TensorComprehensions
TensorComprehensions copied to clipboard
A domain specific language to express machine learning workloads.
trying to add bitwise operators. so far they are parsed correctly in their precedence order. `tc2halide` conversion is also done. I am not sure about how to propagate these through...
There are still a few things that could have been improved, but I think this can be done in subsequent PRs, and I wanted to get some feedback at this...
I'm trying to programatically generate array based code (like one uses LLVM to codegen to assembly). I was hoping to find some sort of LLVM's `IRBuilder` like abstraction, but couldn't...
Most of TC examples are actually for tensor reshape or reduce. Any examples for tensor broadcast? The following try isn't supported by TC: ```sh def broadcast(float(N, M) I0) -> (O)...
Avoid making a DAG of Halide Funcs, and instead just produce a Halide Stmt directly. Should have no observable effect on anything, other than producing slightly simpler variable names (hence...
Tensor Comprehensions Github Issues Guidelines ---------------------------------------------- When submitting a bug report, please include the following information (where relevant): - OS: Ubuntu 18.04 - How you installed TC (docker, conda, source):...
Trying to configure with different versions of aten on ubuntu16.04LTS recieving same error with each version. Cmake is complete but make fails to link executable binary got this when I...
- OS: Ubuntu 16.04 - How you installed TC (docker, conda, source): source - Python version: 3.5 - CUDA/cuDNN version: .9.0/7.4 - Conda version (if using conda): 3.7 - GCC/GXX...
TensorComprehensions is a great lib, it saves a lot of work of writing CUDA kernel and finetune, so I attempt to integrate TC to my Framework. But I am curious...