search

cure-lab / SCINet

The GitHub repository for the paper: “Time Series is a Special Sequence: Forecasting with Sample Convolution and Interaction“. (NeurIPS 2022)
Apache License 2.0
633 stars 129 forks source link

'Tensor' object has no attribute 'dtpye' #28

Closed Jiahui-Gu closed 2 years ago

Jiahui-Gu commented 2 years ago

hello, I got an error

Traceback (most recent call last): File "runpems.py", line 87, in , normalize_statistic = exp.train() File "D:\gjh\workspace\SCINet\experiments\exp_pems.py", line 278, in train forecast = self.model(inputs) File "C:\Users\dell\anaconda3\envs\scinet\lib\site-packages\torch\nn\modules\module.py", line 889 , in _call_impl result = self.forward(*input, **kwargs) File "D:\gjh\workspace\SCINet\models\SCINet.py", line 355, in forward output = torch.zeros(x.shape,dtype=x.dtpye).cuda() AttributeError: 'Tensor' object has no attribute 'dtpye'

when I was running

python run_pems.py --dataset PEMS03 --hidden-size 0.0625 --dropout 0.25 --model_name pems03_h0.0625_dp0.25 --num_decoder_layer 2

I use below command build my environment

conda create -n scinet python=3.8 conda activate scinet pip install -r requirements.txt (without pytorch) conda install pytorch==1.8.0 torchvision==0.9.0 torchaudio==0.8.0 cudatoolkit=10.2 -c pytorch -y

How can I solve this problem? Thanks！

VEWOXIC commented 2 years ago

Hello, It seems like the 'dtpye' is a typo, please try changing it into 'dtype'.

VEWOXIC commented 2 years ago

We have fixed this typo in the repo. If there is no further query. We will close this issue. Thanks for your attention.