276K subscribers
3.94K photos
674 videos
17 files
4.53K links
Погружаемся в машинное обучение и Data Science

Показываем как запускать любые LLm на пальцах.

По всем вопросам - @haarrp

@itchannels_telegram -🔥best channels

Реестр РКН: clck.ru/3Fmqri
Download Telegram
Two Sigma: Using News to Predict Stock Movements

Use news analytics to predict stock price performance

1st place - $25,000
2nd place - $20,000
3rd place - $15,000
4th through 7th place - $10,000 each

This is a two-stage competition consisting of a Submission period and a Scoring period. In the Submission period, entrants will train their models in Kaggle Kernels. During the Scoring period, models submitted at the end of the submission period will be evaluated against regularly updated news and market data.

Start Date: 9/25/2018
Rules Acceptance/Team Merger Deadline: 1/2/2019
Submission Deadline: 1/8/2019 After this date, we will not be taking any more submissions. You can sit back and watch the leaderboard unfold. Remember to select your two best submissions to be rescored. In this competition we will not auto-select your two submissions.
End Date: 7/15/2019

https://www.kaggle.com/c/two-sigma-financial-news
Brilliant post on #CS and #Software about strategy and psychology of Software Development, which is highly applicable to Data Science too.

“Imaginary Problems Are the Root of Bad Software”

https://medium.com/s/story/imaginary-problems-d4f2921bd1b8
Необычные материалы в робототехнике.

У каждого робота должен быть блестящий металлический зад, это знают все. Но в реальных роботах металла не так уж и много — с ним соседствуют пластики, композиты и силикон, а порой и совсем нестандартные субстанции: http://amp.gs/hCG8
How to train an algorithm to successfully pass the "Sonic The Hedgehog" game? Sergey Kolesnikov together with his team took the 4th place out of 900+ in the Open AI contest, and now is telling how to reach it. https://medium.com/swlh/at-the-speed-of-reinforcement-learning-an-openai-contest-story-6ed34fe7a3bb
mmdetection

mmdetection is an open source object detection toolbox based on PyTorch. It is a part of the open-mmlab project developed by Multimedia Laboratory, CUHK.

Major features
- Modular Design
One can easily construct a customized object detection framework by combining different components.
- Support of multiple frameworks out of box
The toolbox directly supports popular detection frameworks, e.g. Faster RCNN, Mask RCNN, RetinaNet, etc.
- Efficient
All basic bbox and mask operations run on GPUs now. The training speed is about 5% ~ 20% faster than Detectron for different models.
- State of the art
This was the codebase of the MMDet team, who won the COCO Detection 2018 challenge.

https://github.com/open-mmlab/mmdetection