glow
glow copied to clipboard
pytorch
Compiler for Neural Network hardware accelerators
I am getting the following error while i am compiling glow ``` /Users/dhruva_macintosh/Calligo/glow/lib/Graph/Nodes.cpp:2252:14: error: no matching function for call to 'expectCompareTrue' isValid &= expectCompareTrue( ^~~~~~~~~~~~~~~~~ /Users/dhruva_macintosh/Calligo/glow/include/glow/Graph/VerifierHelper.h:126:6: note: candidate template ignored:...
Summary: * Glow lowering * loadSlice: allow GlowIValueNode and 4-argument version * loadGetItem: newly added to support aten::__getitem__ * loadSub: allow two GlowIValueNodes as inputs * loadLen: to support aten::len...
After executing `cmake -G Ninja -DCMAKE_BUILD_TYPE=Debug ../glow`, I encountered cmake errors related to pthread and stdc++. I guess they are due to the version of clang, but I haved done...
Hi there, I benchmarked resnet50 in onnx format against onnxruntime. I found out on an Azuredsv3 server that the inference ran in 200ms for glow with AOT against 50ms for...
Due to the read.me , Glow can only be build on MacOs and Linuxs.
Hi, I would like to build a Glow on my Windows x64 machine. For now, I’m working inside a virtual machine. I went to https://github.com/pytorch/glow 1 and I followed **docs/GlowWindowsBuildx86.md**....
Summary: Replace concat with direct partial writes from concat feeders into concat output tensor. If everything matches shapes, strides for all inputs and output of concat, remove concat. Calculate offsets...
Summary: Add offset in tensor for tensor view init. Replace concat with direct partial writes from concat feeders into concat output tensor. If everything matches shapes, strides for all inputs...
I have successfully compiled the Glow C++ API. And I want to test the glow in the python language. That is, I want to use **import torch_glow**. When I go...
Summary: When saving JSON and tensor views are used, do not save shape and stride in compute nodes Differential Revision: D37606431