'2019/12'에 해당되는 글 44건

  1. 2019.12.16 Here's an ultimate data science starter kit: 1. Foundational Skills • Intro to
  2. 2019.12.16 안녕하세요! TensorFlow-KR 논문읽기 모임 #PR12 의 214번째 논문은 2015년 ICCV에서 발표된 FlowNet: Learnin
  3. 2019.12.11 ICYMI: Beautify your face using a GAN-based architecture! https://www.profillic.com/paper/arxiv:1912.03630
  4. 2019.12.11 fast.ai 수강생이 진행한 Top6 프로젝트의 리스트 입니다. 좋아요를 많이 받은 순서인것 같네요. 노트북과 설명이 함께 첨부되어 있습니다 :) 1. segmentation & classification of buildings from drone/aerial imag..
  5. 2019.12.11 Generate 3D Avatars from a Single Image! Wide range of applications from virtual/augmented reality (VR/AR) and telepsychiatry to human-computer interaction and social networks. https://www.profillic.com/paper/arxiv:1912.03455
  6. 2019.12.10 #TFKRNeurIPS 안녕하세요! 오늘 NeurIPS 학회에서 제가 가장 관심있었던 주제는 <Efficient Processing of Deep Neural Network: from Algorithms to Hardware Architectures> 입니다! 저는 주로 네트워크 경량화쪽 기술에 ..
  7. 2019.12.10 안녕하세요! facebookresearch의 Detectron2[1]의 한국어버전 Colab 튜토리얼[2]을 공유합니다. Detectron2은 PyTorch기반의 Object Detection API입니다. Object Detection 하면 Bounding Box Regression 테스크를 많이들 떠..
  8. 2019.12.05 Our paper “Keratinocytic Skin Cancer Detection on the Face using Region-based Convolutional Neural Network” was published on JAMA Dermatology. To my knowledge, the performance of cancer detection was compared with that of dermatologists for the firs..
  9. 2019.12.04 #DEVIEW2019 #발표영상공개 DEVIEW 2019 모든 발표영상이 공개되었습니다. 지금 DEVIEW 홈페이지에서 확인하세요. ▶ DEVIEW 2019 발표 영상|deview.kr/2019/schedule
  10. 2019.12.02 Hello. What Data Science courses would you recommend for a beginner? Thanks.
  11. 2019.12.02 We just released our #NeurIPS2019 Multimodal Model-Agnostic Meta-Learning (MMAML) code for learning few-shot image classification, which extends MAML to multimodal task distributions (e.g. learning from multiple datasets). The code contains #PyTorch imp..
  12. 2019.12.01 [온라인 무료 강의] R로 하는 텍스트 전처리( 박찬엽 SK텔레콤 / T아카데미) 학습내용 1. 단정한 데이터란 무엇인지, 텍스트 데이터에서는 어떻게 접목되는지 이해한다. 2. 한글 데이터 분석에 ..
  13. 2019.12.01 Generate photorealistic facial images under new viewpoints or illumination conditions using this! https://www.profillic.com/paper/arxiv:1911.11999
  14. 2019.12.01 스텐포드 딥러닝 수업이 정말 많네요. 이번학기 새롭게 업데이트된 자료와 코스도 많으니 추운날 방에서 보고 있으면 이번 겨울이 빠르게 지날것 같습니다. 모두 딥러닝/AI와 함께 따뜻한 겨..

Here's an ultimate data science starter kit:

1. Foundational Skills

• Intro to Python - https://lnkd.in/grCsv8v

• Intro to R - https://lnkd.in/gKFiDZn

• Data Wrangling Pydata (90min) - https://lnkd.in/gEhF3-W

• EDA (20min video) - https://lnkd.in/gT8_RKh

• Stats/Prob (Khan Academy) - https://lnkd.in/gsyGpVu

2. Technical Skills

• Data Gathering- Why API Medium https://lnkd.in/gvahtsN

• Intro to SQL: https://lnkd.in/giWs-3N

• Complete SQL Bootcamp: https://lnkd.in/gsgf_fF

• Data Visualization - Medium https://lnkd.in/g3FSRgY

• Machine Learning A-Z: https://lnkd.in/gXqdBsQ

3. Business Skills

• Communication - Data Storytelling https://lnkd.in/gtiCSNT

• Business Analytics- Geckoboard https://lnkd.in/g2X-Xtp

4. Extra Skills

• Natural Language Processing - How to solve 90% of NLP https://lnkd.in/gh8bKe4

• Recommendation Systems - How Spotify Knows You So Well https://lnkd.in/gH2GQKu

• Time Series Analysis - Complete Time Series https://lnkd.in/gFZU2Rb

5. Practice

• Projects/Competitions - Kaggle Kernels https://www.kaggle.com/

• Problem Solving Challenges - HackerRank https://lnkd.in/g9Ps2cb

- - -

Hope these resources help!

If you want more free DS/ML resources, feel free to check out my site: www.ClaoudML.com

Happy learning! #datascience #machinelearning

Posted by uniqueone
,

안녕하세요! TensorFlow-KR 논문읽기 모임 #PR12 의 214번째 논문은 2015년 ICCV에서 발표된 FlowNet: Learning Optical Flow with Convolutional Networks라는 논문입니다. Optical Flow는 비디오의 인접한 Frame에 대하여 각 Pixel이 첫 번째 Frame에서 두 번째 Frame으로 얼마나 이동했는지의 Vector를 모든 위치에 대하여 나타낸 Map입니다. Video에 Motion을 분석하는 일은 매우 중요하기 때문에, 이러한 Optical Flow 역시 굉장히 중요한 요소 중 하나인데요, 이번 영상에서는 고전적인 Computer Vision에서 쓰였던 다양한 Optical Flow 알고리즘들과, Deep Learning Based로 Optical Flow를 구하는 Neural Network인 FlowNet에 대하여 정리해 봤습니다. 감사합니다!! :)

Youtube link: [https://www.youtube.com/watch?v=Z_t0shK98pM](https://www.youtube.com/watch?v=Z_t0shK98pM)

Paper link : http://openaccess.thecvf.com/content_iccv_2015/html/Dosovitskiy_FlowNet_Learning_Optical_ICCV_2015_paper.html

Slide link : https://www.slideshare.net/HyeongminLee3/pr213-flownet-learning-optical-flow-with-convolutional-networks

Posted by uniqueone
,
ICYMI: Beautify your face using a GAN-based architecture!

https://www.profillic.com/paper/arxiv:1912.03630
https://www.facebook.com/groups/DeepNetGroup/permalink/1028034027589479/?sfnsn=mo
Posted by uniqueone
,
fast.ai 수강생이 진행한 Top6 프로젝트의 리스트 입니다. 좋아요를 많이 받은 순서인것 같네요. 노트북과 설명이 함께 첨부되어 있습니다 :)

1. segmentation & classification of buildings from drone/aerial imagery in Zanzibar, Tanzania
https://forums.fast.ai/t/share-your-work-here/27676/577

2. Time series classification: General Transfer Learning with Convolutional Neural Networks
https://forums.fast.ai/t/share-your-work-here/27676/367

3. plotting the loss function and the path followed by SGD
https://forums.fast.ai/t/share-your-work-here/27676/300

4. Urban Sounds database Converted de soundfiles (wav) into spectrograms
https://forums.fast.ai/t/share-your-work-here/27676/61

5. 10 class audio classification task
https://forums.fast.ai/t/share-your-work-here/27676/39

6. applying PCA to the last layer before the predictions => interpreted the top two features as ‘naked/hairy’ and ‘dog/cat’
https://forums.fast.ai/t/share-your-work-here/27676/26
https://www.facebook.com/groups/fastaikr/permalink/2477649515783828/?sfnsn=mo
Posted by uniqueone
,
Generate 3D Avatars from a Single Image! Wide range of applications from virtual/augmented reality (VR/AR) and telepsychiatry to human-computer interaction and social networks.

https://www.profillic.com/paper/arxiv:1912.03455
https://www.facebook.com/groups/TensorFlowKR/permalink/1059469937727397/?sfnsn=mo
Posted by uniqueone
,
#TFKRNeurIPS

안녕하세요! 오늘 NeurIPS 학회에서 제가 가장 관심있었던 주제는

<Efficient Processing of Deep Neural Network: from Algorithms to Hardware Architectures>

입니다!

저는 주로 네트워크 경량화쪽 기술에 관심이 많은데, MIT에 계시는 Vivienne Sze 교수님께서 가장 기초적인 부분부터 최근 연구까지 하나하나 차근차근 설명해주시는 세션이었습니다. 특히, 저는 하드웨어 관점에서는 많이 생각을 해보지는 못했는데 하드웨어 관점에서 여러 논문들을 소개해주셔서 도움이 되었습니다.

공식 슬라이드는

http://eyeriss.mit.edu/2019_neurips_tutorial.pdf

이고,

벌써 비디오로도 볼 수 있는 것 같네요

비디오 링크: https://slideslive.com/38921492/efficient-processing-of-deep-neural-network-from-algorithms-to-hardware-architectures

(제가 찍은 사진에는 참석자가 없는 것 처럼 보이지만, 참석자가 사실 엄청 많았습니다!!)
https://www.facebook.com/groups/TensorFlowKR/permalink/1060767684264289/?sfnsn=mo
Posted by uniqueone
,
안녕하세요!
facebookresearch의 Detectron2[1]의 한국어버전 Colab 튜토리얼[2]을 공유합니다.

Detectron2은 PyTorch기반의 Object Detection API입니다. Object Detection 하면 Bounding Box Regression 테스크를 많이들 떠올리시는데, 요즘은 Object Detection 하면 넓은 의미로 bounding box/keypoints detection, instance/semantic/panoptic segmentation 모두를 지칭하는 용어로 사용하기도 하는 것 같습니다.
Detectron2의 특징으로는
1. PyTorch 기반
2. 다양한 Object Detection 알고리즘 제공
3. 방대한 Pretrained Model Zoo 제공
4. 이를 활용할 수 있는 쉬운 API 환경을 제공
5. 아주 쉬운 커스텀 데이터셋 로더 만들기
(특히 bbox 표현 방식이 데이터셋별로 상이한 경우가 많은데, 4가지 bbox 버전을 제공해줘서 정말 편했습니다)
Detectron2 repo에 colab 튜토리얼 원문이 있는데 이번 한국어버전 튜토리얼의 차이점은.

+ 한국어 번역
+ 튜토리얼 목차 순서를 조금 더 직관적으로 변경
+ RetinaNet(bbox) 이미지/비디오 인퍼런스 튜토리얼 추가
+ RetinaNet(bbox) 커스텀 데이터셋 학습 튜토리얼 추가

으로 제 입맛대로 한번 바꿔봤습니다..
필요한 챕터만 골라서 이것저것 튜닝해보면 기초적인 부분은 쉽게 이해하

요즘 날씨가 정말 춥네요.. 모두들 연말 따뜻하게 보내세요!
항상 감사합니다!
 
REFERENCES
[1] https://github.com/facebookresearch/detectron2
[2] https://colab.research.google.com/github/visionNoob/detectron2_aihub_tutorial/blob/master/Detectron2_Tutorial_(kor_ver).ipynb
https://www.facebook.com/groups/TensorFlowKR/permalink/1059469937727397/?sfnsn=mo
Posted by uniqueone
,
Our paper “Keratinocytic Skin Cancer Detection on the Face using Region-based Convolutional Neural Network” was published on JAMA Dermatology. To my knowledge, the performance of cancer detection was compared with that of dermatologists for the first time in dermatology. Because most of previous studies were classification studies, preselection of end-user was essential. In addition, there were numerous false positives because training data set did not include enough number of common disorders and normal structures.
With the assistance of R-CNN, we trained neural networks with 1,106,886 image crops to localize and diagnose malignancy. The algorithm detects suspected lesion and shows malignancy score and predicts possible diagnosis (178 disease classes).
We used region-based CNN (faster-RCNN; backbone = VGG-16) as a region proposal module, and utilized CNN (SE-ResNet-50) to choose adequate lesion, and utilized CNN (SE-ResNeXt-50 + SENet) to determine malignancy. We chose a multi-step approach to reduce the dimension of problem (object detection -> classification).
The AUC for the validation dataset (2,844 images from 673 patients comprising 185 malignant, 305 benign, and 183 normal conditions) was 0.910. The algorithm’s F1 score and Youden index (sensitivity + specificity - 100%) were comparable with those of 13 dermatologists, while surpassing those of 20 non-dermatologists (325 images from 80 patients comprising 40 malignant, 20 benign, and 20 normal). We are performing an additional work with large scale external validation data set. The pilot result is similar with this report, so I hope I will submit soon.
Web DEMO (https://rcnn.modelderm.com) of the model is accessible via smartphone or PC, to facilitate scientific communication. Sorry for the slowness of the DEMO because it runs on my personal computer despite of the multi-threading and parallel processing with 2080 x1 and 1070 x1.
Thank you.
Paper : https://jamanetwork.com/journals/jamadermatology/article-abstract/2756346
Screenshot : https://i.imgur.com/2TCkdHf.png
Screenshot : https://i.imgur.com/IEZLfOg.jpg
DEMO : https://rcnn.modelderm.com
https://m.facebook.com/groups/107107546348803?view=permalink&id=1021762028216679&sfnsn=mo
Posted by uniqueone
,
#DEVIEW2019 #발표영상공개

DEVIEW 2019 모든 발표영상이 공개되었습니다.
지금 DEVIEW 홈페이지에서 확인하세요.

▶ DEVIEW 2019 발표 영상|deview.kr/2019/schedule
https://m.facebook.com/story.php?story_fbid=2364070777054935&id=353497304778969&sfnsn=mo
Posted by uniqueone
,
Hello. What Data Science courses would you recommend for a beginner?

https://www.facebook.com/groups/DataScienceGroup/permalink/2818646221530583/?sfnsn=mo

Statistics brother, Courseear also has a program they offer it is very useful also don’t get into data science if you are not willing to put in the work brother. Good luck

https://www.udemy.com/course/datascience/

Trust me this will help a lot, and yes its very cheap

https://www.coursera.org/specializations/introduction-data-science

http://hitech360.altervista.org/what-is-data-science-and-why-use-python/

Start Learning Statistics and Linear algebra and calclaus .. there are a lot of website teach that and khan academy is good ... then start with Data Science Course From Coursera this is very good course for beginners then take Machine Learning Course (andrew ng) from Coursera

You should definitely do the Korbit AI course. It is free + you get a certificate on completion.
The best part is that you learn from an AI tutor. It's the best for beginners like you.

http://bit.ly/2PvqOcW

You can try my video channel on YouTube (TechKnowHow). I am a data scientist for a Fortune global 200 company and this video channel has 100's of tutorial videos that are complete walkthroughs of real data science projects we use every day for our execs and directors. The channel is located here: https://YouTube.com/channel/UCwgcmcn_iifLGs_38JIF6kw

I've only started learning a few months ago But, DataQuest.io was a huge help for me grasping python. And once you get comfortable with the python concepts Kaggle offers a much more intuitive learning approach through Jupyter Notebooks. Also if you do a Youtube search for "Data Science Python" There are several 6+ hour walkthrough videos. Youtube will also usually have resources for anything you get stuck on.










Posted by uniqueone
,
We just released our #NeurIPS2019 Multimodal Model-Agnostic Meta-Learning (MMAML) code for learning few-shot image classification, which extends MAML to multimodal task distributions (e.g. learning from multiple datasets). The code contains #PyTorch implementations of our model and two baselines (MAML and Multi-MAML) as well as the scripts to evaluate these models to five popular few-shot learning datasets: Omniglot, Mini-ImageNet, FC100 (CIFAR100), CUB-200-2011, and FGVC-Aircraft.

Code: https://github.com/shaohua0116/MMAML-Classification

Paper: https://arxiv.org/abs/1910.13616

#NeurIPS #MachineLearning #ML #code
Posted by uniqueone
,
[온라인 무료 강의] R로 하는 텍스트 전처리( 박찬엽 SK텔레콤 / T아카데미)

학습내용
1. 단정한 데이터란 무엇인지, 텍스트 데이터에서는 어떻게 접목되는지 이해한다.
2. 한글 데이터 분석에 필요한 Rmecabko / KoLNP 사용법을 알아보고, 한글 데이터 전처리 방법을 알아본다.

<학습대상>
R 프로그래밍이 가능하며, stringr 패키지와 정규표현식에 대하 이해가 있으신 분

<강의목록>
[1강] Tidyverse I - 파이프연산자(%/%), dplyr
[2강] Tidyverse II - tidy data, tidy text
[3강] 형태소분석 패키지 설치 - KoNLP, RmecanKo
[4강] 형태소분석 패키지 사용실습 - Token화, 불용어 제거, 정규표현식
[5강] 정량 지표 I - 단순출현빈도, 동시출현빈도
[6강] 정량 지표 II - tf-idf, 감성분석

* 박찬엽 선생님 github : https://mrchypark.github.io/
* 출처 : https://tacademy.skplanet.com/live/player/onlineLectureDetail.action?seq=166
https://www.facebook.com/113979985329905/posts/2664110800316798/?sfnsn=mo
Posted by uniqueone
,
Generate photorealistic facial images under new viewpoints or illumination conditions using this!

https://www.facebook.com/groups/DeepNetGroup/permalink/1018236828569199/?sfnsn=mo
Posted by uniqueone
,
스텐포드 딥러닝 수업이 정말 많네요. 이번학기 새롭게 업데이트된 자료와 코스도 많으니 추운날 방에서  보고 있으면 이번 겨울이 빠르게 지날것 같습니다. 모두 딥러닝/AI와 함께 따뜻한 겨울 되기실!

Deep Learning

[http://web.stanford.edu/class/cs230/](http://web.stanford.edu/class/cs230/)

[ Natural Language Processing ]

CS 124: From Languages to Information (LINGUIST 180, LINGUIST 280)

[http://web.stanford.edu/class/cs124/](http://web.stanford.edu/class/cs124/)

CS 224N: Natural Language Processing with Deep Learning (LINGUIST 284)

[http://web.stanford.edu/class/cs224n/](http://web.stanford.edu/class/cs224n/)

CS 224U: Natural Language Understanding (LINGUIST 188, LINGUIST 288)

[http://web.stanford.edu/class/cs224u/](http://web.stanford.edu/class/cs224u/)

CS 276: Information Retrieval and Web Search (LINGUIST 286)

[http://web.stanford.edu/class/cs](http://web.stanford.edu/class/cs224u/)276

[ Computer Vision ]
CS 131: Computer Vision: Foundations and Applications

http://[cs131.stanford.edu](http://cs131.stanford.edu/)

CS 205L: Continuous Mathematical Methods with an Emphasis on Machine Learning

[http://web.stanford.edu/class/cs205l/](http://web.stanford.edu/class/cs205l/)

CS 231N: Convolutional Neural Networks for Visual Recognition

[http://cs231n.stanford.edu/](http://cs231n.stanford.edu/)

CS 348K: Visual Computing Systems

[http://graphics.stanford.edu/courses/cs348v-18-winter/](http://graphics.stanford.edu/courses/cs348v-18-winter/)

[ Others ]

CS224W: Machine Learning with Graphs([Yong Dam Kim](https://www.facebook.com/yongdam.kim) )

[http://web.stanford.edu/class/cs224w/](http://web.stanford.edu/class/cs224w/)

 
CS 273B: Deep Learning in Genomics and Biomedicine (BIODS 237, BIOMEDIN 273B, GENE 236)

[https://canvas.stanford.edu/courses/51037](https://canvas.stanford.edu/courses/51037)

CS 236: Deep Generative Models

[https://deepgenerativemodels.github.io/](https://deepgenerativemodels.github.io/)

CS 228: Probabilistic Graphical Models: Principles and Techniques

[https://cs228.stanford.edu/](https://cs228.stanford.edu/)

CS 337: Al-Assisted Care (MED 277)

[http://cs337.stanford.edu/](http://cs337.stanford.edu/)

CS 229: Machine Learning (STATS 229)

[http://cs229.stanford.edu/](http://cs229.stanford.edu/)

CS 229A: Applied Machine Learning

[https://cs229a.stanford.edu](https://cs229a.stanford.edu/)

CS 234: Reinforcement Learning

http://[s234.stanford.edu](http://cs234.stanford.edu/)

CS 221: Artificial Intelligence: Principles and Techniques

[https://stanford-cs221.github.io/autumn2019/](https://stanford-cs221.github.io/autumn2019/)
https://m.facebook.com/groups/255834461424286?view=permalink&id=1051374671870257&sfnsn=mo
Posted by uniqueone
,