TensorComprehensions
TensorComprehensions copied to clipboard
A domain specific language to express machine learning workloads.
Is there any python api?
Templated isl types require the user to specify the domain and range universes of isl objects, allowing the compiler to check whether it makes sense to combine pairs of objects....
Cuda functions can be annotated with launch bounds, that is the maximum number of threads per block (the minimum blocks per multiprocessor can also be specified). This information is used...
These commits implement more functions in the python binders, and also add different options tuning experiments in Python.
Proceeding with ScheduleTree evolution plan (#553) This PR replaces overloaded comparison operators with two functions, `nodeEquals` and `treeEquals`. The main motivation for this change: after grafting ScheduleTreeElem* onto ScheduleTree in...
I got build errors while building CXX object tc/core/CMakeFiles/tc_core_cpu.dir/polyhedral/llvm_jit.cc.o. My settings are in below: - OS: ubuntu 18.04 - How you installed TC (docker, conda, source): source - Python version:...
@nicolasvasilache This is the C++ test for UpSample operator. In OSS env, the test is passing for me without the runtime error. Can you try it out and let me...
This commit is to start support for strided tensors. I made changes to percolate a vector in TensorInfo down to emitCudaKernel to allow codegen to cast strided tensors. This required...