causal_conv1d_ref()参数调用错误
Oytpzy opened this issue · comments
作者您好，我在运行代码的过程中遇到了位置信息输入越界的问题，
Traceback (most recent call last):
File "D:\TSproject\S-D-Mamba-main\S-D-Mamba-main\run.py", line 135, in
exp.train(setting)
File "D:\TSproject\S-D-Mamba-main\S-D-Mamba-main\experiments\exp_long_term_forecasting.py", line 144, in train
outputs = self.model(batch_x, batch_x_mark, dec_inp, batch_y_mark)
File "D:\anaconda\lib\site-packages\torch\nn\modules\module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "D:\TSproject\S-D-Mamba-main\S-D-Mamba-main\model\S_Mamba.py", line 93, in forward
dec_out = self.forecast(x_enc, x_mark_enc, x_dec, x_mark_dec)
File "D:\TSproject\S-D-Mamba-main\S-D-Mamba-main\model\S_Mamba.py", line 80, in forecast
enc_out, attns = self.encoder(enc_out, attn_mask=None)
File "D:\anaconda\lib\site-packages\torch\nn\modules\module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "D:\TSproject\S-D-Mamba-main\S-D-Mamba-main\layers\Mamba_EncDec.py", line 68, in forward
x, attn = attn_layer(x, attn_mask=attn_mask, tau=tau, delta=delta)
File "D:\anaconda\lib\site-packages\torch\nn\modules\module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "D:\TSproject\S-D-Mamba-main\S-D-Mamba-main\layers\Mamba_EncDec.py", line 37, in forward
new_x = self.attention(x) + self.attention_r(x.flip(dims=[1])).flip(dims=[1])
File "D:\anaconda\lib\site-packages\torch\nn\modules\module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "D:\anaconda\lib\site-packages\mamba_ssm\modules\mamba_simple.py", line 146, in forward
out = mamba_inner_fn(
File "D:\anaconda\lib\site-packages\mamba_ssm\ops\selective_scan_interface.py", line 306, in mamba_inner_fn
return mamba_inner_ref(xz, conv1d_weight, conv1d_bias, x_proj_weight, delta_proj_weight,
File "D:\anaconda\lib\site-packages\mamba_ssm\ops\selective_scan_interface.py", line 321, in mamba_inner_ref
x = causal_conv1d_fn(x, rearrange(conv1d_weight, "d 1 w -> d w"), conv1d_bias, "silu")
File "D:\anaconda\lib\site-packages\causal_conv1d\causal_conv1d_interface.py", line 49, in causal_conv1d_fn
return causal_conv1d_ref(x, weight, bias, seq_idx, activation)
TypeError: causal_conv1d_ref() takes from 2 to 4 positional arguments but 5 were given
，请问 causal_conv1d_ref()这个函数只能接受2-4个位置参数，为什么会出现输入5位置信息的情况呀
没有碰到过这种情况, 版本正确了吗？是改参数了吗？同样的，如果有这样的Mamb Block相关的问题，建议去Mamba项目下的issue里找类似的解决方案...