NX-AI
ImportError: DLL load failed while importing slstm_HS64BS8NH1NS4DBfDRbDWbDGbDSbDAfNG4SA1GRCV0GRC0d0FCV0FC0d0: 找不到指定的模块。
'verbose': True, 'with_cuda': True, 'extra_ldflags': ['-LD:\CUDA_manager\CUDA12.0\CUDA1\lib', '-lcublas'], 'extra_cflags': ['-DSLSTM_HIDDEN_SIZE=64', '-DSLSTM_BATCH_SIZE=8', '-DSLSTM_NUM_HEADS=1', '-DSLSTM_NUM_STATES=4', '-DSLSTM_DTYPE_B=float', '-DSLSTM_DTYPE_R=nv_bfloat16', '-DSLSTM_DTYPE_W=nv_bfloat16', '-DSLSTM_DTYPE_G=nv_bfloat16', '-DSLSTM_DTYPE_S=nv_bfloat16', '-DSLSTM_DTYPE_A=float', '-DSLSTM_NUM_GATES=4', '-DSLSTM_SIMPLE_AGG=true', '-DSLSTM_GRADIENT_RECURRENT_CLIPVAL_VALID=false', '-DSLSTM_GRADIENT_RECURRENT_CLIPVAL=0.0', '-DSLSTM_FORWARD_CLIPVAL_VALID=false', '-DSLSTM_FORWARD_CLIPVAL=0.0', '-U__CUDA_NO_HALF_OPERATORS', '-U__CUDA_NO_HALF_CONVERSIONS', '-U__CUDA_NO_BFLOAT16_OPERATORS', '-U__CUDA_NO_BFLOAT16_CONVERSIONS', '-U__CUDA_NO_BFLOAT162_OPERATORS__', '-U__CUDA_NO_BFLOAT162_CONVERSIONS__'], 'extra_cuda_cflags': ['-Xptxas="-v"', '-gencode', 'arch=compute_80,code=compute_80', '-res-usage', '--use_fast_math', '-O3', '-Xptxas -O3', '--extra-device-vectorization', '-DSLSTM_HIDDEN_SIZE=64', '-DSLSTM_BATCH_SIZE=8', '-DSLSTM_NUM_HEADS=1', '-DSLSTM_NUM_STATES=4', '-DSLSTM_DTYPE_B=float', '-DSLSTM_DTYPE_R=nv_bfloat16', '-DSLSTM_DTYPE_W=nv_bfloat16', '-DSLSTM_DTYPE_G=nv_bfloat16', '-DSLSTM_DTYPE_S=nv_bfloat16', '-DSLSTM_DTYPE_A=float', '-DSLSTM_NUM_GATES=4', '-DSLSTM_SIMPLE_AGG=true', '-DSLSTM_GRADIENT_RECURRENT_CLIPVAL_VALID=false', '-DSLSTM_GRADIENT_RECURRENT_CLIPVAL=0.0', '-DSLSTM_FORWARD_CLIPVAL_VALID=false', '-DSLSTM_FORWARD_CLIPVAL=0.0', '-U__CUDA_NO_HALF_OPERATORS', '-U__CUDA_NO_HALF_CONVERSIONS', '-U__CUDA_NO_BFLOAT16_OPERATORS', '-U__CUDA_NO_BFLOAT16_CONVERSIONS', '-U__CUDA_NO_BFLOAT162_OPERATORS__', '-U__CUDA_NO_BFLOAT162_CONVERSIONS__']}
Using C:\Users\a\AppData\Local\torch_extensions\torch_extensions\Cache\py39_cu118 as PyTorch extensions root...
Creating extension directory C:\Users\a\AppData\Local\torch_extensions\torch_extensions\Cache\py39_cu118\slstm_HS64BS8NH1NS4DBfDRbDWbDGbDSbDAfNG4SA1GRCV0GRC0d0FCV0FC0d0...
Detected CUDA files, patching ldflags
Emitting ninja build file C:\Users\a\AppData\Local\torch_extensions\torch_extensions\Cache\py39_cu118\slstm_HS64BS8NH1NS4DBfDRbDWbDGbDSbDAfNG4SA1GRCV0GRC0d0FCV0FC0d0\build.ninja...
Building extension module slstm_HS64BS8NH1NS4DBfDRbDWbDGbDSbDAfNG4SA1GRCV0GRC0d0FCV0FC0d0...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
1.11.1.git.kitware.jobserver-1
Loading extension module slstm_HS64BS8NH1NS4DBfDRbDWbDGbDSbDAfNG4SA1GRCV0GRC0d0FCV0FC0d0...
Traceback (most recent call last):
File "C:\Users\a\Downloads\xlstm-main\xlstm-main\experiments\main.py", line 155, in
I think slstm is not work,I can run parity_xlstm01.yaml,but 11 is not work
