Arch D. Robison

It seems to be useful for dealing with batch dimensions that might be zero. For example, reshaping from [n,a,b] to [0,-1] where n is the batch dimension. We (Nvidia TensorRT...

The install did succeed after I installed MKL.

Just an aside: I noticed the network is using what TensorRT calls "zero as placeholder", which indicates the original ONNX file is not setting the [attribute "allowzero=1" for Reshape](https://onnx.ai/onnx/operators/onnx__Reshape.html). When...

There is an error in TensorRT that affects attempts to use `IFillLayer` with mode `kRANDOM_UNIFORM` or `kRANDOM_NORMAL` to construct a shape tensor. The mistake in TensorRT was that one part...

Thanks for the news. You're the first person I know to get it running on Linux. Okay if I add your notes to the README? - Arch On Wed, Aug...

`@simd` is for exploiting vector arithmetic units, not multiple threads. It's currently limited to a single loop. Though with PR JuliaLang/julia#9876 you can use it on Cartesian ranges, albeit it...

Thinking about this revealed how fuzzy I am on the module system. Modules seem to be objects. For example, `a==b` works for two modules a and b. But if I...