tvm
tvm copied to clipboard
pytorch
TVM integration into PyTorch
This is the error that I get incl. some text before. I make it to ca. 95%. Any idea anyone what is wrong? /usr/include/c++/8/bits/range_access.h:87:5: note: 'std::begin' begin(_Tp (&__arr)[_Nm]) ^~~~~ /home/pi/code/tvm/torch_tvm/compiler.cpp:201:43:...
I am trying to build the project but it is failing errors on llvm-project directory. Here are steps I am following to build the project on my mac laptop: ```...
Pre-installation, I have installed llvm and cmake both of which seem to be working fine. Not sure if this is relevant but i also have a separate install of tvm...
Pre-installation, I have installed llvm and cmake both of which seem to be working fine. Whilst attempting to install torch_tvm the error we keep getting is the following: please note...
Hi all, I'm trying to pass a torch model I trained to TVM. It's a RNN that takes a PackedSequence as an input. `torch_tvm.to_relay` only seems to like tensors as...
Hi, I'm trying to compile maskrcnn-benchmark with pytorch/tvm (directly from the master branch), and even though I know there are currently several operators that are yet to be implemented, I...
Currently, the activation inputs and weight inputs are set together after compilation. For pre-compute pass in TVM, we need to know the weight inputs to do some optimization, such as...
Currently input is setup by a vector in which Value* are pushed. And we call set_input on graph to set input at particular index This makes assumption that the index...
More often than not, we need to "test" single TVM operator, which cannot be fused with other nodes so far. We expect the fusing pass still create one group for...
Hi, for fusing a not that much longer elementary pointwise computation, I get ``` with tvm::CompilationGroup_0 = graph(%0 : Float(*), %1 : Float(*)): %4 : int = prim::Constant[value=1]() %2 :...
