Michael Voronov

Certain visual problems

1

Hi, I just tried the script now and I noticed several visual glitches. 1) When I switch desktops using my keyboard shortcuts, my panels (standard panel and Latte dock) appear...

Alternatives?

7

I really need something to send emails with Flask and I really don't want to program it from scratch. This package doesn't seem to be maintained. Are there any other...

Use Neural-Fortran

4

I have a suggestion: use https://github.com/modern-fortran/neural-fortran. I have implemented MultiHead Attention, LayerNorm and Embedding there. I think, the most sensible thing would be to combine the effort and use that...

Two-Byte Floating Point (half, float16) support for Nvidia GeForce and Quadro

6

## Current Behavior `CLBlastNoHalfPrecision` when trying to use CLBlast on Nvidia GeForce 4060 graphics card. ## Problem Proprietary Nvidia driver does not declare support for `cl_khr_fp16` and yet appears to...

linear2d_layer_no_bias: add optional argument to disable biases

8

### Add option to disable biases Simple additional option. Removing biases helps speed up inference. This technique is used in Llama and, to an extent, in Qwen. Being able to...

Fc2d layer

7

## Fully-Connected Layer for 2D Shapes Also known as MLP, FeedForward, etc. A common component of neural networks, including transformers. The idea is very simple: first linear transformation => activation...

Weight initialization

3

# Weights Initialization Added functions for Xavier and Kaiming. The rule of thumb here: * S-shaped activation (`tanh`, `sigmoid`, etc.) => Xavier * ReLU-shaped activation (`relu`, `gelu`, `silu`, etc.) =>...