-
### Search before asking
- [X] I have searched the Ultralytics YOLO [issues](https://github.com/ultralytics/ultralytics/issues) and [discussions](https://github.com/ultralytics/ultralytics/discussion…
-
Convolutions provided by the [`FastConv` package](https://github.com/aamini/FastConv.jl)
Described in [their paper](https://arxiv.org/pdf/1612.08825.pdf) is considerably outperforming the back ends…
-
There are two things that could be improved about the way wavelengths are currently handled.
1. Log-uniform wavelengths (uniform velocity steps) provide a better sampling for broad wavelength range…
-
Depthwise convolutions, which are currently implemented as a standard `Conv` layer with the number of groups equal to the number of input channels, seem to produce a very large number of allocations c…
-
Hi all,
I am urgently seeking to deploy the TFLite models converted using Larq Compute Engine (LCE) on an ARM32 device, specifically a Cortex-M7 CPU, the STM32F7 series MCU.
I have seen some rel…
-
### 🐛 Describe the bug
On running torchaudio code I noticed that some resampling operations are slower than they should be on the forward pass of the Resample transform. I tracked the slowness to th…
-
The benchmarks this time around are interesting, with some fairly clear trends emerging for the near future.
### Looking Back
First, some appreciation for where things are,
- 9 months ago, we were ~3…
-
### Description
Transpose convolutions are orders of magnitude slower than their complementary regular convolutions and their counterparts in torch (at least for the sizes in the example below). Thi…
-
Implement Fast Fourier Transformations for convolutions
-
TFLite uses int8 per-channel weight quantization for transposed convolutions.
While XNNPACK includes a fast transposed convolution operation it only supports per-tensor weight quantization (i.e. a si…