-
NUST MISIS / SBER AI / AIRI
- Moscow, Russia
-
01:22
(UTC +03:00) - levnovitskiy@gmail.com
- https://t.me/leffffffffffff
- in/lev-novitskiy-022289261
- https://t.me/mlball_days
Highlights
- Pro
Block or Report
Block or report leffff
Contact GitHub support about this user’s behavior. Learn more about reporting abuse.
Report abuse-
euler-scheduler Public
My implementation Diffusers-like Scheduler for performing Euler Method on Conditional Flow Matching models
-
InstructFlow Public
The aim of this repository is to create a Flow-Matching-based model capable of doing text2img, solving inverse problems: jpeg restoration, debluring, denoising, superres, coloring, inpainting, outp…
-
diffusers Public
Forked from huggingface/diffusers🤗 Diffusers: State-of-the-art diffusion models for image and audio generation in PyTorch and FLAX.
-
dragon-diffusion-kandinsky3 Public
My implementation of object moving from DragonDiffusion https://arxiv.org/pdf/2307.02421.pdf
-
-
nsp-response-ranker Public
A model for response quality classification
-
-
My solutions to tasks in NUST MISIS on data analysis
-
-
My Implementation of Adversarial Diffusion Distillation https://arxiv.org/pdf/2311.17042.pdf
-
graphormer-pyg Public
Microsoft Graphormer (https://arxiv.org/abs/2106.05234) rewritten in Pytorch-Geometric
-
-
ml-talent-match Public
Our solution to ML Talent Match hackathon
-
diffusion-project Public
My pet project on Diffusion Models
-
-
-
improved-diffusion-for-physics Public
Forked from openai/improved-diffusionRelease for Improved Denoising Diffusion Probabilistic Models
-
My South Park character generator
-
-
-
sber-beautifulcode-challenge Public
My Solution to SBER Beautiful code challenge
-
RetNet Public
Forked from Jamie-Stirling/RetNetAn implementation of "Retentive Network: A Successor to Transformer for Large Language Models"
Python MIT License UpdatedSep 13, 2023 -
-
-
vgae-pytorch Public
My Vraiational Graph Auto Encoder
-
point-cloud-project Public
My Point Cloud Project
-
any-domain-pretrain-gnns Public
Forked from snap-stanford/pretrain-gnnsStrategies for Pre-training Graph Neural Networks for Any domain
-
-
-
fast-ensemble Public
Library for high level model ensembling