tvm
tvm copied to clipboard
apache
Open deep learning compiler stack for cpu, gpu and specialized accelerators
In rare cases, a `PrimFunc` may be annotated with `StructInfo`, to indicate that it is an impure function with specific shapes for the parameters. If struct info is present, it...
As discussed in several TVM Open Development meetings, we were permitting invalid programs to be parsed by not checking well-formedness during parsing. This PR makes a small change to the...
I was trying to understand the use of **T.assume** in the simplification of expressions and I came across this example in the "**test_tir_transform_simplify.py**" file named as "**TestSimplifyUsingPartiallyKnownBufferExpression**". Here, I tried...
Move cat image url to tlcpack/web-data cc @tqchen
Relax is capable of expressing tensors whose element type is unknown. However, these must typically be replaced with a known dtype prior to compilation, as most operators require known data...
Thanks for participating in the TVM community! We use https://discuss.tvm.ai for any general usage questions and discussions. The issue tracker is used for actionable items such as feature proposals discussion,...
MXNet has been archived and likely future versions of tvm should not depend on it. This is a tracking issue for us to remove the related stale testcases and documentations...
This PR introduces the `multinomial_from_uniform` into nn module for GPU case, and `vm.builtin.multinomial_from_uniform` for CPU. The `multinomial_from_uniform` returns a tensor wherein each row contains an index sampled from a multinomial...
Prior to this commit, a build that used multiple targets needed to provide `tvm::build` with a `Map` specifying which target should be used to compile each `IRModule`. As a result,...
Verifying behavior for subroutines that receive `R.Prim` or `R.Shape` parameters, if the symbolic variables defined by those parameters are already defined by another parameter.