European Distributed Deep Learning (EDDL) library. A general-purpose library initially developed to cover deep learning needs in healthcare use cases within the DeepHealth project.
The mutex must be locked by the current thread of execution, otherwise, the behavior is undefined.
I think you could use std::unique_lock and std::lock_guard to manage semaphore for Tensor.
I don't know if this error also happens with other compilers.
I'm trying to execute the pipeline with Visual Studio 2017, but the application crashes with
unlock of unowned mutex
error.The crash occurs in Tensor destructor when trying to unlock an already unlocked mutex: https://github.com/deephealthproject/eddl/blob/6a84eb9358c077a76ebe0a9bab99c452367e5259/src/tensor/tensor.cpp#L111
The mutex::unlock() reference says:
I think you could use
std::unique_lock
andstd::lock_guard
to manage semaphore for Tensor. I don't know if this error also happens with other compilers.