Translations:AI and Machine Learning/33/en

RNN and multi-head attention API calls may exhibit non-deterministic behaviour when the cuDNN library is built with CUDA Toolkit 10.2 or higher. The user can eliminate the non-deterministic behaviour of cuDNN RNN and multi-head attention APIs by setting a single buffer size in the CUBLAS_WORKSPACE_CONFIG environmental variable, for example, :16:8 or :4096:2, which instructs cuBLAS to allocate eight buffers of 16 KB each in GPU memory or two buffers of 4 MB each.