PyOpenCL not running on a server-node with a Nvidia Tesla card
up vote 0 down vote favorite I am trying to run a script that uses PyOpenCL (or PyCuda) to executes a simulation on the Tesla GPU on a node of a compute cluster. However, I get the following error: LogicError: clGetPlatformIDs failed: PLATFORM_NOT_FOUND_KHR If I run it in plain OpenCL, using a C++ code, there is no problem, so it is not that OpenCL is not installed properly. But I would like to run it from Jupiter (or plain python), and this doesn't work. Switching to PyCuda doesn't help, also gives an error. Some additional information: my python installation is Anaconda3-5.2.0-Linux-x86_64 with PyOpenCL added via conda. Who can help me with this? Thanks in advance, Johan linux opencl pycuda pyopencl share | improve this question edited 2 days ago asked 2 days ago Johan vd Koppel 1 1 New contributor Johan vd Koppel is a new contributor to this site. Take care in asking for clarification, commenting, and answering. Check...