search

cywang97 / StreamingTransformer

Apache License 2.0
271 stars 42 forks source link

IndexKernel.cu:53 errors #11

Closed alan101-tech closed 3 years ago

alan101-tech commented 3 years ago

Hi, I am currently trying to run the librispeech demo. But, when i run ./train.sh, error happens:

/opt/conda/conda-bld/pytorch_1549636813070/work/aten/src/ATen/native/cuda/IndexKernel.cu:53: lambda ->auto::operator()(int)->auto: block: [34,0,0], thread: [94,0,0] Assertion index >= -sizes[i] && index < sizes[i] && "index out of bounds" failed. /opt/conda/conda-bld/pytorch_1549636813070/work/aten/src/ATen/native/cuda/IndexKernel.cu:53: lambda ->auto::operator()(int)->auto: block: [34,0,0], thread: [95,0,0] Assertion index >= -sizes[i] && index < sizes[i] && "index out of bounds" failed. /opt/conda/conda-bld/pytorch_1549636813070/work/aten/src/ATen/native/cuda/IndexKernel.cu:53: lambda ->auto::operator()(int)->auto: block: [34,0,0], thread: [96,0,0] Assertion index >= -sizes[i] && index < sizes[i] && "index out of bounds" failed. /opt/conda/conda-bld/pytorch_1549636813070/work/aten/src/ATen/native/cuda/IndexKernel.cu:53: lambda ->auto::operator()(int)->auto: block: [34,0,0], thread: [97,0,0] Assertion index >= -sizes[i] && index < sizes[i] && "index out of bounds" failed. /opt/conda/conda-bld/pytorch_1549636813070/work/aten/src/ATen/native/cuda/IndexKernel.cu:53: lambda ->auto::operator()(int)->auto: block: [34,0,0], thread: [98,0,0] Assertion index >= -sizes[i] && index < sizes[i] && "index out of bounds" failed. /opt/conda/conda-bld/pytorch_1549636813070/work/aten/src/ATen/native/cuda/IndexKernel.cu:53: lambda ->auto::operator()(int)->auto: block: [34,0,0], thread: [99,0,0] Assertion index >= -sizes[i] && index < sizes[i] && "index out of bounds" failed.

Did you meet this error before?

Thank you very much!

gothiswaysir commented 3 years ago

@alan101-tech Hi, I encountered the same error, did you solve it?

alan101-tech commented 3 years ago

Yes, in file "conf/train_streaming_transformer.yaml", change "transformer-input-layer" from "conv2d" to "custom", the error will be gone.

gothiswaysir commented 3 years ago

Yes, in file "conf/train_streaming_transformer.yaml", change "transformer-input-layer" from "conv2d" to "custom", the error will be gone.

Thank you! My case is different, there are some out-of-bounds data in the data_aligned.json

alan101-tech commented 3 years ago

solved