self.device) File "E:\YD\code\FITS-main\exp\exp_main_F.py", line 39, in _build_model model = model_dict[self.args.model].Model(self.args).float() File "E:\YD\code\FITS-main\models\FITS.py", line 27, in init self.freq_upsampler = nn.Linear(self.dominance_freq, int(self.dominance_freq*self.length_ratio)).to(torch.cfloat) # complex layer for frequency upcampling] File "D:\anaconda\envs\TSL\lib\site-packages\torch\nn\modules\module.py", line 604, in to raise TypeError('nn.Module.to only accepts floating point ' TypeError: nn.Module.to only accepts floating point dtypes, but got desired dtype=torch.complex64 请问您能否给出一些建议，感激不尽！！