Skip to content

lampts/data_science

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

data_science

seeing is believing. A witty saying proves nothing.

"When solving a problem of interest, do not solve a more general problem as an intermediate step." (Vladimir Vapnik)

Must read

My implementations

Chatbot

RecSys

Winining solutions

Stats

  • Good, Hardin. Common Errors in Statistics (and How to Avoid Them) (2003)
  • Kanji. 100 statistical tests (2006)
  • Doing Data Science: Straight Talk from the Frontline

Game Industry:

Case stydies:

DS Coursera

Heroes of DL

Top conferences:

Deep Learning

Events: I will put word cloud for that.

EMNLP 2017: http://noisy-text.github.io/2017/

NLPStan reading

LXMLS16:

ACL2017

VietAI

My SOTA

  • My ATIS: sequence tagging, nb of params: 324335, bi-LSTM
  • Quore question duplicate detection: Accuracy 85% on Wang's test
 - best F1 score: 94.92/94.64
 - train scores: 97.5446666667/96.17
 - val scores: 93.664/92.94

Game industry

Yandex

ICLR 2017 Review

LearningNewThingIn2017

Conf events

NIPs 2016 slides

Theano based DL applications

learn to learn: algos optimization

People

Pin:

Data type: NOQ

  • Nominal (N):cat, dog --> x,o | vis: shape, color
  • Ordinal (O): Jan - Feb - Mar - Apr | vis: area, density
  • Quantitative (Q): numerical 0.42, 0.58 | vis: length, position

People:

Fin data:

Projects:

Wikidata:

Cartoons & Quotes:

Books:

Done:

  1. EMNLP 2016, Austin, 2-4 Nov: http://www.emnlp2016.net/tutorials.html#practical

day 1:

  • Hugo(Twitter): Feed forward NN
  • Kartpathy(OpenAI): Convnet
  • Socher(MetaMind): NLP = word2vec/glove + GRU + MemNet
  • Tensorflow tut: from 5:55:49
  • Ruslan: Deep Unsup Learning: from 7:10:39
  • Andrew Ng: Nuts and bolts in applied DL from 9:09:46

day 2:

AI mistakes:

Keras:

NLP:

Apps:

German word embedding:

PyGotham:

Journalist LDA and ML:

Europython:

Scipy 2016:

Performance Evaluation(PE):

Hypothesis testing

Metrics:

Rock, Metal and NLP:

Financial:

Twitter:

Deep Learning Frameworks/Toolkits:

  • Tensorflow
  • Torch
  • Theano
  • Keras
  • Dynet
  • CNTK

ElasticSearch + Kibana:

Attention based:

ResNet: Residual Networks

Sentiment

NER

ML Stacking

Tensorflow tutorials

Covariate shift

#PydataLondon2017

NLP course

Dataset

Tricks of DL

Pointer network

Attention

Log likelihood test


MLtrainings.ru

GCloud

Current conference

https://github.com/aymericdamien/TensorFlow-Examples

Timeline

WSDM 2019

Computer Vision

ICCV 2019

07.10

13.06

04.06

18.05

17.05

14.05

13.05

08.05

07.05

03.05

28.04

24.04

19.04

10.04

09.04

08.04

05.04

03.04

01.04

31.03

30.03

29.03

28.03

21.03

20.03

14.03

11.03

07.03

06.03

01.03

21.02

20.02

19.02

13.02

12.02

11.02

09.02

03.02

24.01

21.01

18.01

16.01

14.01

03.01

02.01

===== GOODBYE 2018

29.12

25.12

22.12

20.12

19.12

18.12

17.12

12.12

10.12

09.12

-https://hai.stanford.edu/news/the_intertwined_quest_for_understanding_biological_intelligence_and_creating_artificial_intelligence/

07.12

06.12

04.12

02.12

01.12

29.11

26.11

BERT with <3

20.11

15.11

14.11

13.11

12.11

10.11

08.11

07.11

06.11

04.11

01.11

29.10

25.10

23.10

18.10

16.10

10.10

09.10

08.10

03.10

02.10

29.09

27.09

26.09

25.09

24.09

21.09

20.09

19.09

18.09

16.09

13.09

11.09

08.09

07.09

04.09

28.08

27.08

23.08

22.08

21.08

20.08

18.08

17.08

16.08

15.08

14.08

13.08

10.08

08.08

07.08

06.08

03.08

01.08

30.7

27.7

26.07

24.07

20.07

17.07

15.07

14.07

11.07

10.07

05.07

04.07

29.06

28.06

26.06

25.06

22.06

21.06

20.06

19.06

18.06

15.06

14.06

12.06

11.06

09.06

08.06

07.06

06.06

05.06

04.06

02.06

01.06

29.05

28.05

26.05

25.05

24.05

23.05

22.05

21.05

18.05

17.05

15.05

14.05

13.05

10.05

09.05

08.05

07.05

02.05

01.05

30.04

29.04

28.04

24.04

23.04

20.04

19.04

18.04

15.04

10.04

09.04

06.04

05.04

04.04

02.04

01.04

churn:

repeat purchase:

31.03

30.03

28.03

27.03

26.03

24.03

23.03

22.03

21.03

20.03

19.03

18.03

16.03

12.03

08.03

07.03

05.03

04.03

01.03

28.02

27.02

26.02

21.02

20.02

13.02

09.02

07.02

06.02:

05.02

02.02

01.02

31.01

30.01

29.01

26.01

25.01

22.01

20.01

19.01

18.01

17.01

15.01

12.01

11.01

10.01

08.01

04.01

03.01

02.01

22.12

21.12

20.12

18.12

17.12

16.12

15.12

14.12

13.12

12.12

11.12

10.12

07.12

06.12

05.12

04.12

02.12

online marketing applications

01.12

30.11

29.11

28.11

27.11

24.11

23.11

22.11

21.11

17.11

16.11

15.11

14.11

13.11

10.11

09.11

08.11

3.11

2.11

1.11

31.10

30.10

29.10

28.10

27.10

26.10

25.10

24.10

23.10

20.10

19.10

18.10

17.10

16.10

15.10

13.10

12.10

11.10

10.10

07.10

05.10

04.10

03.10

02.10

30.09

29.09

28.09

27.09

25.09

22.09

21.09

19.09

18.09

17.09

16.09

15.09

14.09

13.09

12.09

11.09

10.09

09.09

08.09

07.09

06.09

05.09

04.09

03.09

02.09

01.09

31.08

30.08

29.08

28.08

26.08

25.08

24.08

22.08

21.08

18.08

17.08

16.08

15.08

14.08

13.08

11.08

10.08

09.08

08.08

07.08

06.08

04.08

01.08

31.07

25.07

24.05

23.07

22.07

21.07

20.07

19.07

18.07

17.07

15.07

14.07

13.07

12.07

10.07

06.07

Maxout:

05.07

04.07

03.07

02.07

30.06

29.06

28.06

27.06

26.06

24.06

23.06

22.06

21.06

19.06

14.06

13.06

12.06

09.06

07.06

05.06

02.06

01.06

31.05

30.05

29.05

26.05

25.05

21.05

20.05

19.05

18.05

17.05

16.05

15.05

13.05

12.05

11.05

10.05

09.05

08.05

05.05

04.05

03.05

02.05

30.04

27.04

26.04

25.04

24.04

21.04

20.04

19.04

18.04

17.04

16.04

15.04

14.04

13.04

12.04

10.04

08.04

07.04

06.04

05.04

04.04

03.04

01.04

31.03

30.03

29.03

28.03

27.03

26.03

25.03

23.03

21.03

20.03

I haven't gone back to check what they are suggesting in their original paper, but I can guarantee that recent code written by Christian applies relu before BN. It is still occasionally a topic of debate, though.

17.03

16.03

15.03

14.03

13.03

10.03

09.03

08.03

07.03

06.03

05.03

04.03

02.03

01.03

28.02

27.02

26.02

25.02

24.02

23.02

22.02

21.02

20.02

19.02

18.02

17.02

16.02

15.02

14.02

13.02

12.02

10.02

08.02

07.02

06.02

27.1

26.1

25.1

24.1

23.1

20.1

19.1

18.1

17.1

16.1

15.1

14.1

13.1

12.1

11.1

10.1

9.1

7.1

5.1

4.1

3.1

2.1.17

31.12

30.12

29.12

28.12

27.12

26.12

24.12

23.12

22.12

21.12

20.12

19.12

17.12

16.12

15.12

14.12

13.12

12.12

11.12

9.12

8.12

7.12

6.12

5.12

2.12

1.12

30.11

29.11

28.11

27.11

26.11

25.11

24.11

23.11

Multithread in Theano:

Debug

22.11

21.11

19.11

18.11

17.11

16.11

15.11

14.11

13.11

12.11

11.11

10.11

9.11

8.11

7.11

6.11

04.11

3.11

2.11

About

daily curated links in DS, DL, NLP, ML

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published