@creaf

Deep Learning | Computer Vision Co-creator of #albumentations.

github.com/creafz
Joined March 2008
  • Tweets39
  • Following112
  • Followers260
1Media
New release of the best image augmentations library Albumentations - 1.2.0 - Four new transforms - Improved key points support - Improved documentation - Various bug fixes Stats: - 280k downloads per month - 10.4k stars at GitHub. More details: github.com/albumentations-te…
3
13
Post about #albumentations #OpenSource library. I talk about how it was born from @kaggle competitions, how it evolved, and how we promoted it among developers and scientists. medium.com/@iglovikov/the-bi…

The birth of Albumentations

I would like to talk about Albumentations, an open-source library for image Augmentations. An explanation of how it was born and evolved…

medium.com
19
114
#Albumentations 1.1.0 is out! The new release of a fast and flexible library for image augmentation includes new transformations and improvements for the current ones. Thanks to @Dipetm @viglovikov @cvtalks @AlBuslaev and all our contributors. github.com/albumentations-te…
1
6
1
12
#Albumentations 1.0.0 has been released! New version contains: - 10 new transforms - bug fixes etc See the release notes for details github.com/albumentations-te… Thanks a lot to core team @viglovikov @creaf @cvtalks @AlBuslaev and everyone who helps improve the library
1
23
2
77
Fresh Martian Chronicles are out: a hands-on introduction to a Catalyst framework for #deeplearning by @gazay. Learn how to build your own pipeline for an image classifier and deploy a trained model to Heroku “Beyond Fashion: Deep learning with Catalyst”: amp.gs/JNqB
13
2
37
Another nice contribution to our "Machine Learning with Python" special issue just got published: "Albumentations: Fast and Flexible Image Augmentations" by @alxndrkalinin et al. Paper link: mdpi.com/2078-2489/11/2/125 GitHub Repo: github.com/albumentations-te…
30
1
124
Our peer-reviewed open-access Albumentations paper has been published in the special issue Machine Learning in Python @InformationMDPI: mdpi.com/2078-2489/11/2/125 It covers the design considerations, main features, performance, & adoption of Albumentations albumentations.ai
1
16
44
Now we have an official website for #Albumentations! Check this out at albumentations.ai Big thanks to @creaf for making it real! #DeepLearning #ArtificialIntelligence
3
58
3
203
A fresh release of pytorch-toolbelt 0.3 is out: Changelog: github.com/BloodAxe/pytorch-… If you're unfamiliar - it's a python library with a lot of bells and whistles for PyTorch for fast R&D prototyping and Kaggle farming: linkedin.com/posts/cvtalks_b…

Release Pytorch Toolbel 0.3.0 · BloodAxe/pytorch-toolbelt

Pytorch Toolbel 0.3.0 This release has a huge set of new features, bugfixes and breaking changes. So be careful, when upgrading. pip install pytorch-toolbelt==0.3.0 New features Encoders HRNetV2 D...

github.com
1
10
35
On my new year flight from Lima to SF, I wrote a blog post on the path from my previous job to the current one. TL;DR => it was an ocean of pain and experience obtained at @kaggle was very useful 😀 If you like it => 50 claps. If not => 49 😀 medium.com/@iglovikov/how-i-…

How I found my current job

From the debt collection agency to autonomous vehicles

medium.com
20
69
15
516
I am thrilled that image Augmentation library #albumentations that was born out of @kaggle competitions by @AlBuslaev @creaf @cvtalks @Dipetm and @viglovikov became a part of the @PyTorch ecosystem! pytorch.org/ecosystem

PyTorch Landscape

landscape.pytorch.org
4
95
5
413
datafuturology.com/podcast/2… My interview with the Data Futurology podcast.
14
1
33
Catalyst, high-level utils for @PyTorch DL & RL research. Release 19.05 thread PS. 600 starts on Github - done 🚀
1
10
32
We have finally released the source code and pre-trained models of our recent work "DGC-Net: Dense Geometric Correspondence network"🎉 Paper: arxiv.org/abs/1810.08393 Github: github.com/AaltoVision/DGC-N… Project page: aaltovision.github.io/dgc-ne… @tiulpin @mapo1 @PyTorch @CSAalto
1
17
61
Hell yeah, We have released catalyst 19.03 final version! - tests, lots of new tests for train/infer pipeline validation - registry refactoring for simplified customisation - stablelized API - minor improvements github.com/catalyst-team/cat…

GitHub - catalyst-team/catalyst: Accelerated deep learning R&D

Accelerated deep learning R&D. Contribute to catalyst-team/catalyst development by creating an account on GitHub.

github.com
4
15
In #albumentations v0.2.2 @AlBuslaev @creaf @cvtalks and @viglovikov added experimental support for weather transforms from the #automold library. Let us know if they help or you have ideas on how to improve them.
23
3
70
Are your interested in reproducible RL? Or want a competitive benchmark of current off-policy RL algorithms? check out arxiv.org/abs/1903.00027, catalyst.rl – framework for distributed RL training on top of @PyTorch Various RL algorithms and auxiliary tricks included

Catalyst.RL: A Distributed Framework for Reproducible RL Research

Despite the recent progress in deep reinforcement learning field (RL), and, arguably because of it, a large body of work remains to be done in reproducing and carefully comparing different RL...

arxiv.org
2
33
1
101
The Interview with @kaggle Grandmaster and Senior CV Engineer @LyftLevel5: Vladimir Iglovikov @viglovikov just got published @hackernoon. The Grandmaster has really been kind enough to share *ALL* of his secrets, you can find all of them here: hackernoon.com/interview-wit…

Interview with Kaggle Grandmaster, Senior CV Engineer at Lyft: Dr. Vladimir I. Iglovikov | Hacker...

Today, I’m honored to be talking to another great kaggler from the ODS community: (kaggle: iglovikov) Competitions Grandmaster (Ranked #97), Discussions Expert (Ranked #30): Dr. Vladimir I. Iglovikov

hackernoon.com
8
84
2
355
creaf retweeted
Kaggle
@kaggle12 Mar 2019
Bookmark this amazing library of image augmentations 😵 by #Kaggle Masters @AlBuslaev @creaf @viglovikov and @cvtalks | github.com/albu/albumentatio… #resources #machinelearning
2
57
2
214
Catalyst.dl - high-level utils for @Pytorch DL research v19.03 You get a training loop with metrics, early-stopping, model checkpointing and other features without the boilerplate. Break the cycle - use the Catalyst! github.com/catalyst-team/cat…
1
16
37
Load more