yorkr rocks women’s One Day International (ODI) and International T20!!

“Life is not measured by the number of breaths we take, but by the moments that take our breath away.” Maya Angelou

“Life shrinks or expands in proportion to one’s courage.” Anais Nin

“Devotion to the truth is the hallmark of morality; there is no greater, nobler, more heroic form of devotion than the act of a man who assumes the responsibility of thinking.” Ayn Rand in Atlas Shrugged

Introduction

yorkr, this time, rocks women’s cricket!!! In this post, my R package yorkr analyzes women’s One Day International and International T20. The latest changes in my R package yorkr, as mentioned in my last post Revitalizing R package yorkr, included the modifications for the segregation men’s and women’s ODI and T20 matches into separate folders while converting them from YAML to R data frames. As the data was already converted I could just use the yorkr functions 90+ to analyze the women’s ODI and women’s T20. The data for this is taken from Cricsheet

My R package yorkr has 4 classes of functions

ODI Functions

Class 1: Analysis of ODI matches – See ODI-Part 1
Class 2: Analysis of all ODI matches between 2 ODI teams – See ODI Part 2
Class 3 : Analysis of all matches played by a ODI team againsta all other ODI teams – See ODI Part 3
Class 4 : Analysis of ODI batsmen and bowlers – See ODI Part 4

Note
-The converted data is available at yorkrData
-This RMarkdown file has been published at RPubs at yorkrAnalyzesWomensODIT20
-You can download this as a PDF at yorkrAnalyzesWomensODIT20

install.packages("../../../yorkrgit/yorkr_0.0.9.tar.gz",repos = NULL, type="source")

1. Analysis of women’s ODI matches

library(yorkr)

Save all matches between 2 teams

#saveAllMatchesBetweenTeams("../../../yorkrData2020/odi/odiWomenMatches/","../../../yorkrData2020/odi/odiWomenMatches2Teams/")

Save all matches played by an ODI team against all other ODI teams

#saveAllMatchesAllOpposition("../../../yorkrData2020/odi/odiWomenMatches/","../../../yorkrData2020/odi/odiWomenAllMatchesAllTeams/")

Since there are several functions in each class, I have randomly selected a few functions to demonstrate yorkr’s analysis ## ODI Match Analysis (Class 1) In the functions below ODI women matches are analyzed as in the India-Australia ODI in 7 Feb 2016.

1.Scorecard

load("../../../yorkrData2020/odi/odiWomenMatches/Australia-India-2016-02-07.RData")
aus_ind <- overs
teamBattingScorecardMatch(aus_ind,'India')

## Total= 223

## # A tibble: 7 x 5
##   batsman         ballsPlayed fours sixes  runs
##   <chr>                 <int> <int> <dbl> <dbl>
## 1 H Kaur                   42     2     0    22
## 2 J Goswami                 4     1     0     4
## 3 M Raj                   113    12     0    89
## 4 PG Raut                  31     2     0    24
## 5 S Mandhana               52     7     0    55
## 6 S Pandey                 18     2     0    17
## 7 V Krishnamurthy          21     2     0    12

2.Batting Partnerships

The partnerships in this match between India and Australia. Mithali Raj tops the list, with partnerships with Smriti Mandhana, Harmanpreet Kaur and Punam Raut. The next highest partnership is Smriti Mandhana

teamBatsmenPartnershipMatch(aus_ind,"India","Australia")

Analyze bowling in the women’s ODI England-New Zealand match on 15 Feb 2013

3.Wicket kind

load("../../../yorkrData2020/odi/odiWomenMatches/England-New Zealand-2013-02-15.RData")
eng_nz <- overs
teamBowlingWicketKindMatch(eng_nz,"England","New Zealand")

4.Match worm graph

Plot the match worm graph for Pakistan-South Africa women’s ODI 25 Jun 2017

load("../../../yorkrData2020/odi/odiWomenMatches/Pakistan-South Africa-2017-06-25.RData")
pak_sa <-overs
matchWormGraph(pak_sa,'Pakistan',"South Africa")

Analysis of team in all matches against another team (Class 2)

5. Team Batsmen partnerships

The functions below analyze all matches between South Africa and Sri Lanka.

load("../../../yorkrData2020/odi/odiWomenMatches2Teams/South Africa-Sri Lanka-allMatches.RData")
sa_sl_matches <- matches
m <-teamBatsmenPartnershiOppnAllMatches(sa_sl_matches,'South Africa',report="summary")
m

## # A tibble: 16 x 2
##    batsman        totalRuns
##    <chr>              <dbl>
##  1 M du Preez           241
##  2 M Kapp               194
##  3 L Wolvaardt          168
##  4 D van Niekerk        138
##  5 L Lee                138
##  6 T Chetty             136
##  7 A Steyn              118
##  8 L Goodall             89
##  9 S Luus                71
## 10 N de Klerk            35
## 11 CL Tryon              15
## 12 F Tunnicliffe         15
## 13 S Ismail               9
## 14 M Klaas                2
## 15 Y Fourie               1
## 16 B Bezuidenhout         0

teamBatsmenPartnershipOppnAllMatchesChart(sa_sl_matches,"Sri Lanka","South Africa")

6. Team bowler wicketkind

The plot below gives the performance if women Indian ODI bowlers in all ODI matches against England. The top wicket takers are Jhulan Goswami, Ekta Bisht, Gouher Sultana

load("../../../yorkrData2020/odi/odiWomenMatches2Teams/India-England-allMatches.RData")
ind_eng_matches <- matches
teamBowlersWicketsOppnAllMatches(ind_eng_matches,"India","England")

Performance of women ODI teams against all other teams in all matches (Class 3)

7. Overall batting scorecard

West Indies top scorers in ODI in all matches. The top scorers in West Indies are 1. Stafanie Taylor 2. Deandra Dottin 3. Hayley Matthews

load("../../../yorkrData2020/odi/odiWomenAllMatchesAllteams/allMatchesAllOpposition-West Indies.RData")
wi_matches <- matches
m <-teamBattingScorecardAllOppnAllMatches(wi_matches,theTeam="West Indies")

## Total= 4629

## # A tibble: 31 x 5
##    batsman          ballsPlayed fours sixes  runs
##    <chr>                  <int> <int> <int> <dbl>
##  1 SR Taylor               1087    83     7   766
##  2 DJS Dottin               778    69    21   641
##  3 HK Matthews              734    71     4   527
##  4 SA Campbelle             649    39     4   396
##  5 Kycia A Knight           517    35     2   284
##  6 CN Nation                554    31     1   274
##  7 Kyshona A Knight         578    35    NA   264
##  8 MR Aguilleira            481    20     3   252
##  9 B Cooper                 289    19     3   176
## 10 NY McLean                230    18     2   155
## # … with 21 more rows

Individual batsman and bowler performances (Class 4)

8. Batsmen performances

The functions below perform individual batsman and bowler analysis. I chose the top women ODI batsman

Mithali Raj (Ind) has the highest ODI runs with a career average of 50.64
Charlotte Edwards (Eng)
Suzie Bates (NX)

#india_details <- getTeamBattingDetails("India",dir="../../../yorkrData2020/odi/odiWomenMatches", save=TRUE,odir="../../../yorkrData2020/odi/odiWomenBattingBowlingDetails")
#eng_details <- getTeamBattingDetails("England",dir="../../../yorkrData2020/odi/odiWomenMatches", save=TRUE,odir="../../../yorkrData2020/odi/odiWomenBattingBowlingDetails")
#nz_details <- getTeamBattingDetails("New Zealand",dir="../../../yorkrData2020/odi/odiWomenMatches", save=TRUE,odir="../../../yorkrData2020/odi/odiWomenBattingBowlingDetails")

mithali <- getBatsmanDetails(team="India",name="M Raj",dir="../../../yorkrData2020/odi/odiWomenBattingBowlingDetails")

## [1] "../../../yorkrData2020/odi/odiWomenBattingBowlingDetails/India-BattingDetails.RData"

charlotte <- getBatsmanDetails(team="England",name="CM Edwards",dir="../../../yorkrData2020/odi/odiWomenBattingBowlingDetails")

## [1] "../../../yorkrData2020/odi/odiWomenBattingBowlingDetails/England-BattingDetails.RData"

suzie<- getBatsmanDetails(team="New Zealand",name="SW Bates",dir="../../../yorkrData2020/odi/odiWomenBattingBowlingDetails")

## [1] "../../../yorkrData2020/odi/odiWomenBattingBowlingDetails/New Zealand-BattingDetails.RData"

Plot Runs vs Strike Rate

library(grid)
library(gridExtra)
p1 <-batsmanRunsVsStrikeRate(mithali,"Mithali Raj")
p2 <- batsmanRunsVsStrikeRate(charlotte, "Charlotte E")
p3 <- batsmanRunsVsStrikeRate(suzie, "Suzie Bates")
grid.arrange(p1,p2,p3, ncol=2)

Plot the moving average

p1 <-batsmanMovingAverage(mithali,"Mithali Raj")
p2 <- batsmanMovingAverage(charlotte, "Charlotte E")
p3 <- batsmanMovingAverage(suzie, "Suzie Bates")
grid.arrange(p1,p2,p3, ncol=2)

p1 <-batsmanCumulativeAverageRuns(mithali,"Mithali Raj")
p2 <- batsmanCumulativeAverageRuns(charlotte, "Charlotte E")
p3 <- batsmanCumulativeAverageRuns(suzie, "Suzie Bates")
grid.arrange(p1,p2,p3, ncol=2)

Analyze ODI bowler performances

9. Bowler performances

The following 3 bowlers have been chosen for analysis

Jhulan Goswami (Ind) is the highest overwall wicket taker with 225 wicket
Anisa Mohammed (WI)
Sana Mir (Pak)

#india_details <- getTeamBowlingDetails("India",dir="../../../yorkrData2020/odi/odiWomenMatches", save=TRUE,odir="../../../yorkrData2020/odi/odiWomenBattingBowlingDetails")
#wi_details <- getTeamBowlingDetails("West Indies",dir="../../../yorkrData2020/odi/odiWomenMatches", save=TRUE,odir="../../../yorkrData2020/odi/odiWomenBattingBowlingDetails")
#pak_details <- getTeamBowlingDetails("Pakistan",dir="../../../yorkrData2020/odi/odiWomenMatches", save=TRUE,odir="../../../yorkrData2020/odi/odiWomenBattingBowlingDetails")

jhulan <- getBowlerWicketDetails(team="India",name="J Goswami",dir="../../../yorkrData2020/odi/odiWomenBattingBowlingDetails")
anisa <- getBowlerWicketDetails(team="West Indies",name="A Mohammed",dir="../../../yorkrData2020/odi/odiWomenBattingBowlingDetails")
sana <- getBowlerWicketDetails(team="Pakistan",name="Sana Mir",dir="../../../yorkrData2020/odi/odiWomenBattingBowlingDetails")

Plot the bowler Mean Economy Rate

p1<-bowlerMeanEconomyRate(jhulan,"Jhulan G")
p2<-bowlerMeanEconomyRate(anisa, "Anisa M")
p3<-bowlerMeanEconomyRate(sana, "Sana Mir")
grid.arrange(p1,p2,p3, ncol=2)

Plot the cumulative average wickets taken by the bowlers

p1<-bowlerCumulativeAvgWickets(jhulan,"Jhulan G")
p2<-bowlerCumulativeAvgWickets(anisa, "Anisa M")
p3<-bowlerCumulativeAvgWickets(sana, "Sana Mir")
grid.arrange(p1,p2,p3, ncol=2)

2. Analysis of women’s International Twenty 20 matches

I have chosen some random yorkr functions to show the analysis of T20 players and matches

T20 Functions

There are the following class of T20 functions

Class 1: Analysis of T20 matches – See T20-Part 1
Class 2: Analysis of all T20 matches between 2 T20 teams – See T20 Part 2
Class 3 : Analysis of all matches played by a T20 team againsta All other T20 teams – See T20 Part 3
Class 4 : Analysis of T20 batsmen and bowlers – See T20 Part 4

You can also refer to the yorkr template that I created Analysis of International T20 matches with yorkr templates

Save all matches between teams

#saveAllMatchesBetweenTeams("../../../yorkrData2020/t20/t20WomenMatches/","../../../yorkrData2020/t20/t20WomenMatches2Teams/")

Save all T20 matches played by a team against all other teams

#saveAllMatchesAllOpposition("../../../yorkrData2020/t20/t20WomenMatches/","../../../yorkrData2020/t20/t20WomenAllMatchesAllTeams/")

T20 Match Analysis (Class 1)

10. Batting scorecard

Print the scorecard for the Bangladesh- Ireland match played on 3 Apr 2014

load("../../../yorkrData2020/t20/t20WomenMatches/Bangladesh-Ireland-2014-04-03.RData")
ban_ire <- overs
teamBattingScorecardMatch(ban_ire,'Bangladesh')

## Total= 95

## # A tibble: 9 x 5
##   batsman         ballsPlayed fours sixes  runs
##   <chr>                 <int> <dbl> <dbl> <dbl>
## 1 Ayasha Rahman            19     2     0    12
## 2 Fahima Khatun             2     0     0     0
## 3 Lata Mondal              12     1     0     8
## 4 Panna Ghosh               3     0     0     4
## 5 Rumana Ahmed             14     3     0    16
## 6 Salma Khatun              6     1     0     7
## 7 Shaila Sharmin            7     0     0     6
## 8 Shamima Sultana          11     0     0     7
## 9 Sharmin Akhter           46     3     0    35

Plot the performance of T20 batsmen against in bowlers in Germany – Netherlands.

load("../../../yorkrData2020/t20/t20WomenMatches/Germany-Netherlands-2019-06-27.RData")
ger_net <- overs
teamBatsmenVsBowlersMatch(ger_net,'Netherlands',"Germany",plot=TRUE)

11. Bowling scorecard

Print the bowling scorecard of Hong Kong-Kuwait T20 match played on 25 Feb 2019

load("../../../yorkrData2020/t20/t20WomenMatches/Hong Kong-Kuwait-2019-02-25.RData")
hk_kuw <-overs
teamBowlingScorecardMatch(hk_kuw,'Hong Kong')

## # A tibble: 5 x 5
##   bowler      overs maidens  runs wickets
##   <chr>       <int>   <int> <dbl>   <int>
## 1 Chan Ka Man     2       0     5       1
## 2 KY Chan         3       1     2       4
## 3 M Hill          2       0     6       1
## 4 M Wai Siu       2       0    11       1
## 5 M Yousaf        1       1     0       3

Head to head between 2 women’s T20 teams (Class 2)

12. Team batting partnerships

Print the partnership among Indian T20 women in all matches against England

load("../../../yorkrData2020/t20/t20WomenMatches2Teams/India-England-allMatches.RData")
ind_eng_matches <- matches
m <-teamBatsmenPartnershiOppnAllMatches(ind_eng_matches,'India',report="detailed")
m[1:30,]

##       batsman      nonStriker partnershipRuns totalRuns
## 1       M Raj        A Sharma               2       233
## 2       M Raj      BS Fulmali              25       233
## 3       M Raj       DB Sharma              16       233
## 4       M Raj          H Kaur              18       233
## 5       M Raj       J Goswami               6       233
## 6       M Raj         KV Jain               5       233
## 7       M Raj        L Kumari               5       233
## 8       M Raj     N Niranjana               3       233
## 9       M Raj        N Tanwar              17       233
## 10      M Raj         PG Raut              41       233
## 11      M Raj      R Malhotra               5       233
## 12      M Raj      S Mandhana              17       233
## 13      M Raj          S Naik              10       233
## 14      M Raj        S Pandey              19       233
## 15      M Raj        SK Naidu              37       233
## 16      M Raj V Krishnamurthy               7       233
## 17 S Mandhana          H Deol              20       145
## 18 S Mandhana    JI Rodrigues              47       145
## 19 S Mandhana           M Raj              32       145
## 20 S Mandhana   Shafali Verma              46       145
## 21     H Kaur        A Sharma               1       137
## 22     H Kaur        AA Patil               8       137
## 23     H Kaur       DB Sharma              14       137
## 24     H Kaur         E Bisht               3       137
## 25     H Kaur       J Goswami              11       137
## 26     H Kaur    JI Rodrigues              12       137
## 27     H Kaur           M Raj              19       137
## 28     H Kaur      MR Meshram              33       137
## 29     H Kaur        N Tanwar               2       137
## 30     H Kaur         PG Raut               0       137

13. Team batting partnerships (plot)

Plot the batting partnership of Indian T20 womern against England

The best batsmen are Mithali Raj, Smriti Mandhana and Harmanpreet Kaur in that order

teamBatsmenPartnershipOppnAllMatchesChart(ind_eng_matches,"India","England")

14. Team Wicketkind

Plot the wicket kind taken by the bowlers of Scotland against USA

load("../../../yorkrData2020/t20/t20WomenMatches2Teams/Scotland-United States of America-allMatches.RData")
sco_usa_matches <- matches
teamBowlersWicketsOppnAllMatches(sco_usa_matches,"Scotalnd","USA")

Performance of teams against all other teams in all T20 matches (Class 3)

15. Overall team scorecard

Print the batting scorecard of Zimbabwe against all other teams

load("../../../yorkrData2020/t20/t20WomenAllMatchesAllTeams/allMatchesAllOpposition-Zimbabwe.RData")
zim_matches <- matches
m <-teamBattingScorecardAllOppnAllMatches(zim_matches,theTeam="Zimbabwe")

## Total= 571

## # A tibble: 7 x 5
##   batsman      ballsPlayed fours sixes  runs
##   <chr>              <int> <int> <int> <dbl>
## 1 SM Mayers            181    20     3   216
## 2 M Mupachikwa         139     9    NA   125
## 3 CS Mugeri             88     9     2   119
## 4 M Musonda             38     2     1    46
## 5 J Nkomo               25     3    NA    34
## 6 A Ndiraya             14     3    NA    18
## 7 AC Mushangwe          13    NA    NA    13

15. Team batting partnerships

Print the batting partnership of West Indies. The best performances are by 1. Stafanie Taylor 2. Deandra Dottin 3. Hayley Matthews

load("../../../yorkrData2020/t20/t20WomenAllMatchesAllTeams/allMatchesAllOpposition-West Indies.RData")
wi_matches <- matches
m <- teamBatsmenPartnershipAllOppnAllMatches(wi_matches,theTeam='West Indies')
m

## # A tibble: 29 x 2
##    batsman        totalRuns
##    <chr>              <dbl>
##  1 SR Taylor           1199
##  2 DJS Dottin           912
##  3 HK Matthews          458
##  4 SA Campbelle         407
##  5 B Cooper             300
##  6 SACA King            287
##  7 MR Aguilleira        250
##  8 CN Nation            243
##  9 Kycia A Knight       240
## 10 NY McLean            142
## # … with 19 more rows

16. Team bowling wicketkind

The plot below shows the women T20 bowlers who have performed the best against India namely 1. Katherine Brunt (Eng) 2. Elysse Perry (Aus) 3. Anya Shrubsole

load("../../../yorkrData2020/t20/t20WomenAllMatchesAllTeams/allMatchesAllOpposition-India.RData")
ind_matches <- matches
teamBowlingWicketKindAllOppnAllMatches(ind_matches,t1="India",t2="All")

Analyze women T20 batsmen & bowlers (Class 4)

17. T20 batsmen performances

The following 4 players were chosen

Harmanpreet Kaur (Ind)
Suzie Bates (NZ)
Meg Lanning (Aus)
Stafanie Tay;or (WI)

#india_details <- getTeamBattingDetails("India",dir="../../../yorkrData2020/t20/t20WomenMatches", save=TRUE,odir="../../../yorkrData2020/t20/t20WomenBattingBowlingDetails")
#eng_details <- getTeamBattingDetails("England",dir="../../../yorkrData2020/t20/t20WomenMatches", save=TRUE,odir="../../../yorkrData2020/t20/t20WomenBattingBowlingDetails")
#aus_details <- getTeamBattingDetails("Australia",dir="../../../yorkrData2020/t20/t20WomenMatches", save=TRUE,odir="../../../yorkrData2020/t20/t20WomenBattingBowlingDetails")
#wi_details <-  getTeamBattingDetails("West Indies",dir="../../../yorkrData2020/t20/t20WomenMatches", save=TRUE,odir="../../../yorkrData2020/t20/t20WomenBattingBowlingDetails")
#nz_details <-  getTeamBattingDetails("New Zealand",dir="../../../yorkrData2020/t20/t20WomenMatches", save=TRUE,odir="../../../yorkrData2020/t20/t20WomenBattingBowlingDetails")

harmanpreet <- getBatsmanDetails(team="India",name="H Kaur",dir="../../../yorkrData2020/t20/t20WomenBattingBowlingDetails")

## [1] "../../../yorkrData2020/t20/t20WomenBattingBowlingDetails/India-BattingDetails.RData"

suzie <- getBatsmanDetails(team="New Zealand",name="SW Bates",dir="../../../yorkrData2020/t20/t20WomenBattingBowlingDetails")

## [1] "../../../yorkrData2020/t20/t20WomenBattingBowlingDetails/New Zealand-BattingDetails.RData"

meg <- getBatsmanDetails(team="Australia",name="MM Lanning",dir="../../../yorkrData2020/t20/t20WomenBattingBowlingDetails")

## [1] "../../../yorkrData2020/t20/t20WomenBattingBowlingDetails/Australia-BattingDetails.RData"

stafanie <- getBatsmanDetails(team="West Indies",name="SR Taylor",dir="../../../yorkrData2020/t20/t20WomenBattingBowlingDetails")

## [1] "../../../yorkrData2020/t20/t20WomenBattingBowlingDetails/West Indies-BattingDetails.RData"

Plot the performance of the players against opposition.

batsmanRunsAgainstOpposition(harmanpreet,"Harmanpreet")

batsmanRunsAgainstOpposition(suzie,"Suzie Bates")

batsmanRunsAgainstOpposition(stafanie,"Stafanie Taylor")

batsmanRunsAgainstOpposition(meg,"Meg Lanning")

Plot the cumulative strike rate of the players. Meg Lanning has the best strike rate of the lot. Stafanie and Suzie also touch a strike rate of 100

p1<-batsmanCumulativeStrikeRate(harmanpreet,"Harmanpreet")
p2<-batsmanCumulativeStrikeRate(suzie,"Suzie Bates")
p3<-batsmanCumulativeStrikeRate(stafanie,"Stafanie Taylor")
p4 <-batsmanCumulativeStrikeRate(meg,"Meg Lanning")
grid.arrange(p1,p2,p3,p4, ncol=2)

Analyze women’s T20 bowlers.

18. T20 bowler performances

The following bowlers were chosen for analysis

Poonam Yadav (Ind)
Anisa Mohammed (WI)
Ellyse Perry (Aus)
Anya Shrubsole (England)

#india_details <- getTeamBowlingDetails("India",dir="../../../yorkrData2020/t20/t20WomenMatches", save=TRUE,odir="../../../yorkrData2020/t20/t20WomenBattingBowlingDetails")
#wi_details <- getTeamBowlingDetails("West Indies",dir="../../../yorkrData2020/t20/t20WomenMatches", save=TRUE,odir="../../../yorkrData2020/t20/t20WomenBattingBowlingDetails")
#aus_details <- getTeamBowlingDetails("Australia",dir="../../../yorkrData2020/t20/t20WomenMatches", save=TRUE,odir="../../../yorkrData2020/t20/t20WomenBattingBowlingDetails")
#eng_details <- getTeamBowlingDetails("England",dir="../../../yorkrData2020/t20/t20WomenMatches", save=TRUE,odir="../../../yorkrData2020/t20/t20WomenBattingBowlingDetails")

poonam <- getBowlerWicketDetails(team="India",name="Poonam Yadav",dir="../../../yorkrData2020/t20/t20WomenBattingBowlingDetails")
anisa <- getBowlerWicketDetails(team="West Indies",name="A Mohammed",dir="../../../yorkrData2020/t20/t20WomenBattingBowlingDetails")
ellyse <- getBowlerWicketDetails(team="Australia",name="EA Perry",dir="../../../yorkrData2020/t20/t20WomenBattingBowlingDetails")
anya <- getBowlerWicketDetails(team="England",name="A Shrubsole",dir="../../../yorkrData2020/t20/t20WomenBattingBowlingDetails")

Plot the bowler’s moving average

p1<-bowlerMovingAverage(poonam,"Poonam Yadav")
p2<-bowlerMovingAverage(anisa,"Anisa M")
p3 <-bowlerMovingAverage(ellyse,"Ellyse Perry")
p4 <-bowlerMovingAverage(anya,"Anya Shrubsole")
grid.arrange(p1,p2,p3,p4, ncol=2)

Plot the bowlers Cumulative Average Wickets

p1<-bowlerCumulativeAvgWickets(poonam,"Poonam Yadav")
p2<-bowlerCumulativeAvgWickets(anisa,"Anisa M")
p3 <-bowlerCumulativeAvgWickets(ellyse,"Ellyse Perry")
p4 <-bowlerCumulativeAvgWickets(anya,"Anya Shrubsole")
grid.arrange(p1,p2,p3,p4, ncol=2)

3a. Rank women ODI batsmen

Note: Mithali Raj (Ind) tops the ODI table with the most runs and highest average in ODI. The Cricsheet data does not have the earlier years in which she played. Hence you may see a much lower average for Mithali Raj

library(yorkr)
dir="/Users/tvganesh/backup/software/cricket-package/yorkr-cricsheet/yorkrData2020/odi/odiWomenMatches"
odir="/Users/tvganesh/backup/software/cricket-package/yorkr-cricsheet/yorkrData2020/odi/odiWomenBattingBowlingDetails"

rankODIBatsmen(dir=dir,odir=odir,minMatches=30)

## # A tibble: 24 x 4
##    batsman          matches meanRuns meanSR
##    <chr>              <int>    <dbl>  <dbl>
##  1 AE Satterthwaite      32     61.5   81.2
##  2 MM Lanning            47     49.5   85.0
##  3 TT Beaumont           35     45.8   68.5
##  4 EA Perry              42     45.7   74.3
##  5 SW Bates              42     44.0   70.9
##  6 NR Sciver             35     43.0   94.7
##  7 M Raj                 35     42.8   64.1
##  8 AC Jayangani          48     38.6   59.9
##  9 NE Bolton             32     36.5   60.5
## 10 T Chetty              34     33.1   70.3
## # … with 14 more rows

3b. Rank women ODI bowlers

Note: Jhulan Goswami tops the ODI bowlers with the most wickets. However the rank below is based on the available data in Cricsheet

library(yorkr)
dir="/Users/tvganesh/backup/software/cricket-package/yorkr-cricsheet/yorkrData2020/odi/odiWomenMatches"
odir="/Users/tvganesh/backup/software/cricket-package/yorkr-cricsheet/yorkrData2020/odi/odiWomenBattingBowlingDetails"

rankODIBowlers(dir=dir,odir=odir,minMatches=30)

## # A tibble: 19 x 4
##    bowler        matches totalWickets meanER
##    <chr>           <int>        <dbl>  <dbl>
##  1 JL Jonassen        44           76   3.90
##  2 M Kapp             49           70   3.80
##  3 S Ismail           44           65   3.82
##  4 KH Brunt           42           62   3.57
##  5 EA Perry           43           58   4.44
##  6 A Shrubsole        41           58   4.07
##  7 J Goswami          33           58   3.59
##  8 S Luus             41           54   4.82
##  9 D van Niekerk      40           53   3.84
## 10 ML Schutt          35           48   4.46
## 11 A Khaka            33           47   4.08
## 12 JL Gunn            30           43   4.23
## 13 I Ranaweera        35           42   4.89
## 14 Sana Mir           32           41   4.29
## 15 LA Marsh           30           40   4.16
## 16 NR Sciver          36           37   4.64
## 17 NR Sciver          36           37   4.64
## 18 NR Sciver          36           37   4.64
## 19 NR Sciver          36           37   4.64

4a. Rank women T20 batsman

library(yorkr)
dir="/Users/tvganesh/backup/software/cricket-package/yorkr-cricsheet/yorkrData2020/t20/t20WomenMatches"
odir="/Users/tvganesh/backup/software/cricket-package/yorkr-cricsheet/yorkrData2020/t20/t20WomenBattingBowlingDetails"

rankT20Batsmen(dir=dir,odir=odir,minMatches=30)

## # A tibble: 30 x 4
##    batsman       matches meanRuns meanSR
##    <chr>           <int>    <dbl>  <dbl>
##  1 SR Taylor          39     33.1   96.7
##  2 MM Lanning         53     29.3  102. 
##  3 EJ Villani         32     28.2   94.8
##  4 D van Niekerk      41     27.3   88.2
##  5 SJ Taylor          46     26.7  100. 
##  6 SW Bates           35     26.1   99.8
##  7 AC Jayangani       41     25.5   94.7
##  8 Bismah Maroof      52     24.5   83.0
##  9 DJS Dottin         38     24    109. 
## 10 CM Edwards         44     23.7   94.1
## # … with 20 more rows

4b. Rank women T20 bowlers

library(yorkr)
dir="/Users/tvganesh/backup/software/cricket-package/yorkr-cricsheet/yorkrData2020/t20/t20WomenMatches"
odir="/Users/tvganesh/backup/software/cricket-package/yorkr-cricsheet/yorkrData2020/t20/t20WomenBattingBowlingDetails"

rankT20Bowlers(dir=dir,odir=odir,minMatches=30)

## # A tibble: 20 x 4
##    bowler           matches totalWickets meanER
##    <chr>              <int>        <dbl>  <dbl>
##  1 A Shrubsole           50           76   5.95
##  2 Nida Dar              50           59   5.99
##  3 KH Brunt              49           57   5.93
##  4 JL Jonassen           50           55   5.31
##  5 EA Perry              51           52   5.67
##  6 S Ismail              50           52   5.40
##  7 ML Schutt             39           50   6.17
##  8 D van Niekerk         39           47   5.45
##  9 D Hazell              35           44   4.95
## 10 NR Sciver             44           43   6.30
## 11 JL Gunn               30           41   6.14
## 12 A Mohammed            43           41   5.80
## 13 M Kapp                31           39   5.08
## 14 Asmavia Iqbal         33           36   6.39
## 15 Sana Mir              46           36   5.85
## 16 HASD Siriwardene      35           33   6.31
## 17 EA Osborne            30           31   5.62
## 18 S Luus                37           29   7.13
## 19 KDU Prabodhani        33           25   4.87
## 20 Bismah Maroof         35           22   6.49

Conclusion

While I have just shown how to use a small subset of functions, you can use the entire set of yorkr functions to analyze individual matches, head-2-head confrontation of two teams, performance of a teams against all other teams and finally performance of individual batsmen and bowlers in women’s ODI and T20 games.

Revitalizing R package yorkr

There is nothing so useless as doing efficiently that which should not be done at all. Peter Drucker

The most important thing in communication is to hear what isn’t being said. Peter Drucker

“Work expands to fill the time available for its completion.” Corollary: “Expenditure rises to meet income.” Parkinson’s law

Introduction

“Operation successful!!!, the Programmer Surgeon in me, thought to himself. What should have been a routine surgery, turned out to be a major operation in the end, which involved several grueling hours. The surgeon looked at the large chunks of programming logic in the operation tray, which had been surgically removed, as they had outlived their utility and had partly become dysfunctional. The surgeon glanced at the new, concise code logic which had replaced the earlier somewhat convoluted logic, with a smile of satisfaction,

To, those who tuned in late, I am referring to my R package yorkr which I had created in many years ago, in early 2016. The package had worked well for quite some time on data from Cricsheet. Cricsheet went into a hiatus in late 2017-2018, and came alive back in 2019. Unfortunately, a key function in the package, started to malfunction. The diagnosis was that the format of the YAML files had changed, in newer files, which resulted in the problem. I had got mails from users mentioning that yorkr was not converting the new YAML files. This was on my to do list for a long time, and a week or two back, I decided to “bite the bullet” and fix the issue. I hoped the fix would be trivial but it was anything but. Finally, I took the hard decision of re-designing the core of the yorkr package, which involved converting YAML files to RData (dataframes). Also, since it has been a while since I did R code, having done more of Python stuff in recent times, I had to jog my memory with my earlier 2 posts Essential R and R vs Python

I spent many hours, tweaking and fixing the new logic so that it worked on the older and new files. Finally, I am happy to say that the new code is much more compact and probably less error prone.

I also had to ensure that the converted files performed exactly on all the other yorkr functions. I ran all the my yorkr functions in my yorkr posts on ODI, Intl. T20 and IPL and made sure the results were identical. (Phew!!)

The changes will be available in CRAN in yorkr_0.0.8

Do take a look at my yorkr posts. All the functions work correctly. Do use help, as I have changed a few functions. I will have my posts reflect the correct usage, but some function or other may slip the cracks.

One Day Internationals ODI-Part1, ODI-Part2, ODI-Part3, ODI-Part4
International T20s – T20-Part1,T20-Part2,T20-Part3,T20-Part4
Indian Premier League IPL-Part1, IPL-Part2,IPL-Part3, IPL-Part4

While making the changes, I also touched up some functions and made them more user friendly (added additional arguments etc). But by and large, yorkr is still yorkr and is intact.It just sports some spanking, new YAML conversion logic.

Note:

The code is available in Github yorkr
This RMarkdown has been published at RPubs Revitalizing yorkr
I have already converted the YAML files for ODI, Intl T20 and IPL. You can access and download the converted data from Github at yorkrData2020

setwd("/Users/tvganesh/backup/software/cricket-package/yorkr-cricsheet/yorkrgit")
install.packages("yorkr_0.0.8.tar.gz",repos = NULL, type="source")
library(yorkr)

Checkout my interactive Shiny apps GooglyPlus2021 (interactive plots ) and GooglyPlusPlus2021 (analysis in specific intervals) which can be used to analyze IPL players, teams and matches.

Below I rank batsmen and bowlers in ODIs, T20 and IPL based on the data from Cricsheet.

1a. Rank ODI Batsmen

dir="/Users/tvganesh/backup/software/cricket-package/yorkr-cricsheet/yorkrData2020/odi/odiMenMatches"
odir="/Users/tvganesh/backup/software/cricket-package/yorkr-cricsheet/yorkrData2020/odi/odiBattingBowlingDetails"

rankODIBatsmen(dir=dir,odir=odir,minMatches=50)

## # A tibble: 151 x 4
##    batsman        matches meanRuns meanSR
##    <chr>            <int>    <dbl>  <dbl>
##  1 Babar Azam          52     50.2   87.2
##  2 SD Hope             51     48.7   71.0
##  3 V Kohli            207     48.4   79.4
##  4 HM Amla            159     46.6   82.4
##  5 DA Warner          114     46.1   88.0
##  6 AB de Villiers     190     45.5   94.5
##  7 JE Root            108     44.9   82.5
##  8 SR Tendulkar        96     43.9   77.1
##  9 IJL Trott           63     43.1   68.9
## 10 Q de Kock          106     42.0   82.7
## # … with 141 more rows

1b. Rank ODI Bowlers

dir="/Users/tvganesh/backup/software/cricket-package/yorkr-cricsheet/yorkrData2020/odi/odiMenMatches"
odir="/Users/tvganesh/backup/software/cricket-package/yorkr-cricsheet/yorkrData2020/odi/odiBattingBowlingDetails"

rankODIBowlers(dir=dir,odir=odir,minMatches=30)

## # A tibble: 265 x 4
##    bowler           matches totalWickets meanER
##    <chr>              <int>        <dbl>  <dbl>
##  1 SL Malinga           191          308   5.25
##  2 MG Johnson           142          238   4.73
##  3 Shakib Al Hasan      157          214   4.72
##  4 Shahid Afridi        166          213   4.69
##  5 JM Anderson          143          207   4.96
##  6 KMDN Kulasekara      161          190   4.94
##  7 SCJ Broad            115          189   5.31
##  8 DW Steyn             114          188   4.96
##  9 Mashrafe Mortaza     139          180   4.97
## 10 Saeed Ajmal          106          180   4.17
## # … with 255 more rows

2a. Rank T20 Batsmen

dir="/Users/tvganesh/backup/software/cricket-package/yorkr-cricsheet/yorkrData2020/t20/t20MenMatches"
odir="/Users/tvganesh/backup/software/cricket-package/yorkr-cricsheet/yorkrData2020/t20/t20BattingBowlingDetails"

rankT20Batsmen(dir=dir,odir=odir,minMatches=50)

## # A tibble: 43 x 4
##    batsman          matches meanRuns meanSR
##    <chr>              <int>    <dbl>  <dbl>
##  1 V Kohli               61     39.0   132.
##  2 Mohammad Shahzad      52     31.8   123.
##  3 CH Gayle              50     31.1   124.
##  4 BB McCullum           69     30.7   126.
##  5 PR Stirling           66     29.6   116.
##  6 MJ Guptill            70     29.6   125.
##  7 DA Warner             75     29.1   128.
##  8 AD Hales              50     28.1   120.
##  9 TM Dilshan            78     26.7   105.
## 10 RG Sharma             72     26.4   120.
## # … with 33 more rows

2b. Rank T20 Bowlers

dir="/Users/tvganesh/backup/software/cricket-package/yorkr-cricsheet/yorkrData2020/t20/t20MenMatches"
odir="/Users/tvganesh/backup/software/cricket-package/yorkr-cricsheet/yorkrData2020/t20/t20BattingBowlingDetails"

rankT20Bowlers(dir=dir,odir=odir,,minMatches=30)

## # A tibble: 153 x 4
##    bowler          matches totalWickets meanER
##    <chr>             <int>        <dbl>  <dbl>
##  1 SL Malinga           78          115   7.39
##  2 Shahid Afridi        89           98   6.80
##  3 Saeed Ajmal          62           92   6.30
##  4 Umar Gul             56           87   7.40
##  5 KMDN Kulasekara      56           72   7.25
##  6 TG Southee           55           69   8.68
##  7 DJ Bravo             60           69   8.41
##  8 DW Steyn             47           69   7.00
##  9 Shakib Al Hasan      57           69   6.82
## 10 SCJ Broad            55           68   7.83
## # … with 143 more rows

3a. Rank IPL Batsmen

dir="/Users/tvganesh/backup/software/cricket-package/yorkr-cricsheet/yorkrData2020/ipl/iplMatches"
odir="/Users/tvganesh/backup/software/cricket-package/yorkr-cricsheet/yorkrData2020/ipl/iplBattingBowlingDetails"


rankIPLBatsmen(dir=dir,odir=odir,,minMatches=50)

## # A tibble: 69 x 4
##    batsman        matches meanRuns meanSR
##    <chr>            <int>    <dbl>  <dbl>
##  1 DA Warner          130     37.9   128.
##  2 CH Gayle           125     36.2   134.
##  3 SE Marsh            67     35.9   120.
##  4 MEK Hussey          59     33.8   105.
##  5 KL Rahul            59     33.5   128.
##  6 V Kohli            175     31.6   119.
##  7 AM Rahane          116     30.7   108.
##  8 AB de Villiers     141     30.3   135.
##  9 F du Plessis        65     29.4   117.
## 10 S Dhawan           140     29.0   114.
## # … with 59 more rows

3a. Rank IPL Bowlers

dir="/Users/tvganesh/backup/software/cricket-package/yorkr-cricsheet/yorkrData2020/ipl/iplMatches"
odir="/Users/tvganesh/backup/software/cricket-package/yorkr-cricsheet/yorkrData2020/ipl/iplBattingBowlingDetails"

rankIPLBowlers(dir=dir,odir=odir,,minMatches=30)

## # A tibble: 143 x 4
##    bowler          matches totalWickets meanER
##    <chr>             <int>        <dbl>  <dbl>
##  1 SL Malinga          120          184   6.99
##  2 SP Narine           108          137   6.71
##  3 Harbhajan Singh     131          134   7.11
##  4 DJ Bravo             85          118   8.18
##  5 B Kumar              86          116   7.43
##  6 YS Chahal            82          102   7.85
##  7 R Ashwin             92           98   6.81
##  8 JJ Bumrah            76           91   7.47
##  9 PP Chawla            85           87   8.02
## 10 RA Jadeja            89           85   7.93
## # … with 133 more rows

##Conclusion

Go ahead and give yorkr a spin once yorkr_0.0.8 is available in CRAN. I hope you have fun. Do get back to me if you have any issues.

I’ll be back. Watch this space!!

Deconstructing Convolutional Neural Networks with Tensorflow and Keras

I have been very fascinated by how Convolution Neural Networks have been able to, so efficiently, do image classification and image recognition CNN’s have been very successful in in both these tasks. A good paper that explores the workings of a CNN Visualizing and Understanding Convolutional Networks by Matthew D Zeiler and Rob Fergus. They show how through a reverse process of convolution using a deconvnet.

In their paper they show how by passing the feature map through a deconvnet ,which does the reverse process of the convnet, they can reconstruct what input pattern originally caused a given activation in the feature map

In the paper they say “A deconvnet can be thought of as a convnet model that uses the same components (filtering, pooling) but in reverse, so instead of mapping pixels to features, it does the opposite. An input image is presented to the CNN and features activation computed throughout the layers. To examine a given convnet activation, we set all other activations in the layer to zero and pass the feature maps as input to the attached deconvnet layer. Then we successively (i) unpool, (ii) rectify and (iii) filter to reconstruct the activity in the layer beneath that gave rise to the chosen activation. This is then repeated until input pixel space is reached.”

I started to scout the internet to see how I can implement this reverse process of Convolution to understand what really happens under the hood of a CNN. There are a lot of good articles and blogs, but I found this post Applied Deep Learning – Part 4: Convolutional Neural Networks take the visualization of the CNN one step further.

This post takes VGG16 as the pre-trained network and then uses this network to display the intermediate visualizations. While this post was very informative and also the visualizations of the various images were very clear, I wanted to simplify the problem for my own understanding.

Hence I decided to take the MNIST digit classification as my base problem. I created a simple 3 layer CNN which gives close to 99.1% accuracy and decided to see if I could do the visualization.

As mentioned in the above post, there are 3 major visualisations

Feature activations at the layer
Visualisation of the filters
Visualisation of the class outputs

Feature Activation – This visualization the feature activation at the 3 different layers for a given input image. It can be seen that first layer activates based on the edge of the image. Deeper layers activate in a more abstract way.

Visualization of the filters: This visualization shows what patterns the filters respond maximally to. This is implemented in Keras here

To do this the following is repeated in a loop

Choose a loss function that maximizes the value of a convnet filter activation
Do gradient ascent (maximization) in input space that increases the filter activation

Visualizing Class Outputs of the MNIST Convnet: This process is similar to determining the filter activation. Here the convnet is made to generate an image that represents the category maximally.

You can access the Google colab notebook here – Deconstructing Convolutional Neural Networks in Tensoflow and Keras

import numpy as np
import pandas as pd
import os
import tensorflow as tf
import matplotlib.pyplot as plt
from keras.layers import Dense, Dropout, Flatten
from keras.layers import Conv2D, MaxPooling2D, Input
from keras.models import Model
from sklearn.model_selection import train_test_split
from keras.utils import np_utils

Using TensorFlow backend.

In [0]:

mnist=tf.keras.datasets.mnist
# Set training and test data and labels
(training_images,training_labels),(test_images,test_labels)=mnist.load_data()

In [0]:

#Normalize training data
X =np.array(training_images).reshape(training_images.shape[0],28,28,1) 
# Normalize the images by dividing by 255.0
X = X/255.0
X.shape
# Use one hot encoding for the labels
Y = np_utils.to_categorical(training_labels, 10)
Y.shape
# Split training data into training and validation data in the ratio of 80:20
X_train, X_validation, y_train, y_validation = train_test_split(X,Y,test_size=0.20, random_state=42)

In [4]:

# Normalize test data
X_test =np.array(test_images).reshape(test_images.shape[0],28,28,1) 
X_test=X_test/255.0
#Use OHE for the test labels
Y_test = np_utils.to_categorical(test_labels, 10)
X_test.shape

Out[4]:

(10000, 28, 28, 1)

Display data

Display the training data and the corresponding labels

In [5]:

print(training_labels[0:10])
f, axes = plt.subplots(1, 10, sharey=True,figsize=(10,10))
for i,ax in enumerate(axes.flat):
    ax.axis('off')
    ax.imshow(X[i,:,:,0],cmap="gray")

Create a Convolutional Neural Network

The CNN consists of 3 layers

Conv2D of size 28 x 28 with 24 filters
Perform Max pooling
Conv2D of size 14 x 14 with 48 filters
Perform max pooling
Conv2d of size 7 x 7 with 64 filters
Flatten
Use Dense layer with 128 units
Perform 25% dropout
Perform categorical cross entropy with softmax activation function

In [0]:

num_classes=10
inputs = Input(shape=(28,28,1))
x = Conv2D(24,kernel_size=(3,3),padding='same',activation="relu")(inputs)
x = MaxPooling2D(pool_size=(2, 2))(x)
x = Conv2D(48, (3, 3), padding='same',activation='relu')(x)
x = MaxPooling2D(pool_size=(2, 2))(x)
x = Conv2D(64, (3, 3), padding='same',activation='relu')(x)
x = MaxPooling2D(pool_size=(2, 2))(x)
x = Flatten()(x)
x = Dense(128, activation='relu')(x)
x = Dropout(0.25)(x)
output = Dense(num_classes,activation="softmax")(x)

model = Model(inputs,output)

model.compile(loss='categorical_crossentropy', 
          optimizer='adam', 
          metrics=['accuracy'])

Summary of CNN

Display the summary of CNN

In [7]:

model.summary()

Model: "model_1"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
input_1 (InputLayer)         (None, 28, 28, 1)         0         
_________________________________________________________________
conv2d_1 (Conv2D)            (None, 28, 28, 24)        240       
_________________________________________________________________
max_pooling2d_1 (MaxPooling2 (None, 14, 14, 24)        0         
_________________________________________________________________
conv2d_2 (Conv2D)            (None, 14, 14, 48)        10416     
_________________________________________________________________
max_pooling2d_2 (MaxPooling2 (None, 7, 7, 48)          0         
_________________________________________________________________
conv2d_3 (Conv2D)            (None, 7, 7, 64)          27712     
_________________________________________________________________
max_pooling2d_3 (MaxPooling2 (None, 3, 3, 64)          0         
_________________________________________________________________
flatten_1 (Flatten)          (None, 576)               0         
_________________________________________________________________
dense_1 (Dense)              (None, 128)               73856     
_________________________________________________________________
dropout_1 (Dropout)          (None, 128)               0         
_________________________________________________________________
dense_2 (Dense)              (None, 10)                1290      
=================================================================
Total params: 113,514
Trainable params: 113,514
Non-trainable params: 0
_________________________________________________________________

Perform Gradient descent and validate with the validation data

In [8]:

epochs = 20
batch_size=256
history = model.fit(X_train,y_train,
         epochs=epochs,
         batch_size=batch_size,
         validation_data=(X_validation,y_validation))

————————————————

acc = history.history[ ‘accuracy’ ]

val_acc = history.history[ ‘val_accuracy’ ]

loss = history.history[ ‘loss’ ]

val_loss = history.history[‘val_loss’ ]

epochs = range(len(acc)) # Get number of epochs

#————————————————

# Plot training and validation accuracy per epoch

#————————————————

plt.plot ( epochs, acc,label=”training accuracy” )

plt.plot ( epochs, val_acc, label=’validation acuracy’ )

plt.title (‘Training and validation accuracy’)

plt.legend()

plt.figure()

#————————————————

# Plot training and validation loss per epoch

#————————————————

plt.plot ( epochs, loss , label=”training loss”)

plt.plot ( epochs, val_loss,label=”validation loss” )

plt.title (‘Training and validation loss’ )

plt.legend()

Test model on test data

f, axes = plt.subplots(1, 10, sharey=True,figsize=(10,10))

for i,ax in enumerate(axes.flat):

ax.axis(‘off’)

ax.imshow(X_test[i,:,:,0],cmap=”gray”)

l=[]
for i in range(10):
  x=X_test[i].reshape(1,28,28,1)
  y=model.predict(x)
  m = np.argmax(y, axis=1)
  print(m)

[7]
[2]
[1]
[0]
[4]
[1]
[4]
[9]
[5]
[9]

Generate the filter activations at the intermediate CNN layers

In [12]:

img = test_images[51].reshape(1,28,28,1)
fig = plt.figure(figsize=(5,5))
print(img.shape)
plt.imshow(img[0,:,:,0],cmap="gray")
plt.axis('off')

Display the activations at the intermediate layers

This displays the intermediate activations as the image passes through the filters and generates these feature maps

In [13]:

layer_names = ['conv2d_4', 'conv2d_5', 'conv2d_6']

layer_outputs = [layer.output for layer in model.layers if 'conv2d' in layer.name]
activation_model = Model(inputs=model.input,outputs=layer_outputs)
intermediate_activations = activation_model.predict(img)
images_per_row = 8
max_images = 8

for layer_name, layer_activation in zip(layer_names, intermediate_activations):
    print(layer_name,layer_activation.shape)
    n_features = layer_activation.shape[-1]
    print("features=",n_features)
    n_features = min(n_features, max_images)
    print(n_features)

    size = layer_activation.shape[1]
    print("size=",size)
    n_cols = n_features // images_per_row
    display_grid = np.zeros((size * n_cols, images_per_row * size))


    for col in range(n_cols):
      for row in range(images_per_row):
          channel_image = layer_activation[0,:, :, col * images_per_row + row]

          channel_image -= channel_image.mean()
          channel_image /= channel_image.std()
          channel_image *= 64
          channel_image += 128
          channel_image = np.clip(channel_image, 0, 255).astype('uint8')
          display_grid[col * size : (col + 1) * size,
                         row * size : (row + 1) * size] = channel_image
    scale = 2. / size
    plt.figure(figsize=(scale * display_grid.shape[1],
                        scale * display_grid.shape[0]))
    plt.axis('off')
    plt.title(layer_name)
    plt.grid(False)
    plt.imshow(display_grid, aspect='auto', cmap='viridis')
    
plt.show()

It can be seen that at the higher layers only abstract features of the input image are captured

# To fix the ImportError: cannot import name 'imresize' in the next cell. Run this cell. Then comment and restart and run all
#!pip install scipy==1.1.0

Visualize the pattern that the filters respond to maximally

Choose a loss function that maximizes the value of the CNN filter in a given layer
Start from a blank input image.
Do gradient ascent in input space. Modify input values so that the filter activates more
Repeat this in a loop.

In [14]:

from vis.visualization import visualize_activation, get_num_filters
from vis.utils import utils
from vis.input_modifiers import Jitter

max_filters = 24
selected_indices = []
vis_images = [[], [], [], [], []]
i = 0
selected_filters = [[0, 3, 11, 15, 16, 17, 18, 22], 
    [8, 21, 23, 25, 31, 32, 35, 41], 
    [2, 7, 11, 14, 19, 26, 35, 48]]

# Set the layers
layer_name = ['conv2d_4', 'conv2d_5', 'conv2d_6']
# Set the layer indices
layer_idx = [1,3,5]
for layer_name,layer_idx in zip(layer_name,layer_idx):


    # Visualize all filters in this layer.
    if selected_filters:
        filters = selected_filters[i]
    else:
        # Randomly select filters
        filters = sorted(np.random.permutation(get_num_filters(model.layers[layer_idx]))[:max_filters])
    selected_indices.append(filters)

    # Generate input image for each filter.
    # Loop through the selected filters in each layer and generate the activation of these filters
    for idx in filters:
        img = visualize_activation(model, layer_idx, filter_indices=idx, tv_weight=0., 
                                   input_modifiers=[Jitter(0.05)], max_iter=300) 
        vis_images[i].append(img)

    # Generate stitched image palette with 4 cols so we get 2 rows.
    stitched = utils.stitch_images(vis_images[i], cols=4)    
    plt.figure(figsize=(20, 30))
    plt.title(layer_name)
    plt.axis('off')
    stitched = stitched.reshape(1,61,127,1)
    plt.imshow(stitched[0,:,:,0])
    plt.show()
    i += 1

from vis.utils import utils
new_vis_images = [[], [], [], [], []]
i = 0
layer_name = ['conv2d_4', 'conv2d_5', 'conv2d_6']
layer_idx = [1,3,5]
for layer_name,layer_idx in zip(layer_name,layer_idx):
   
    # Generate input image for each filter.
    for j, idx in enumerate(selected_indices[i]):
        img = visualize_activation(model, layer_idx, filter_indices=idx, 
                                   seed_input=vis_images[i][j], input_modifiers=[Jitter(0.05)], max_iter=300) 
        #img = utils.draw_text(img, 'Filter {}'.format(idx))  
        new_vis_images[i].append(img)

    stitched = utils.stitch_images(new_vis_images[i], cols=4)   
    plt.figure(figsize=(20, 30))
    plt.title(layer_name)
    plt.axis('off')
    print(stitched.shape)
    stitched = stitched.reshape(1,61,127,1)
    plt.imshow(stitched[0,:,:,0])
    plt.show()
    i += 1

Visualizing Class Outputs

Here the CNN will generate the image that maximally represents the category. Each of the output represents one of the digits as can be seen below

In [16]:

from vis.utils import utils
from keras import activations
codes = '''
zero 0
one 1
two 2
three 3
four 4
five 5
six 6
seven 7
eight 8
nine 9
'''
layer_idx=10
initial = []
images = []
tuples = []
# Swap softmax with linear for better visualization
model.layers[layer_idx].activation = activations.linear
model = utils.apply_modifications(model)
for line in codes.split('\n'):
    if not line:
        continue
    name, idx = line.rsplit(' ', 1)
    idx = int(idx)
    img = visualize_activation(model, layer_idx, filter_indices=idx, 
                               tv_weight=0., max_iter=300, input_modifiers=[Jitter(0.05)])

    initial.append(img)
    tuples.append((name, idx))

i = 0
for name, idx in tuples:
    img = visualize_activation(model, layer_idx, filter_indices=idx,
                               seed_input = initial[i], max_iter=300, input_modifiers=[Jitter(0.05)])
    #img = utils.draw_text(img, name) # Unable to display text on gray scale image
    i += 1
    images.append(img)

stitched = utils.stitch_images(images, cols=4)
plt.figure(figsize=(20, 20))
plt.axis('off')
stitched = stitched.reshape(1,94,127,1)
plt.imshow(stitched[0,:,:,0])

plt.show()

In the grid below the class outputs show the MNIST digits to which output responds to maximally. We can see the ghostly outline
of digits 0 – 9. We can clearly see the outline if 0,1, 2,3,4,5 (yes, it is there!),6,7, 8 and 9. If you look at this from a little distance the digits are clearly visible. Isn’t that really cool!!

Conclusion:

It is really interesting to see the class outputs which show the image to which the class output responds to maximally. In the
post Applied Deep Learning – Part 4: Convolutional Neural Networks the class output show much more complicated images and is worth a look. It is really interesting to note that the model has adjusted the filter values and the weights of the fully connected network to maximally respond to the MNIST digits

References

1. Visualizing and Understanding Convolutional Networks
2. Applied Deep Learning – Part 4: Convolutional Neural Networks
3. Visualizing Intermediate Activations of a CNN trained on the MNIST Dataset
4. How convolutional neural networks see the world
5. Keras – Activation_maximization

Also see

1. Using Reinforcement Learning to solve Gridworld
2. Deep Learning from first principles in Python, R and Octave – Part 8
3. Cricketr learns new tricks : Performs fine-grained analysis of players
4. Video presentation on Machine Learning, Data Science, NLP and Big Data – Part 1
5. Big Data-2: Move into the big league:Graduate from R to SparkR
6. OpenCV: Fun with filters and convolution
7. Powershell GUI – Adding bells and whistles

To see all posts click Index of posts

Understanding Neural Style Transfer with Tensorflow and Keras

Neural Style Transfer (NST) is a fascinating area of Deep Learning and Convolutional Neural Networks. NST is an interesting technique, in which the style from an image, known as the ‘style image’ is transferred to another image ‘content image’ and we get a third a image which is a generated image which has the content of the original image and the style of another image.

NST can be used to reimagine how famous painters like Van Gogh, Claude Monet or a Picasso would have visualised a scenery or architecture. NST uses Convolutional Neural Networks (CNNs) to achieve this artistic style transfer from one image to another. NST was originally implemented by Gati et al., in their paper Neural Algorithm of Artistic Style. Convolutional Neural Networks have been very successful in image classification image recognition et cetera. CNN networks have also been have also generated very interesting pictures using Neural Style Transfer which will be shown in this post. An interesting aspect of CNN’s is that the first couple of layers in the CNN capture basic features of the image like edges and pixel values. But as we go deeper into the CNN, the network captures higher level features of the input image.

To get started with Neural Style transfer we will be using the VGG19 pre-trained network. The VGG19 CNN is a compact pre-trained your network which can be used for performing the NST. However, we could have also used Resnet or InceptionV3 networks for this purpose but these are very large networks. The idea of using a network trained on a different task and applying it to a new task is called transfer learning.

What needs to be done to transfer the style from one of the image to another image. This brings us to the question – What is ‘style’? What is it that distinguishes Van Gogh’s painting or Picasso’s cubist art. Convolutional Neural Networks capture basic features in the lower layers and much more complex features in the deeper layers. Style can be computed by taking the correlation of the feature maps in a layer L. This is my interpretation of how style is captured. Since style is intrinsic to the image, it implies that the style feature would exist across all the filters in a layer. Hence, to pick up this style we would need to get the correlation of the filters across channels of a lawyer. This is computed mathematically, using the Gram matrix which calculates the correlation of the activation of a the filter by the style image and generated image

To transfer the style from one image to the content image we need to do two parallel operations while doing forward propagation
– Compute the content loss between the source image and the generated image
– Compute the style loss between the style image and the generated image
– Finally we need to compute the total loss

In order to get transfer the style from the ‘style’ image to the ‘content ‘image resulting in a ‘generated’ image the total loss has to be minimised. Therefore backward propagation with gradient descent is done to minimise the total loss comprising of the content and style loss.

Initially we make the Generated Image ‘G’ the same as the source image ‘S’

The content loss at layer ‘l’

$L_{content} = 1/2 \sum_{i}^{j} ( F^{l}_{i,j} - P^{l}_{i,j})^{2}$

where $F^{l}_{i,j}$ and $P^{l}_{i,j}$ represent the activations at layer ‘l’ in a filter i, at position ‘j’. The intuition is that the activations will be same for similar source and generated image. We need to minimise the content loss so that the generated stylized image is as close to the original image as possible. An intermediate layer of VGG19 block5_conv2 is used

The Style layers that are are used are

style_layers = [‘block1_conv1’,

‘block2_conv1’,

‘block3_conv1’,

‘block4_conv1’,

‘block5_conv1’]

To compute the Style Loss the Gram matrix needs to be computed. The Gram Matrix is computed by unrolling the filters as shown below (source: Convolutional Neural Networks by Prof Andrew Ng, Coursera). The result is a matrix of size $n_{c}$ x $n_{c}$ where $n_{c}$ is the number of channels

The above diagram shows the filters of height $n_{H}$ and width $n_{W}$ with $n_{C}$ channels

The contribution of layer ‘l’ to style loss is given by

$L^{'}_{style} = \frac{\sum_{i}^{j} (G^{2}_{i,j} - A^l{i,j})^2}{4N^{2}_{l}M^{2}_{l}}$

where $G_{i,j}$ and $A_{i,j}$ are the Gram matrices of the style and generated images respectively. By minimising the distance in the gram matrices of the style and generated image we can ensure that generated image is a stylized version of the original image similar to the style image

The total loss is given by

$L_{total} = \alpha L_{content} + \beta L_{style}$

Back propagation with gradient descent works to minimise the content loss between the source and generated image, while the style loss tries to minimise the discrepancies in the style of the style image and generated image. Running through forward and backpropagation through several epochs successfully transfers the style from the style image to the source image.

You can check the Notebook at Neural Style Transfer

Note: The code in this notebook is largely based on the Neural Style Transfer tutorial from Tensorflow, though I may have taken some changes from other blogs. I also made a few changes to the code in this tutorial, like removing the scaling factor, or the class definition (Personally, I belong to the old school (C language) and am not much in love with the ‘self.”..All references are included below

Note: Here is a interesting thought. Could we do a Neural Style Transfer in music? Imagine Carlos Santana playing ‘Hotel California’ or Brian May style in ‘Another brick in the wall’. While our first reaction would be that it may not sound good as we are used to style of these songs, we may be surprised by a possible style transfer. This is definitely music to the ears!

Here are few runs from this

A) Run 1

1. Neural Style Transfer – a) Content Image – My portrait. b) Style Image – Wassily Kadinsky Oil on canvas, 1913, Vassily Kadinsky’s composition

2. Result of Neural Style Transfer

2) Run 2

a) Content Image – Portrait of my parents b) Style Image – Vincent Van Gogh’s ,Starry Night Oil on canvas 1889

2. Result of Neural Style Transfer

Run 3

1. Content Image – Caesar 2 (Masai Mara- 20 Jun 2018). Style Image – The Great Wave at Kanagawa – Katsushika Hokosai, 1826-1833

Screenshot 2020-04-12 at 12.40.44 PM

2. Result of Neural Style Transfer

lkg

Run 4

1. Content Image – Junagarh Fort , Rajasthan Sep 2016 b) Style Image – Le Pont Japonais by Claude Monet, Oil on canvas, 1920

2. Result of Neural Style Transfer

Neural Style Transfer is a very ingenious idea which shows that we can segregate the style of a painting and transfer to another image.

References

1. A Neural Algorithm of Artistic Style, Leon A. Gatys, Alexander S. Ecker, Matthias Bethge
2. Neural style transfer
3. Neural Style Transfer: Creating Art with Deep Learning using tf.keras and eager execution
4. Convolutional Neural Network, DeepLearning.AI Specialization, Prof Andrew Ng
5. Intuitive Guide to Neural Style Transfer

The mechanics of Convolutional Neural Networks in Tensorflow and Keras

Convolutional Neural Networks (CNNs), have been very popular in the last decade or so. CNNs have been used in multiple applications like image recognition, image classification, facial recognition, neural style transfer etc. CNN’s have been extremely successful in handling these kind of problems. How do they work? What makes them so successful? What is the principle behind CNN’s ?

Note: this post is based on two Coursera courses I did, namely namely Deep Learning specialisation by Prof Andrew Ng and Tensorflow Specialisation by Laurence Moroney.

In this post I show you how CNN’s work. To understand how CNNs work, we need to understand the concept behind machine learning algorithms. If you take a simple machine learning algorithm in which you are trying to do multi-class classification using softmax or binary classification with the sigmoid function, for a set of for a set of input features against a target variable we need to create an objective function of the input features versus the target variable. Then we need to minimise this objective function, while performing gradient descent, such that the cost is the lowest. This will give the set of weights for the different variables in the objective function.

The central problem in ML algorithms is to do feature selection, i.e. we need to find the set of features that actually influence the target. There are various methods for doing features selection – best fit, forward fit, backward fit, ridge and lasso regression. All these methods try to pick out the predictors that influence the output most, by making the weights of the other features close to zero. Please look at my post – Practical Machine Learning in R and Python – Part 3, where I show you the different methods for doing features selection.

In image classification or Image recognition we need to find the important features in the image. How do we do that? Many years back, have played around with OpenCV. While working with OpenCV I came across are numerous filters like the Sobel ,the Laplacian, Canny, Gaussian filter et cetera which can be used to identify key features of the image. For example the Canny filter feature can be used for edge detection, Gaussian for smoothing, Sobel for determining the derivative and we have other filters for detecting vertical or horizontal edges. Take a look at my post Computer Vision: Ramblings on derivatives, histograms and contours So for handling images we need to apply these filters to pick out the key features of the image namely the edges and other features. So rather than using the entire image’s pixels against the target class we can pick out the features from the image and use that as predictors of the target output.

Note: that in Convolutional Neural Network, fixed filter values like the those shown above are not used directly. Rather the filter values are learned through back propagation and gradient descent as shown below.

In CNNs the filter values are considered to be weights which are then learned and updated in each forward/backward propagation cycle much like the way a fully connected Deep Learning Network learns the weights of the network.

Here is a short derivation of the most important parts of how a CNNs work

The convolution of a filter F with the input X can be represented as.

Convolving we get

This the forward propagation as it passes through a non-linear function like Relu

To go through back propagation we need to compute the $\partial L$ at every node of Convolutional Neural network

The loss with respect to the output is $\partial L/\partial O$ . $\partial O/\partial X$ & $\partial O/\partial F$ are the local derivatives

We need these local derivatives because we can learn the filter values using gradient descent

where $\alpha$ is the learning rate. Also $\partial L/\partial X$ is the loss which is back propagated to the previous layers. You can see the detailed derivation of back propagation in my post Deep Learning from first principles in Python, R and Octave – Part 3 in a L-layer, multi-unit Deep Learning network.

In the fully connected layers the weights associated with each connection is computed in every cycle of forward and backward propagation using gradient descent. Similarly, the filter values are also computed and updated in each forward and backward propagation cycle. This is done so as to minimize the loss at the output layer.

By using the chain rule and simplifying the back propagation for the Convolutional layers we get these 2 equations. The first equation is used to learn the filter values and the second is used pass the loss to layers before

(for the detailed derivation see Convolutions and Backpropagations

An important aspect of performing convolutions is to reduce the size of the flattened image that is passed into the fully connected DL network. Successively convolving with 2D filters and doing a max pooling helps to reduce the size of the features that we can use for learning the images. Convolutions also enable a sparsity of connections as you can see in the diagram below. In the LeNet-5 Convolution Neural Network of Yann Le Cunn, successive convolutions reduce the image size from 28 x 28=784 to 120 flattened values.

Here is an interesting Deep Learning problem. Convolutions help in picking out important features of images and help in image classification/ detection. What would be its equivalent if we wanted to identify the Carnatic ragam of a song? A Carnatic ragam is roughly similar to Western scales (major, natural, melodic, blues) with all its modes Lydian, Aeolion, Phyrgian etc. Except in the case of the ragams, it is more nuanced, complex and involved. Personally, I can rarely identify a ragam on which a carnatic song is based (I am tone deaf when it comes to identifying ragams). I have come to understand that each Carnatic ragam has its own character, which is made up of several melodic phrases which are unique to that flavor of a ragam. What operation like convolution would be needed so that we can pick out these unique phrases in a Carnatic ragam? Of course, we would need to use it in Recurrent Neural Networks with LSTMs as a song is a time sequence of notes to identify sequences. Nevertheless, if there was some operation with which we can pick up the distinct, unique phrases from a song and then run it through a classifier, maybe we would be able to identify the ragam of the song.

Below I implement 3 simple CNN using the Dogs vs Cats Dataset from Kaggle. The first CNN uses regular Convolutions a Fully connected network to classify the images. The second approach uses Image Augmentation. For some reason, I did not get a better performance with Image Augumentation. Thirdly I use the pre-trained Inception v3 network.

1. Basic Convolutional Neural Network in Tensorflow & Keras

You can view the Colab notebook here – Cats_vs_dogs_1.ipynb

Here some important parts of the notebook

Create CNN Model

Use 3 Convolution + Max pooling layers with 32,64 and 128 filters respectively
Flatten the data
Have 2 Fully connected layers with 128, 512 neurons with relu activation
Use sigmoid for binary classification

In [0]:

model = tf.keras.models.Sequential([
    tf.keras.layers.Conv2D(32,(3,3),activation='relu',input_shape=(150,150,3)),
    tf.keras.layers.MaxPooling2D(2,2),
    tf.keras.layers.Conv2D(64,(3,3),activation='relu'),
    tf.keras.layers.MaxPooling2D(2,2),
    tf.keras.layers.Conv2D(128,(3,3),activation='relu'),
    tf.keras.layers.MaxPooling2D(2,2),
    tf.keras.layers.Flatten(),
    tf.keras.layers.Dense(128,activation='relu'),
    tf.keras.layers.Dense(512,activation='relu'),
    tf.keras.layers.Dense(1,activation='sigmoid')
])

Print model summary

In [13]:

model.summary()

Model: "sequential"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
conv2d (Conv2D)              (None, 148, 148, 32)      896       
_________________________________________________________________
max_pooling2d (MaxPooling2D) (None, 74, 74, 32)        0         
_________________________________________________________________
conv2d_1 (Conv2D)            (None, 72, 72, 64)        18496     
_________________________________________________________________
max_pooling2d_1 (MaxPooling2 (None, 36, 36, 64)        0         
_________________________________________________________________
conv2d_2 (Conv2D)            (None, 34, 34, 128)       73856     
_________________________________________________________________
max_pooling2d_2 (MaxPooling2 (None, 17, 17, 128)       0         
_________________________________________________________________
flatten (Flatten)            (None, 36992)             0         
_________________________________________________________________
dense (Dense)                (None, 128)               4735104   
_________________________________________________________________
dense_1 (Dense)              (None, 512)               66048     
_________________________________________________________________
dense_2 (Dense)              (None, 1)                 513       
=================================================================
Total params: 4,894,913
Trainable params: 4,894,913
Non-trainable params: 0
_________________________________________________________________

Use the Adam Optimizer with binary cross entropy

model.compile(optimizer='adam',
             loss='binary_crossentropy',
             metrics=['accuracy'])

Perform Gradient Descent

Do Gradient Descent for 15 epochs

history=model.fit(train_generator,
                 validation_data=validation_generator,
                 steps_per_epoch=100,
                 epochs=15,
                 validation_steps=50,
                 verbose=2)

Epoch 1/15
100/100 - 13s - loss: 0.6821 - accuracy: 0.5425 - val_loss: 0.6484 - val_accuracy: 0.6131
Epoch 2/15
100/100 - 13s - loss: 0.6227 - accuracy: 0.6456 - val_loss: 0.6161 - val_accuracy: 0.6394
Epoch 3/15
100/100 - 13s - loss: 0.5975 - accuracy: 0.6719 - val_loss: 0.5558 - val_accuracy: 0.7206
Epoch 4/15
100/100 - 13s - loss: 0.5480 - accuracy: 0.7241 - val_loss: 0.5431 - val_accuracy: 0.7138
Epoch 5/15
100/100 - 13s - loss: 0.5182 - accuracy: 0.7447 - val_loss: 0.4839 - val_accuracy: 0.7606
Epoch 6/15
100/100 - 13s - loss: 0.4773 - accuracy: 0.7781 - val_loss: 0.5029 - val_accuracy: 0.7506
Epoch 7/15
100/100 - 13s - loss: 0.4466 - accuracy: 0.7972 - val_loss: 0.4573 - val_accuracy: 0.7912
Epoch 8/15
100/100 - 13s - loss: 0.4395 - accuracy: 0.7997 - val_loss: 0.4252 - val_accuracy: 0.8119
Epoch 9/15
100/100 - 13s - loss: 0.4314 - accuracy: 0.8019 - val_loss: 0.4931 - val_accuracy: 0.7481
Epoch 10/15
100/100 - 13s - loss: 0.4309 - accuracy: 0.7969 - val_loss: 0.4203 - val_accuracy: 0.8109
Epoch 11/15
100/100 - 13s - loss: 0.4329 - accuracy: 0.7916 - val_loss: 0.4189 - val_accuracy: 0.8069
Epoch 12/15
100/100 - 13s - loss: 0.4248 - accuracy: 0.8050 - val_loss: 0.4476 - val_accuracy: 0.7925
Epoch 13/15
100/100 - 13s - loss: 0.3868 - accuracy: 0.8306 - val_loss: 0.3900 - val_accuracy: 0.8236
Epoch 14/15
100/100 - 13s - loss: 0.3710 - accuracy: 0.8328 - val_loss: 0.4520 - val_accuracy: 0.7900
Epoch 15/15
100/100 - 13s - loss: 0.3654 - accuracy: 0.8353 - val_loss: 0.3999 - val_accuracy: 0.8100

Plot results

- Plot training and validation accuracy

Plot training and validation loss

#-----------------------------------------------------------
# Retrieve a list of list results on training and test data
# sets for each training epoch
#-----------------------------------------------------------
acc      = history.history[     'accuracy' ]
val_acc  = history.history[ 'val_accuracy' ]
loss     = history.history[    'loss' ]
val_loss = history.history['val_loss' ]

epochs   = range(len(acc)) # Get number of epochs

#------------------------------------------------
# Plot training and validation accuracy per epoch
#------------------------------------------------
plt.plot  ( epochs,     acc,label="training accuracy" )
plt.plot  ( epochs, val_acc, label='validation acuracy' )
plt.title ('Training and validation accuracy')
plt.legend()

plt.figure()

#------------------------------------------------
# Plot training and validation loss per epoch
#------------------------------------------------
plt.plot  ( epochs,     loss , label="training loss")
plt.plot  ( epochs, val_loss,label="validation loss" )
plt.title ('Training and validation loss'   )
plt.legend()

2. CNN with Image Augmentation

You can check the Cats_vs_Dogs_2.ipynb

Including the important parts of this implementation below

Use Image Augumentation

Use Image Augumentation to improve performance

Use the same model parameters as before
Perform the following image augmentation
- width, height shift
- shear and zoom
Note: Adding rotation made the performance worse

import tensorflow as tf
from tensorflow import keras
from tensorflow.keras.optimizers import RMSprop
from tensorflow.keras.preprocessing.image import ImageDataGenerator
model = tf.keras.models.Sequential([
    tf.keras.layers.Conv2D(32,(3,3),activation='relu',input_shape=(150,150,3)),
    tf.keras.layers.MaxPooling2D(2,2),
    tf.keras.layers.Conv2D(64,(3,3),activation='relu'),
    tf.keras.layers.MaxPooling2D(2,2),
    tf.keras.layers.Conv2D(128,(3,3),activation='relu'),
    tf.keras.layers.MaxPooling2D(2,2),
    tf.keras.layers.Flatten(),
    tf.keras.layers.Dense(128,activation='relu'),
    tf.keras.layers.Dense(512,activation='relu'),
    tf.keras.layers.Dense(1,activation='sigmoid')
])


train_datagen = ImageDataGenerator(
      rescale=1./255,
      #rotation_range=90,
      width_shift_range=0.2,
      height_shift_range=0.2,
      shear_range=0.2,
      zoom_range=0.2)
      #horizontal_flip=True,
      #fill_mode='nearest')

validation_datagen = ImageDataGenerator(rescale=1./255)
#
train_generator = train_datagen.flow_from_directory(train_dir,
                                                    batch_size=32,
                                                    class_mode='binary',
                                                    target_size=(150, 150))     
# --------------------
# Flow validation images in batches of 20 using test_datagen generator
# --------------------
validation_generator =  validation_datagen.flow_from_directory(validation_dir,
                                                         batch_size=32,
                                                         class_mode  = 'binary',
                                                         target_size = (150, 150))

# Use Adam Optmizer 
model.compile(optimizer='adam',
             loss='binary_crossentropy',
             metrics=['accuracy'])

Found 20000 images belonging to 2 classes.
Found 5000 images belonging to 2 classes.

Perform Gradient Descent

history=model.fit(train_generator,
                 validation_data=validation_generator,
                 steps_per_epoch=100,
                 epochs=15,
                 validation_steps=50,
                 verbose=2)

Epoch 1/15
100/100 - 27s - loss: 0.5716 - accuracy: 0.6922 - val_loss: 0.4843 - val_accuracy: 0.7744
Epoch 2/15
100/100 - 27s - loss: 0.5575 - accuracy: 0.7084 - val_loss: 0.4683 - val_accuracy: 0.7750
Epoch 3/15
100/100 - 26s - loss: 0.5452 - accuracy: 0.7228 - val_loss: 0.4856 - val_accuracy: 0.7665
Epoch 4/15
100/100 - 27s - loss: 0.5294 - accuracy: 0.7347 - val_loss: 0.4654 - val_accuracy: 0.7812
Epoch 5/15
100/100 - 27s - loss: 0.5352 - accuracy: 0.7350 - val_loss: 0.4557 - val_accuracy: 0.7981
Epoch 6/15
100/100 - 26s - loss: 0.5136 - accuracy: 0.7453 - val_loss: 0.4964 - val_accuracy: 0.7621
Epoch 7/15
100/100 - 27s - loss: 0.5249 - accuracy: 0.7334 - val_loss: 0.4959 - val_accuracy: 0.7556
Epoch 8/15
100/100 - 26s - loss: 0.5035 - accuracy: 0.7497 - val_loss: 0.4555 - val_accuracy: 0.7969
Epoch 9/15
100/100 - 26s - loss: 0.5024 - accuracy: 0.7487 - val_loss: 0.4675 - val_accuracy: 0.7728
Epoch 10/15
100/100 - 27s - loss: 0.5015 - accuracy: 0.7500 - val_loss: 0.4276 - val_accuracy: 0.8075
Epoch 11/15
100/100 - 26s - loss: 0.5002 - accuracy: 0.7581 - val_loss: 0.4193 - val_accuracy: 0.8131
Epoch 12/15
100/100 - 27s - loss: 0.4733 - accuracy: 0.7706 - val_loss: 0.5209 - val_accuracy: 0.7398
Epoch 13/15
100/100 - 27s - loss: 0.4999 - accuracy: 0.7538 - val_loss: 0.4109 - val_accuracy: 0.8075
Epoch 14/15
100/100 - 27s - loss: 0.4550 - accuracy: 0.7859 - val_loss: 0.3770 - val_accuracy: 0.8288
Epoch 15/15
100/100 - 26s - loss: 0.4688 - accuracy: 0.7688 - val_loss: 0.4764 - val_accuracy: 0.7786

Plot results

Plot training and validation accuracy
Plot training and validation loss

In [15]:

import matplotlib.pyplot as plt
#-----------------------------------------------------------
# Retrieve a list of list results on training and test data
# sets for each training epoch
#-----------------------------------------------------------
acc      = history.history[     'accuracy' ]
val_acc  = history.history[ 'val_accuracy' ]
loss     = history.history[    'loss' ]
val_loss = history.history['val_loss' ]

epochs   = range(len(acc)) # Get number of epochs

#------------------------------------------------
# Plot training and validation accuracy per epoch
#------------------------------------------------
plt.plot  ( epochs,     acc,label="training accuracy" )
plt.plot  ( epochs, val_acc, label='validation acuracy' )
plt.title ('Training and validation accuracy')
plt.legend()

plt.figure()

#------------------------------------------------
# Plot training and validation loss per epoch
#------------------------------------------------
plt.plot  ( epochs,     loss , label="training loss")
plt.plot  ( epochs, val_loss,label="validation loss" )
plt.title ('Training and validation loss'   )
plt.legend()

Implementation using Inception Network V3

The implementation is in the Colab notebook Cats_vs_Dog_3.ipynb

This is implemented as below

Use Inception V3

import os

from tensorflow.keras import layers
from tensorflow.keras import Model

  
from tensorflow.keras.applications.inception_v3 import InceptionV3
pre_trained_model = InceptionV3(input_shape = (150, 150, 3), 
                                include_top = False, 
                                weights = 'imagenet')


for layer in pre_trained_model.layers:
  layer.trainable = False
  
# pre_trained_model.summary()

last_layer = pre_trained_model.get_layer('mixed7')
print('last layer output shape: ', last_layer.output_shape)
last_output = last_layer.output

Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/inception_v3/inception_v3_weights_tf_dim_ordering_tf_kernels_notop.h5
87916544/87910968 [==============================] - 1s 0us/step
last layer output shape:  (None, 7, 7, 768)

Use Layer 7 of Inception Network

Use Image Augumentation
Use Adam Optimizer

In [0]:

import tensorflow as tf
from tensorflow import keras
from tensorflow.keras.optimizers import RMSprop
from tensorflow.keras.preprocessing.image import ImageDataGenerator
# Flatten the output layer to 1 dimension
x = layers.Flatten()(last_output)
# Add a fully connected layer with 1,024 hidden units and ReLU activation
x = layers.Dense(1024, activation='relu')(x)
# Add a dropout rate of 0.2
x = layers.Dropout(0.2)(x)                  
# Add a final sigmoid layer for classification
x = layers.Dense  (1, activation='sigmoid')(x)           

model = Model( pre_trained_model.input, x) 
#train_datagen = ImageDataGenerator( rescale = 1.0/255. )
#validation_datagen = ImageDataGenerator( rescale = 1.0/255. )

train_datagen = ImageDataGenerator(
      rescale=1./255,
      #rotation_range=90,
      width_shift_range=0.2,
      height_shift_range=0.2,
      shear_range=0.2,
      zoom_range=0.2)
      #horizontal_flip=True,
      #fill_mode='nearest')

validation_datagen = ImageDataGenerator(rescale=1./255)
#
train_generator = train_datagen.flow_from_directory(train_dir,
                                                    batch_size=32,
                                                    class_mode='binary',
                                                    target_size=(150, 150))     
# --------------------
# Flow validation images in batches of 20 using test_datagen generator
# --------------------
validation_generator =  validation_datagen.flow_from_directory(validation_dir,
                                                         batch_size=32,
                                                         class_mode  = 'binary',
                                                         target_size = (150, 150))


model.compile(optimizer='adam',
             loss='binary_crossentropy',
             metrics=['accuracy'])

Found 20000 images belonging to 2 classes.
Found 5000 images belonging to 2 classes.

Fit model

history=model.fit(train_generator,
                 validation_data=validation_generator,
                 steps_per_epoch=100,
                 epochs=15,
                 validation_steps=50,
                 verbose=2)

Epoch 1/15
100/100 - 31s - loss: 0.5961 - accuracy: 0.8909 - val_loss: 0.1919 - val_accuracy: 0.9456
Epoch 2/15
100/100 - 30s - loss: 0.2002 - accuracy: 0.9259 - val_loss: 0.1025 - val_accuracy: 0.9550
Epoch 3/15
100/100 - 30s - loss: 0.1618 - accuracy: 0.9366 - val_loss: 0.0920 - val_accuracy: 0.9581
Epoch 4/15
100/100 - 29s - loss: 0.1442 - accuracy: 0.9381 - val_loss: 0.0960 - val_accuracy: 0.9600
Epoch 5/15
100/100 - 30s - loss: 0.1402 - accuracy: 0.9381 - val_loss: 0.0703 - val_accuracy: 0.9794
Epoch 6/15
100/100 - 30s - loss: 0.1437 - accuracy: 0.9413 - val_loss: 0.1090 - val_accuracy: 0.9531
Epoch 7/15
100/100 - 30s - loss: 0.1325 - accuracy: 0.9428 - val_loss: 0.0756 - val_accuracy: 0.9670
Epoch 8/15
100/100 - 29s - loss: 0.1341 - accuracy: 0.9491 - val_loss: 0.0625 - val_accuracy: 0.9737
Epoch 9/15
100/100 - 29s - loss: 0.1186 - accuracy: 0.9513 - val_loss: 0.0934 - val_accuracy: 0.9581
Epoch 10/15
100/100 - 29s - loss: 0.1171 - accuracy: 0.9513 - val_loss: 0.0642 - val_accuracy: 0.9727
Epoch 11/15
100/100 - 29s - loss: 0.1018 - accuracy: 0.9591 - val_loss: 0.0930 - val_accuracy: 0.9606
Epoch 12/15
100/100 - 29s - loss: 0.1190 - accuracy: 0.9541 - val_loss: 0.0737 - val_accuracy: 0.9719
Epoch 13/15
100/100 - 29s - loss: 0.1223 - accuracy: 0.9494 - val_loss: 0.0740 - val_accuracy: 0.9695
Epoch 14/15
100/100 - 29s - loss: 0.1158 - accuracy: 0.9516 - val_loss: 0.0659 - val_accuracy: 0.9744
Epoch 15/15
100/100 - 29s - loss: 0.1168 - accuracy: 0.9591 - val_loss: 0.0788 - val_accuracy: 0.9669

Plot results

Plot training and validation accuracy
Plot training and validation loss

In [14]:

import matplotlib.pyplot as plt
#-----------------------------------------------------------
# Retrieve a list of list results on training and test data
# sets for each training epoch
#-----------------------------------------------------------
acc      = history.history[     'accuracy' ]
val_acc  = history.history[ 'val_accuracy' ]
loss     = history.history[    'loss' ]
val_loss = history.history['val_loss' ]

epochs   = range(len(acc)) # Get number of epochs

#------------------------------------------------
# Plot training and validation accuracy per epoch
#------------------------------------------------
plt.plot  ( epochs,     acc,label="training accuracy" )
plt.plot  ( epochs, val_acc, label='validation acuracy' )
plt.title ('Training and validation accuracy')
plt.legend()

plt.figure()

#------------------------------------------------
# Plot training and validation loss per epoch
#------------------------------------------------
plt.plot  ( epochs,     loss , label="training loss")
plt.plot  ( epochs, val_loss,label="validation loss" )
plt.title ('Training and validation loss'   )
plt.legend()

I intend to do some interesting stuff with Convolutional Neural Networks.

Watch this space!

To see all posts click Index of posts

Big Data 6: The T20 Dance of Apache NiFi and yorkpy

“I don’t count my sit-ups. I only start counting once it starts hurting. ”

Muhammad Ali

“Hard work beats talent when talent doesn’t work hard.”

Tim Notke

In my previous post Big Data 5: kNiFI-ing through cricket data with Apache NiFi and yorkpy, I created a Big Data Pipeline that takes raw data in YAML format from a Cricsheet to processing and ranking IPL T20 players. In that post I had mentioned that we could create a similar pipeline to create a real time dashboard of IPL Analytics. I could have have done this but I needed to know how to create a Web UI. After digging and poking around, I have been able to create a simple Web UI running off Apache Web server. This UI uses basic JQuery and CSS to display a real time IPL T20 dashboard. As in my previous post, this is an end-2-end Big Data pipeline which can handle large data sets at scheduled times, process them and generate real time dashboards.

We could imagine an inter-galactic T20 championship league where T20 data comes in every hour or sooner and we need to perform analytics to see if us earthlings are any better than people with pointy heads or little green men. The NiFi pipeline could be used as-is, however the yorkpy package would have to be rewritten in Pyspark. That is in another eon, though.

My package yorkpy has around ~45+ functions which fall in the following main categories

1. Pitching yorkpy . short of good length to IPL – Part 1 :Class 1: This includes functions that convert the yaml data of IPL matches into Pandas dataframe which are then saved as CSV. This part can perform analysis of individual IPL matches.
2. Pitching yorkpy.on the middle and outside off-stump to IPL – Part 2 :Class 2:This part includes functions to create a large data frame for head-to-head confrontation between any 2IPL teams says CSK-MI, DD-KKR etc, which can be saved as CSV. Analysis is then performed on these team-2-team confrontations.
3. Pitching yorkpy.swinging away from the leg stump to IPL – Part 3 Class 3:The 3rd part includes the performance of any IPL team against all other IPL teams. The data can also be saved as CSV.
4. Pitching yorkpy … in the block hole – Part 4 :Class 4: This part performs analysis of individual IPL batsmen and bowlers

Watch the live demo of the end-2-end NiFi pipeline at ‘The T20 Dance‘

You can download the NiFi template and associated code from Github at T20 Dance

The Apache NiFi Pipeline is shown below

1. T20 Dance – Overall NiFi Pipeline

There are 5 process groups

2. ListAndConvertYaml2DataFrames

This post starts with having the YAML files downloaded and unpacked from Cricsheet. The individual YAML files are converted into Pandas dataframes and saved as CSV. A concurrency of 12 is used to increase performance and process YAML files in parallel. The processor MergeContent creates a merged content to signal the completion of conversion and triggers the other Process Groups through a funnel.

3. Analyse individual IPL T20 matches

This Process Group ‘Analyse T20 matches’ used the yorkpy’s Class 1 functions which can perform analysis of individual IPL T20 matches. The matchWorm() and matchScorecard() functions are used, through any other function could have been used. The Process Group is shown below

4. Analyse performance of an IPL team in all matches against another IPL team

This Process Group ‘Analyse performance of IPL team in all matched against another IPL team‘ does analysis in all matches between any 2 IPL teams (Class 2) as shown below

5. Analyse performance of IPL team in all matches against all other IPL teams

This uses Class 3 functions. Individual data sets for each IPL team versus all other IPL teams is created before Class 3 yorkpy functions are invoked. This is included below

6. Analyse performances of IPL batsmen and bowlers

This Process Group uses Class 4 yorkpy functions. The match CSV files are processed to get batting and bowling details before calling the individual functions as shown below

7. IPL T20 Dashboard

The IPL T20 Dashboard is shown

Conclusion

This NiFI pipeline was done for IPL T20 however, it could be done for any T20 format like Intl T20, BBL, Natwest etc which are posted in Cricsheet. Also, only a subset of the yorkpy functions were used. There is a much wider variety of functions available.

Hope the T20 dance got your foot a-tapping!

To see all posts click Index of posts

Big Data-5: kNiFi-ing through cricket data with yorkpy

“The temptation to form premature theories upon insufficient data is the bane of our profession.”

Sherlock Holmes in the Valley of fear by Arthur Conan Doyle

“If we have data, let’s look at data. If all we have are opinions, let’s go with mine.”

Jim Barksdale, former CEO Netscape

In this post I use Apache NiFi Dataflow Pipeline along with my Python package yorkpy to crunch through cricket data from Cricsheet. The Data Pipelne flows all the way from the source to target analytics output. Apache NiFi was created to automate the flow of data between systems. NiFi dataflows enable the automated and managed flow of information between systems. This post automates the flow of data from Cricsheet, from where the zip file it is downloaded, unpacked, processed, transformed and finally T20 players are ranked.

While this is a straight forward example of what can be done, this pattern can be applied to real Big Data systems. For example hypothetically, we could consider that we get several parallel streams of cricket data or for that matter any sports related data. There could be parallel Data flow pipelines that get the data from the sources. This would then be followed by data transformation modules and finally a module for generating analytics. At the other end a UI based on AngularJS or ReactJS could display the results in a cool and awesome way.

Incidentally, the NiFi pipeline that I discuss in this post, is a simplistic example, and does not use the Big Data stack like HDFS, Hive, Spark etc. Nevertheless, the pattern used, has all the modules for a Big Data pipeline namely ingestion, unpacking, transformation and finally analytics. This NiF pipeline demonstrates the flow using the regular file system of Mac and my python based package yorkpy. The concepts mentioned could be used in a real Big Data scenario which has much fatter pipes of data coming. If this was the case the NiFi pipeline would utilize HDFS/Hive for storing the ingested data and Pyspark/Scala for the transformation and analytics and other related technologies.

A pictorial representation is given below

In the diagram above each of the vertical boxes could be any technology from the ever proliferating Big Data stack namely HDFS, Hive, Spark, Sqoop, Kafka, Impala and so on. Such a dataflow automation could be created when any big sporting event happens, as long as the data generated large, and there is a need for dynamic and automated reporting. The UI could be based on AngularJS/ReactJS and could display analytical tables and charts.

This post demonstrates one such scenario in which IPL T20 data is downloaded from Cricsheet site, unpacked and stored in a specific directory. This dataflow automation is based on my yorkpy package. To know more about the yorkpy package see Pitching yorkpy … short of good length to IPL – Part 1 and the associated parts. The zip file, from Cricsheet, contains individual IPL T20 matches in YAML format. The convertYaml2DataframeT20() function is used to convert the YAML files into Pandas dataframes before storing them as CSV files. After this done, the function rankIPLT20batting() function is used to perform the overall ranking of the T20 players. My yorkpy Python package has about ~ 50+ functions that perform various analytics on any T20 data for e.g it has the following classes of functions

analyze T20 matches
analyze performance of a T20 team in all matches against another T20 team
analyze performance of a T20 team against all other T20 teams
analyze performance of T20 batsman and bowlers
rank T20 batsmen and bowlers

The functions of yorkpy generate tables or charts. While this post demonstrates one scenario, we could use any of the yorkpy T20 functions, generate the output and display on a widget in the UI display, created with cool technologies like AngularJS/ReactJS, possibly in near real time as data keeps coming in.,

To use yorkpy with NiFI the following packages have to be installed in your environment

-pip install yorkpy
-pip install pyyaml
-pip install pandas
-yum install python-devel (equivalent in Windows)
-pip install matplotlib
-pip install seaborn
-pip install sklearn
-pip install datetime

I have created a video of the NiFi Pipeline with the real dataflow fro source to the ranked IPL T20 batsmen. Take a look at RankingT20PlayersWithNiFiYorkpy

You can clone/fork the NiFi template from rankT20withNiFiYorkpy

The NiFi Data Flow Automation is shown below

1. Overall flow

The overall NiFi flow contains 2 Process Groups a) DownloadAnd Unpack. b) Convert and Rank IPL batsmen. While it appears that the Process Groups are disconnected, they are not. The first process group downloads the T20 zip file, unpacks the. zip file and saves the YAML files in a specific folder. The second process group monitors this folder and starts processing as soon the YAML files are available. It processes the YAML converting it into dataframes before storing it as CSV file. The next processor then does the actual ranking of the batsmen before writing the output into IPLrank.txt

1.1 DownloadAndUnpack Process Group

This process group is shown below

1.1.1 GetT20Data

The GetT20Data Processor downloads the zip file given the URL

The ${T20data} variable points to the specific T20 format that needs to be downloaded. I have set this to https://cricsheet.org/downloads/ipl.zip. This could be set any other data set. In fact we could have parallel data flows for different T20/ Sports data sets and generate

1.1.2 SaveUnpackedData

This processor stores the YAML files in a predetermined folder, so that the data can be picked up by the 2nd Process Group for processing

1.2 ProcessAndRankT20Players Process Group

This is the second process group which converts the YAML files to pandas dataframes before storing them as. CSV files. The RankIPLPlayers will then read all the CSV files, stack them and then proceed to rank the IPL players. The Process Group is shown below

1.2.1 ListFile and FetchFile Processors

The left 2 Processors ListFile and FetchFile get all the YAML files from the folder and pass it to the next processor

1.2.2 convertYaml2DataFrame Processor

The convertYaml2DataFrame Processor uses the ExecuteStreamCommand which call a python script. The Python script invoked the yorkpy function convertYaml2Dataframe() as shown below

The ${convertYaml2Dataframe} variable points to the python file below which invoked the yorkpy function yka.convertYaml2PandasDataframeT20()

import yorkpy.analytics as yka
import argparse
parser = argparse.ArgumentParser(description='convert')
parser.add_argument("yamlFile",help="YAML File")
args=parser.parse_args()
yamlFile=args.yamlFile
yka.convertYaml2PandasDataframeT20(yamlFile,"/Users/tvganesh/backup/software/nifi/ipl","/Users/tvganesh/backup/software/nifi/ipldata")

This function takes as input $filename which comes from FetchFile processor which is a FlowFile. So I have added a concurrency of 8 to handle upto 8 Flowfiles at a time. The thumb rule as I read on the internet is 2x, 4x the number of cores of your system. Since I have an 8 core Mac, I could possibly have gone ~ 30 concurrent threads. Also the number of concurrent threads is less when the flow is run in a Oracle Box VirtualMachine. Box since a vCore < actual Core

The scheduling tab is as below

Here are the 8 concurrent Python threads on Mac at bottom right… (pretty cool!)

I have not fully tested how latency vs throughput slider changes, affects the performance.

1.2.3 MergeContent Processor

This processor’s only job is to trigger the rankIPLPlayers when all the FlowFiles have merged into 1 file.

1.2.4 RankT20Players

This processor is an ExecuteStreamCommand Processor that executes a Python script which invokes a yorkpy function rankIPLT20Batting()

import yorkpy.analytics as yka
rank=yka.rankIPLT20Batting("/Users/tvganesh/backup/software/nifi/ipldata")
print(rank.head(15))

1.2.5 OutputRankofT20Player Processor

This processor writes the generated rank to an output file.

1.3 Final Ranking of IPL T20 players

The Nodejs based web server picks up this file and displays on the web page the final ranks (the code is based on a good youtube for reading from file)

2. Final thoughts

As I have mentioned above though the above NiFi Cricket Dataflow automation does not use the Hadoop ecosystem, the pattern used is valid and can be used with some customization in Big Data flows as parallel stream. I could have also done this on Oracle VirtualBox but I thought since the code is based on Python and Pandas there is no real advantage of running on the VirtualBox. GIve the NiFi flow a shot. Have fun!!!

Also see
1.My book ‘Deep Learning from first Practical Machine Learning with R and Python – Part 5
Edition’ now on Amazon
2. Introducing QCSimulator: A 5-qubit quantum computing simulator in R
3.De-blurring revisited with Wiener filter using OpenCV
4. Practical Machine Learning with R and Python – Part 5
5. Natural language processing: What would Shakespeare say?
6.Getting started with Tensorflow, Keras in Python and R
7.Revisiting World Bank data analysis with WDI and gVisMotionChart

To see all posts click Index of posts

Ranking T20 players in Intl T20, IPL, BBL and Natwest using yorkpy

There is a voice that doesn’t use words, listen.
When someone beats a rug, the blows are not against the rug, but against the dust in it.
I lost my hat while gazing at the moon, and then I lost my mind.
Rumi

Introduction

After a long hiatus, I am back to my big, bad, blogging ways! In this post I rank T20 players from several different leagues namely

International T20
Indian Premier League (IPL) T20
Big Bash League (BBL) T20
Natwest Blast (NTB) T20

I have added 8 new functions to my Python Package yorkpy, which will perform the ranking for the above 4 T20 League formats. To know more about my Python package see Pitching yorkpy . short of good length to IPL – Part 1, and the related posts on yorkpy. The code can be easily extended to other leagues which have a the same ‘yaml’ format for the matches. I also fixed some issues which started to crop up, possibly because a few things have changed in the new data.

The new functions are

rankIntlT20Batting()
rankIntlT20Batting()
rankIPLT20Batting()
rankIPLT20Batting
rankBBLT20Batting()
rankBBLT20Batting()
rankNTBT20Batting()
rankNTBT20Batting()

The yorkpy package uses data from Cricsheet

You can clone/fork the code for yorkpy at yorkpy

You can download the PDF of the post from Rank T20

yorkpy can be installed with ‘pip install yorkpy‘

1. International T20

The steps to do before ranking for International T20 matches are 1. Download International T20 zip file from Cricsheet Intl T20 2. Unzip the file. This will create a folder with yaml files

import yorkpy.analytics as yka
#yka.convertAllYaml2PandasDataframesT20("../t20s","../data")

This above step will convert the yaml files into CSV files. Now do the ranking as below

1a. Ranking of International T20 batsmen

import yorkpy.analytics as yka
intlT20RankBatting=yka.rankIntlT20Batting("C:\\software\\cricket-package\\yorkpyPkg\\data\\data")
intlT20RankBatting.head(15)

##                      matches  runs_mean     SR_mean
## batsman                                            
## V Kohli                   58  38.672414  125.212402
## KS Williamson             42  32.595238  122.884631
## Mohammad Shahzad          52  31.942308  118.212288
## CH Gayle                  50  31.140000  111.869984
## BB McCullum               69  29.492754  117.011666
## MM Lanning                48  28.812500   98.582663
## SJ Taylor                 44  28.659091   98.684856
## MJ Guptill                68  28.573529  117.673702
## DA Warner                 71  28.507042  121.142746
## DPMD Jayawardene          53  27.584906  107.787092
## KC Sangakkara             54  26.407407  106.039838
## JP Duminy                 68  26.294118  114.606717
## TM Dilshan                78  26.243590   97.910384
## RG Sharma                 65  25.907692  113.056548
## H Masakadza               53  25.566038   99.453880

1b. Ranking of International T20 bowlers

import yorkpy.analytics as yka
intlT20RankBowling=yka.rankIntlT20Bowling("C:\\software\\cricket-package\\yorkpyPkg\\data\\data")

intlT20RankBowling.head(15)

##                       matches  wicket_mean  econrate_mean
## bowler                                                   
## Umar Gul                   58     1.603448       7.637931
## SL Malinga                 78     1.500000       7.409188
## Saeed Ajmal                63     1.492063       6.451058
## DW Steyn                   46     1.478261       7.014855
## A Shrubsole                45     1.422222       6.294444
## M Morkel                   41     1.292683       7.680894
## KMDN Kulasekara            57     1.280702       7.476608
## TG Southee                 51     1.274510       8.759804
## SCJ Broad                  53     1.264151            inf
## Shakib Al Hasan            58     1.241379       6.836207
## R Ashwin                   44     1.204545       7.162879
## Nida Dar                   44     1.204545       6.083333
## KH Brunt                   44     1.204545       5.982955
## KD Mills                   42     1.166667       8.289683
## SR Watson                  46     1.152174       8.246377

2. Indian Premier League (IPL) T20

The steps to do before ranking for IPL T20 matches are 1. Download IPL T20 zip file from Cricsheet IPL T20 2. Unzip the file. This will create a folder with yaml files

import yorkpy.analytics as yka
#yka.convertAllYaml2PandasDataframesT20("../ipl","../ipldata")

This above step will convert the yaml files into CSV files in the /ipldata folder. Now do the ranking as below

2a. Ranking of batsmen in IPL T20

import yorkpy.analytics as yka
IPLT20RankBatting=yka.rankIPLT20Batting("C:\\software\\cricket-package\\yorkpyPkg\\data\\ipldata")
IPLT20RankBatting.head(15)

##                    matches  runs_mean     SR_mean
## batsman                                          
## DA Warner              129  37.589147  119.917864
## CH Gayle               123  36.723577  125.256818
## SE Marsh                70  36.314286  114.707578
## KL Rahul                59  33.542373  123.424971
## MEK Hussey              60  33.400000  100.439187
## V Kohli                174  32.413793  115.830849
## KS Williamson           42  31.690476  120.443172
## AB de Villiers         143  30.923077  128.967081
## JC Buttler              45  30.800000  132.561154
## AM Rahane              118  30.330508  102.240398
## SR Tendulkar            79  29.949367  101.651959
## F du Plessis            65  29.415385  112.462114
## Q de Kock               51  29.333333  110.973836
## SS Iyer                 47  29.170213  102.144222
## G Gambhir              155  28.741935  103.997558

2b. Ranking of bowlers in IPL T20

import yorkpy.analytics as yka
IPLT20RankBowling=yka.rankIPLT20Bowling("C:\\software\\cricket-package\\yorkpyPkg\\data\\ipldata")
IPLT20RankBowling.head(15)

##                      matches  wicket_mean  econrate_mean
## bowler                                                  
## SL Malinga               122     1.540984       7.173361
## Imran Tahir               43     1.465116       8.155039
## A Nehra                   88     1.375000       7.923295
## MJ McClenaghan            56     1.339286       8.638393
## Rashid Khan               46     1.304348       6.543478
## Sandeep Sharma            79     1.303797       7.860759
## MM Patel                  63     1.301587       7.530423
## DJ Bravo                 131     1.282443       8.458333
## M Morkel                  70     1.257143       7.760714
## SP Narine                109     1.256881       6.747706
## YS Chahal                 83     1.228916       8.103659
## R Vinay Kumar            104     1.221154       8.556090
## RP Singh                  82     1.219512       8.149390
## CH Morris                 52     1.211538       7.854167
## B Kumar                  117     1.205128       7.536325

3. Natwest T20

The steps to do before ranking for Natwest T20 matches are 1. Download Natwest T20 zip file from Cricsheet NTB T20 2. Unzip the file. This will create a folder with yaml files

import yorkpy.analytics as yka
#yka.convertAllYaml2PandasDataframesT20("../ntb","../ntbdata")

This above step will convert the yaml files into CSV files in the /ntbdata folder. Now do the ranking as below

3a. Ranking of NTB batsmen

import yorkpy.analytics as yka
NTBT20RankBatting=yka.rankNTBT20Batting("C:\\software\\cricket-package\\yorkpyPkg\\data\\ntbdata")
NTBT20RankBatting.head(15)

##                      matches  runs_mean     SR_mean
## batsman                                            
## Babar Azam                13  44.461538  121.268809
## T Banton                  13  42.230769  139.376274
## JJ Roy                    12  41.250000  142.182147
## DJM Short                 12  40.250000  131.182294
## AN Petersen               12  37.916667  132.522727
## IR Bell                   13  37.615385  130.104721
## M Klinger                 26  35.346154  112.682922
## EJG Morgan                16  35.062500  129.817650
## AJ Finch                  19  34.578947  137.093465
## MH Wessels                26  33.884615  116.300969
## S Steel                   11  33.545455  140.118207
## DJ Bell-Drummond          21  33.142857  108.566309
## Ashar Zaidi               11  33.000000  178.553331
## DJ Malan                  26  33.000000  120.127202
## T Kohler-Cadmore          23  32.956522  112.493019

3b. Ranking of NTB bowlers

import yorkpy.analytics as yka
NTBT20RankBowling=yka.rankNTBT20Bowling("C:\\software\\cricket-package\\yorkpyPkg\\data\\ntbdata")
NTBT20RankBowling.head(15)

##                        matches  wicket_mean  econrate_mean
## bowler                                                    
## MW Parkinson                11     2.000000       7.628788
## HF Gurney                   23     1.956522       8.831884
## GR Napier                   12     1.916667       8.694444
## R Rampaul                   19     1.736842       7.131579
## P Coughlin                  11     1.727273       8.909091
## AJ Tye                      26     1.692308       8.227564
## GC Viljoen                  12     1.666667       7.708333
## BAC Howell                  21     1.666667       6.857143
## BW Sanderson                12     1.583333       7.902778
## KJ Abbott                   14     1.571429       9.398810
## JE Taylor                   13     1.538462       9.839744
## JDS Neesham                 12     1.500000      10.812500
## MJ Potts                    12     1.500000       8.486111
## TT Bresnan                  21     1.476190       8.817460
## T van der Gugten            13     1.461538       7.211538

4. Big Bash Leagure (BBL) T20

The steps to do before ranking for BBL T20 matches are 1. Download BBL T20 zip file from Cricsheet BBL T20 2. Unzip the file. This will create a folder with yaml files

import yorkpy.analytics as yka
#yka.convertAllYaml2PandasDataframesT20("../bbl","../bbldata")

This above step will convert the yaml files into CSV files in the /bbldata folder. Now do the ranking as below

4a. Ranking of BBL batsmen

import yorkpy.analytics as yka
BBLT20RankBatting=yka.rankBBLT20Batting("C:\\software\\cricket-package\\yorkpyPkg\\data\\bbldata")
BBLT20RankBatting.head(15)

##                 matches  runs_mean     SR_mean
## batsman                                       
## DJM Short            43  40.883721  118.773047
## SE Marsh             47  39.148936  113.616053
## AJ Finch             62  36.306452  120.271231
## AT Carey             37  34.945946  120.125341
## UT Khawaja           41  31.268293  107.355655
## CA Lynn              74  31.162162  121.746578
## MS Wade              46  30.782609  120.310081
## TM Head              45  30.000000  126.769564
## MEK Hussey           23  29.173913  109.492934
## BJ Hodge             29  29.000000  124.438040
## BR Dunk              39  28.230769  106.149913
## AD Hales             31  27.161290  117.678008
## BB McCullum          34  27.058824  115.486392
## GJ Bailey            57  27.000000  121.159220
## MR Marsh             47  26.510638  114.994909

4b. Ranking of BBL bowlers

import yorkpy.analytics as yka
BBLT20RankBowling=yka.rankBBLT20Bowling("C:\\software\\cricket-package\\yorkpyPkg\\data\\bbldata")
BBLT20RankBowling.head(15)

##                    matches  wicket_mean  econrate_mean
## bowler                                                
## Yasir Arafat            15     2.000000       7.587778
## CH Morris               15     1.733333       8.572222
## TK Curran               27     1.629630       8.716049
## TT Bresnan              13     1.615385       8.775641
## JR Hazlewood            18     1.555556       7.361111
## CJ McKay                15     1.533333       8.555556
## DR Sams                 36     1.527778       8.581019
## AC McDermott            14     1.500000       9.166667
## JP Faulkner             20     1.500000       8.345833
## SP Narine               12     1.500000       7.395833
## AJ Tye                  51     1.490196       8.101307
## M Kelly                 21     1.476190       8.908730
## SA Abbott               73     1.438356       8.737443
## B Laughlin              82     1.426829       8.332317
## SW Tait                 31     1.419355       8.895161

Conclusion

You should be able to now rank players in the above formats as new data is added to Cricsheet. yorkpy can also be used for other leagues which follow the Cricsheet format.

Also see
1. Deep Learning from first principles in Python, R and Octave – Part 5
2. Using Linear Programming (LP) for optimizing bowling change or batting lineup in T20 cricket
3. Using Reinforcement Learning to solve Gridworld
4. Big Data-4: Webserver log analysis with RDDs, Pyspark, SparkR and SparklyR
5. My book ‘Practical Machine Learning in R and Python: Third edition’ on Amazon
6. Deblurring with OpenCV: Weiner filter reloaded
7. Rock N’ Roll with Bluemix, Cloudant & NodeExpress
8. Modeling a Car in Android

To see all posts click Index of posts

Using Reinforcement Learning to solve Gridworld

“Take up one idea. Make that one idea your life — think of it, dream of it, live on that idea. Let the brain, muscles, nerves, every part of your body, be full of that idea, and just leave every other idea alone. This is the way to success.”

– Swami Vivekananda

“Be the change you want to see in the world”

– Mahatma Gandhi

“If you want to shine like the sun, first burn like the sun”

-Shri A.P.J Abdul Kalam

Reinforcement Learning

Reinforcement Learning (RL) involves decision making under uncertainty which tries to maximize return over successive states.There are four main elements of a Reinforcement Learning system: a policy, a reward signal, a value function. The policy is a mapping from the states to actions or a probability distribution of actions. Every action the agent takes results in a numerical reward. The agent’s sole purpose is to maximize the reward in the long run.

Reinforcement Learning is very different from Supervised, Unsupervised and Semi-supervised learning where the data is either labeled, unlabeled or partially labeled and the learning algorithm tries to learn the target values from the input features which is then used either for inference or prediction. In unsupervised the intention is to extract patterns from the data. In Reinforcement Learning the agent/robot takes action in each state based on the reward it would get for a particular action in a specific state with the goal of maximizing the reward. In many ways Reinforcement Learning is similar to how human beings and animals learn. Every action we take is with the goal of increasing our overall happiness, contentment, money,fame, power over the opposite!

RL has been used very effectively in many situations, the most famous is AlphaGo from Deep Mind, the first computer program to defeat a professional Go player in the Go game, which is supposed to be extremely complex. Also AlphaZero, from DeepMind has a higher ELO rating than that of Stockfish and was able to beat Stockfish 800+ times in 1000 chess matches. Take a look at DeepMind

In this post, I use some of the basic concepts of Reinforcment Learning to solve Grids (mazes). With this we can solve mazes, with arbitrary size, shape and complexity fairly easily. The RL algorithm can find the optimal path through the maze. Incidentally, I recollect recursive algorithms in Data Structures book which take a much more complex route with a lot of back tracking to solve maze problems

Reinforcement Learning involves decision making under uncertainty which tries to maximize return over successive states.There are four main elements of a Reinforcement Learning system: a policy, a reward signal, a value function. The policy is a mapping from the states to actions or a probability distribution of actions. Every action the agent takes results in a numerical reward. The agent’s sole purpose is to maximize the reward in the long run.

The reward indicates the immediate return, a value function specifies the return in the long run. Value of a state is the expected reward that an agent can accrue.

The agent/robot takes an action in At in state St and moves to state S’t anf gets a reward Rt+1 as shown

An agent will seek to maximize the overall return as it transition across states
The expected return can be expressed as
$G_{t} = R_{t+1} + \gamma G_{t+1}$ where $G_{t}$ is the expected return in time t and the discounted expected return $G_{t+1}$ in time t+1

A policy is a mapping from states to probabilities of selecting each possible action. If the agent is following policy $\pi$ at time t, then $\pi(a|s)$ is the probability that $A_{t}$ = a if $S_{t}$ = s.

The value function of a state s under a policy $\pi$ , denoted $v_{\pi}(s)$ , is the expected return when starting in s and following $\pi$ thereafter

This can be written as

$v_{\pi}(s) = E_{\pi}[G_{t} |S_{t}=s] = E_{\pi}[\sum_{k=0}^{k=Inf} \gamma^{k}R_{t+k+1}|S_{t}=s]$

= $E_{\pi}[R_{t+1} + \gamma G_{t+1} |S_{t}=s]$

$v_{\pi}(s)=\sum_{a} \pi(a|s) \sum_{s',r} p(s',r|s,a)[r+\gamma*v_{\pi}(s')]$

Similarly the action value function gives the expected return when taking an action ‘a’ in state ‘s’
$q_{\pi}(s,a)= \sum_{s',r} p(s',r|s,a)[r+\gamma*\pi(a|s)q_{\pi}(s',a')]$

These are Bellman’s equation for the state value function

The Bellman equations give the equation for each of the state

The Bellman optimality equations give the optimal policy of choosing specific actions in specific states to achieve the maximum reward and reach the goal efficiently. They are given as

$v_{*}(s)=max_{a}\sum_{s',r} p(s',r|s,a)[r+\gamma*v_{*}(s')]$

$q_{*}(s,a)=\sum_{s',r} p(s',r|s,a)[r+\gamma*max_{a}q_{*}(s',a')]$

The Bellman equations cannot be used directly in goal directed problems and dynamic programming is used instead where the value functions are computed iteratively

n this post I solve Grids using Reinforcement Learning. In the problem below the Maze has 2 end states as shown in the corner. There are four possible actions in each state up, down, right and left. If an action in a state takes it out of the grid then the agent remains in the same state. All actions have a reward of -1 while the end states have a reward of 0

This is shown as

where the reward for any transition is Rt=−1 except the transition to the end states at the corner which have a reward of 0. The policy is a uniform policy with all actions being equi-probable with a probability of 1/4 or 0.25

You can fork/clone the code from my Github repository – Gridworld

Note: This post shows 3 different grids each with slightly more complexity and uses 3 methods

a) Bellman Update

b) Greedification

c) Bellman Optimality Update

with dynamic programming to solve the Grids

1. Gridworld-1

In [1]:

import numpy as np
import random

In [2]:

gamma = 1 # discounting rate
gridSize = 4
rewardValue = -1
terminationStates = [[0,0], [gridSize-1, gridSize-1]]
actions = [[-1, 0], [1, 0], [0, 1], [0, -1]]
numIterations = 1000

The action value provides the next state for a given action in a state and the accrued reward

In [3]:

def actionValue(initialPosition,action):
    if initialPosition in terminationStates:
        finalPosition = initialPosition
        reward=0
    else:
        #Compute final position
        finalPosition = np.array(initialPosition) + np.array(action)
        reward= rewardValue
    # If the action moves the finalPosition out of the grid, stay in same cell
    if -1 in finalPosition or gridSize in finalPosition:
        finalPosition = initialPosition
        reward= rewardValue
    
    #print(finalPosition)
    return finalPosition, reward

1a. Bellman Update

In [4]:

# Initialize valueMap and valueMap1
valueMap = np.zeros((gridSize, gridSize))
valueMap1 = np.zeros((gridSize, gridSize))
states = [[i, j] for i in range(gridSize) for j in range(gridSize)]

In [5]:

def policy_evaluation(numIterations,gamma,theta,valueMap):
    for i in range(numIterations):
        delta=0
        for state in states:
            weightedRewards=0
            for action in actions:
                finalPosition,reward = actionValue(state,action)
                weightedRewards += 1/4* (reward + gamma * valueMap[finalPosition[0],finalPosition][1])
            valueMap1[state[0],state[1]]=weightedRewards
            delta =max(delta,abs(weightedRewards-valueMap[state[0],state[1]]))
        valueMap = np.copy(valueMap1)
        if(delta < 0.01):                                                
            print(valueMap)
            break

In [6]:

valueMap = np.zeros((gridSize, gridSize))
valueMap1 = np.zeros((gridSize, gridSize))
states = [[i, j] for i in range(gridSize) for j in range(gridSize)]
policy_evaluation(1000,1,0.001,valueMap)

[[  0.         -13.89528403 -19.84482978 -21.82635535]
 [-13.89528403 -17.86330422 -19.84586777 -19.84482978]
 [-19.84482978 -19.84586777 -17.86330422 -13.89528403]
 [-21.82635535 -19.84482978 -13.89528403   0.        ]]

Findings

The valueMap is the result of several sweeps through all the states. It can be seen that the cells in the corner state have a higher value. We can start on any cell in the grid and move in the direction which is greater than the current state and we will reach the end state

1b. Greedify

The previous alogirthm while it works is somewhat inefficient as we have to sweep over the states to compute the state value function. The approach below works on the same problem but after each computation of the value function, a greedifications takes place to ensure that the action with the highest return is selected after which the policy π is followed

To make the transitions clearer I also create another grid which shows the path from any cell to the end states as

‘u’ – up

‘d’ – down

‘r’ – right

‘l’ – left

Important note: If there are several alternative actions with equal value then the algorithm will break the tie randomly

In [7]:

valueMap = np.zeros((gridSize, gridSize))
valueMap1 = np.zeros((gridSize, gridSize))
states = [[i, j] for i in range(gridSize) for j in range(gridSize)]
pi = np.ones((gridSize,gridSize))/4
pi1 = np.chararray((gridSize, gridSize))
pi1[:] = 'a'

In [8]:

# Compute the value state function for the Grid
def policy_evaluate(states,actions,gamma,valueMap):
    #print("iterations=",i)
    for state in states:
        weightedRewards=0
        for action in actions:
            finalPosition,reward = actionValue(state,action)
            weightedRewards += 1/4* (reward + gamma * valueMap[finalPosition[0],finalPosition][1])
        # Set the computed weighted rewards to valueMap1
        valueMap1[state[0],state[1]]=weightedRewards
    # Copy to original valueMap
    valueMap = np.copy(valueMap1)
    return(valueMap)

In [9]:

def argmax(q_values):
    idx=np.argmax(q_values)
    return(np.random.choice(np.where(a==a[idx])[0].tolist()))


# Compute the best action in each state
def greedify_policy(state,pi,pi1,gamma,valueMap):  
        q_values=np.zeros(len(actions))
        for idx,action in enumerate(actions):
            finalPosition,reward = actionValue(state,action)
            q_values[idx] += 1/4* (reward + gamma * valueMap[finalPosition[0],finalPosition][1])
        # Find the index of the action for which the q_value is 
        idx=q_values.argmax()
        pi[state[0],state[1]]=idx 
        if(idx == 0):
            pi1[state[0],state[1]]='u'
        elif(idx == 1):
            pi1[state[0],state[1]]='d'
        elif(idx == 2):
            pi1[state[0],state[1]]='r'
        elif(idx == 3):
            pi1[state[0],state[1]]='l'

In [10]:

def improve_policy(pi, pi1,gamma,valueMap):
    policy_stable = True
    for state in states:
        old = pi[state].copy()
        # Greedify policy for state
        greedify_policy(state,pi,pi1,gamma,valueMap)
        if not np.array_equal(pi[state], old):
            policy_stable = False
    print(pi)
    print(pi1)
    return pi, pi1, policy_stable

In [11]:

def policy_iteration(gamma, theta):
    valueMap = np.zeros((gridSize, gridSize))
    pi = np.ones((gridSize,gridSize))/4
    pi1 = np.chararray((gridSize, gridSize))
    pi1[:] = 'a'
    policy_stable = False
    print("here")
    while not policy_stable:
        valueMap = policy_evaluate(states,actions,gamma,valueMap)
        pi,pi1, policy_stable = improve_policy(pi,pi1,  gamma,valueMap)
    return valueMap, pi,pi1

In [12]:

theta=0.1
valueMap, pi,pi1 = policy_iteration(gamma, theta)

[[0. 3. 0. 0.]
 [0. 0. 0. 0.]
 [0. 0. 0. 1.]
 [0. 0. 2. 0.]]
[[b'u' b'l' b'u' b'u']
 [b'u' b'u' b'u' b'u']
 [b'u' b'u' b'u' b'd']
 [b'u' b'u' b'r' b'u']]
[[0. 3. 3. 0.]
 [0. 0. 0. 1.]
 [0. 0. 1. 1.]
 [0. 2. 2. 0.]]
[[b'u' b'l' b'l' b'u']
 [b'u' b'u' b'u' b'd']
 [b'u' b'u' b'd' b'd']
 [b'u' b'r' b'r' b'u']]
[[0. 3. 3. 1.]
 [0. 0. 1. 1.]
 [0. 0. 1. 1.]
 [0. 2. 2. 0.]]
[[b'u' b'l' b'l' b'd']
 [b'u' b'u' b'd' b'd']
 [b'u' b'u' b'd' b'd']
 [b'u' b'r' b'r' b'u']]
[[0. 3. 3. 1.]
 [0. 0. 1. 1.]
 [0. 0. 1. 1.]
 [0. 2. 2. 0.]]
[[b'u' b'l' b'l' b'd']
 [b'u' b'u' b'd' b'd']
 [b'u' b'u' b'd' b'd']
 [b'u' b'r' b'r' b'u']]

Findings

From the above valueMap we can see that greedification solves this much faster as below

1c. Bellman Optimality update

The Bellman optimality update directly updates the value state function for the action that results in the maximum return in a state

In [13]:

gamma = 1 # discounting rate
rewardValue = -1
gridSize = 4
terminationStates = [[0,0], [gridSize-1, gridSize-1]]
actions = [[-1, 0], [1, 0], [0, 1], [0, -1]]
numIterations = 1000

In [14]:

valueMap = np.zeros((gridSize, gridSize))
valueMap1 = np.zeros((gridSize, gridSize))
states = [[i, j] for i in range(gridSize) for j in range(gridSize)]
pi = np.ones((gridSize,gridSize))/4
pi1 = np.chararray((gridSize, gridSize))
pi1[:] = 'a'

In [15]:

def bellman_optimality_update(valueMap, state, gamma):

    q_values=np.zeros(len(actions))
    
    for idx,action in enumerate(actions):
        finalPosition,reward = actionValue(state,action)
        q_values[idx] += 1/4* (reward + gamma * valueMap[finalPosition[0],finalPosition][1])
    # Find the index of the action for which the q_value is 
    idx=q_values.argmax()
            
    max = np.argmax(q_values)
    valueMap[state[0],state[1]] = q_values[max]    
    #print(q_values[max])

In [16]:

def value_iteration(gamma, theta):
    valueMap = np.zeros((gridSize, gridSize))
    while True:
        delta = 0
        for state in states:
            v_old=valueMap[state[0],state[1]]
            bellman_optimality_update(valueMap, state, gamma)
            delta = max(delta, abs(v_old - valueMap[state[0],state[1]]))
        if delta < theta:
            break
    pi = np.ones((gridSize,gridSize))/4
    for state in states:
        greedify_policy(state,pi,pi1,gamma,valueMap)
    print(pi)
    print(pi1)
    return valueMap, pi,pi1

In [17]:

gamma = 1
theta = 0.01
valueMap,pi,pi1=value_iteration(gamma, theta)
pi
pi1

[[0. 3. 3. 1.]
 [0. 0. 0. 1.]
 [0. 0. 1. 1.]
 [0. 2. 2. 0.]]
[[b'u' b'l' b'l' b'd']
 [b'u' b'u' b'u' b'd']
 [b'u' b'u' b'd' b'd']
 [b'u' b'r' b'r' b'u']]

Out[17]:

chararray([[b'u', b'l', b'l', b'd'],
           [b'u', b'u', b'u', b'd'],
           [b'u', b'u', b'd', b'd'],
           [b'u', b'r', b'r', b'u']], dtype='|S1')

Findings

The above valueMap shows the optimal path from any state

2.Gridworld 2

To make the problem more interesting, I created a 2nd grid which has more interesting structure as shown below <img src=”fig5.png”

The end state is the grey cell. Transitions to the black cells have a negative reward of -10. All other transitions have a reward of -1, while the end state has a reward of 0

In [2]:

##2a. Bellman Update

In [3]:

gamma = 1 # discounting rate
gridSize = 4

terminationStates = [[0,0]]
#terminationStates = [[0,0]]
actions = [[-1, 0], [1, 0], [0, 1], [0, -1]]
numIterations = 1000

In [4]:

rewardValue = np.zeros((gridSize,gridSize)) -1
rewardValue[0]=np.array([-1,-10,-10,-10])
rewardValue[2]=np.array([-10,-10,-10,-1])
rewardValue

Out[4]:

array([[ -1., -10., -10., -10.],
       [ -1.,  -1.,  -1.,  -1.],
       [-10., -10., -10.,  -1.],
       [ -1.,  -1.,  -1.,  -1.]])

In [5]:

def actionValue(initialPosition,action):
    if initialPosition in terminationStates:
        finalPosition = initialPosition
        reward=0
    else:
        #Compute final position
        finalPosition = np.array(initialPosition) + np.array(action)
        
        # If the action moves the finalPosition out of the grid, stay in same cell
        if -1 in finalPosition or gridSize in finalPosition:
                finalPosition = initialPosition
                reward= rewardValue[finalPosition[0],finalPosition[1]]
        else:
                reward= rewardValue[finalPosition[0],finalPosition[1]]
    
    #print(finalPosition)
    return finalPosition, reward

In [6]:

valueMap = np.zeros((gridSize, gridSize))
valueMap1 = np.zeros((gridSize, gridSize))
states = [[i, j] for i in range(gridSize) for j in range(gridSize)]

In [7]:

def policy_evaluation(numIterations,gamma,theta,valueMap):
    for i in range(numIterations):
        delta=0
        #print("iterations=",i)
        for state in states:
            weightedRewards=0
            for action in actions:
                finalPosition,reward = actionValue(state,action)
                #print("reward=",reward,"valueMap=",valueMap[finalPosition[0],finalPosition][1])
                weightedRewards += 1/4* (reward + gamma * valueMap[finalPosition[0],finalPosition][1])
            #print(weightedRewards)
            valueMap1[state[0],state[1]]=weightedRewards
            #print("wr=",weightedRewards,"va=",valueMap[state[0],state[1]]) 
            delta =max(delta,abs(weightedRewards-valueMap[state[0],state[1]]))
        valueMap = np.copy(valueMap1)
        #print(valueMap1)
        if(delta < 0.01):
            print(delta)                                                   
            print(valueMap)
            break

In [8]:

valueMap = np.zeros((gridSize, gridSize))
valueMap1 = np.zeros((gridSize, gridSize))
states = [[i, j] for i in range(gridSize) for j in range(gridSize)]
policy_evaluation(1000,1,0.0001,valueMap)

0.009862596190146178
[[   0.         -137.28514189 -209.19560831 -239.01378395]
 [-129.2494276  -180.67825796 -220.31626448 -237.86482779]
 [-194.08846546 -213.88769305 -231.5579035  -241.29920147]
 [-217.15664109 -227.25768494 -237.76348718 -241.51200989]]

2b. Greedify

In [9]:

valueMap = np.zeros((gridSize, gridSize))
valueMap1 = np.zeros((gridSize, gridSize))
states = [[i, j] for i in range(gridSize) for j in range(gridSize)]
pi = np.ones((gridSize,gridSize))/4
pi1 = np.chararray((gridSize, gridSize))
pi1[:] = 'a'

In [10]:

# Compute the value state function for the Grid
def policy_evaluate(states,actions,gamma,valueMap):
    #print("iterations=",i)
    for state in states:
        weightedRewards=0
        for action in actions:
            finalPosition,reward = actionValue(state,action)
            weightedRewards += 1/4* (reward + gamma * valueMap[finalPosition[0],finalPosition][1])
        # Set the computed weighted rewards to valueMap1
        valueMap1[state[0],state[1]]=weightedRewards
    # Copy to original valueMap
    valueMap = np.copy(valueMap1)
    return(valueMap)

In [11]:

def argmax(q_values):
    idx=np.argmax(q_values)
    return(np.random.choice(np.where(a==a[idx])[0].tolist()))


# Compute the best action in each state
def greedify_policy(state,pi,pi1,gamma,valueMap):  
        q_values=np.zeros(len(actions))
        for idx,action in enumerate(actions):
            finalPosition,reward = actionValue(state,action)
            q_values[idx] += 1/4* (reward + gamma * valueMap[finalPosition[0],finalPosition][1])
        # Find the index of the action for which the q_value is 
        idx=q_values.argmax()
        pi[state[0],state[1]]=idx 
        if(idx == 0):
            pi1[state[0],state[1]]='u'
        elif(idx == 1):
            pi1[state[0],state[1]]='d'
        elif(idx == 2):
            pi1[state[0],state[1]]='r'
        elif(idx == 3):
            pi1[state[0],state[1]]='l'

In [12]:

def improve_policy(pi, pi1,gamma,valueMap):
    policy_stable = True
    for state in states:
        old = pi[state].copy()
        # Greedify policy for state
        greedify_policy(state,pi,pi1,gamma,valueMap)
        if not np.array_equal(pi[state], old):
            policy_stable = False
    print(pi)
    print(pi1)
    return pi, pi1, policy_stable

In [13]:

def policy_iteration(gamma, theta):
    valueMap = np.zeros((gridSize, gridSize))
    pi = np.ones((gridSize,gridSize))/4
    pi1 = np.chararray((gridSize, gridSize))
    pi1[:] = 'a'
    policy_stable = False
    print("here")
    while not policy_stable:
        valueMap = policy_evaluate(states,actions,gamma,valueMap)
        pi,pi1, policy_stable = improve_policy(pi,pi1,  gamma,valueMap)
    return valueMap, pi,pi1

In [14]:

theta=0.1
valueMap, pi,pi1 = policy_iteration(gamma, theta)

here
[[0. 3. 1. 1.]
 [0. 3. 2. 1.]
 [0. 1. 1. 1.]
 [1. 1. 2. 1.]]
[[b'u' b'l' b'd' b'd']
 [b'u' b'l' b'r' b'd']
 [b'u' b'd' b'd' b'd']
 [b'd' b'd' b'r' b'd']]
[[0. 3. 1. 1.]
 [0. 3. 2. 1.]
 [0. 1. 1. 1.]
 [1. 2. 2. 1.]]
[[b'u' b'l' b'd' b'd']
 [b'u' b'l' b'r' b'd']
 [b'u' b'd' b'd' b'd']
 [b'd' b'r' b'r' b'd']]
[[0. 3. 1. 1.]
 [0. 3. 2. 1.]
 [0. 1. 1. 1.]
 [2. 2. 2. 1.]]
[[b'u' b'l' b'd' b'd']
 [b'u' b'l' b'r' b'd']
 [b'u' b'd' b'd' b'd']
 [b'r' b'r' b'r' b'd']]
[[0. 3. 1. 1.]
 [0. 3. 2. 1.]
 [0. 1. 1. 1.]
 [2. 2. 2. 1.]]
[[b'u' b'l' b'd' b'd']
 [b'u' b'l' b'r' b'd']
 [b'u' b'd' b'd' b'd']
 [b'r' b'r' b'r' b'd']]

In [15]:

## 2c. Bellman Optimality update

In [16]:

gamma = 1 # discounting rate
rewardValue = np.zeros((gridSize,gridSize)) -1
rewardValue[0]=np.array([-1,-10,-10,-10])
rewardValue[2]=np.array([-10,-10,-10,-1])
rewardValue
gridSize = 4
terminationStates = [[0,0]]
actions = [[-1, 0], [1, 0], [0, 1], [0, -1]]
numIterations = 1000

In [17]:

valueMap = np.zeros((gridSize, gridSize))
valueMap1 = np.zeros((gridSize, gridSize))
states = [[i, j] for i in range(gridSize) for j in range(gridSize)]
pi = np.ones((gridSize,gridSize))/4
pi1 = np.chararray((gridSize, gridSize))
pi1[:] = 'a'

In [18]:

2c. Bellman Optimality Update

def bellman_optimality_update(valueMap, state, gamma):

    q_values=np.zeros(len(actions))
    
    for idx,action in enumerate(actions):
        finalPosition,reward = actionValue(state,action)
        q_values[idx] += 1/4* (reward + gamma * valueMap[finalPosition[0],finalPosition][1])
    # Find the index of the action for which the q_value is 
    idx=q_values.argmax()
            
    max = np.argmax(q_values)
    valueMap[state[0],state[1]] = q_values[max]    
    #print(q_values[max])

In [19]:

def value_iteration(gamma, theta):
    valueMap = np.zeros((gridSize, gridSize))
    while True:
        delta = 0
        for state in states:
            v_old=valueMap[state[0],state[1]]
            bellman_optimality_update(valueMap, state, gamma)
            delta = max(delta, abs(v_old - valueMap[state[0],state[1]]))
        if delta < theta:
            break
    pi = np.ones((gridSize,gridSize))/4
    for state in states:
        greedify_policy(state,pi,pi1,gamma,valueMap)
    print(pi)
    print(pi1)
    return valueMap, pi,pi1

In [20]:

gamma = 1
theta = 0.000001
valueMap,pi,pi1=value_iteration(gamma, theta)
pi
pi1

[[0. 3. 1. 1.]
 [0. 3. 3. 3.]
 [0. 0. 0. 0.]
 [2. 2. 2. 0.]]
[[b'u' b'l' b'd' b'd']
 [b'u' b'l' b'l' b'l']
 [b'u' b'u' b'u' b'u']
 [b'r' b'r' b'r' b'u']]

Out[20]:

chararray([[b'u', b'l', b'd', b'd'],
           [b'u', b'l', b'l', b'l'],
           [b'u', b'u', b'u', b'u'],
           [b'r', b'r', b'r', b'u']], dtype='|S1')

Findings

The above shows the path from any cell to the stop cell as

3. Another maze

This is the third grid world which I create where the green cell is the end state and has a reward of 0. Transitions to the black cell will receive a reward of -10 and all other transitions will receive a reward of -1

In [2]:

gamma = 1 # discounting rate
gridSize = 5
terminationStates = [[2,2]]
actions = [[-1, 0], [1, 0], [0, 1], [0, -1]]
numIterations = 1000

In [3]:

rewardValue = np.zeros((gridSize,gridSize)) -1
rewardValue[1]=np.array([-1,-10,-1,-10,-1])
rewardValue[3]=np.array([-1,-10,-1,-10,-1])
rewardValue

Out[3]:

array([[ -1.,  -1.,  -1.,  -1.,  -1.],
       [ -1., -10.,  -1., -10.,  -1.],
       [ -1.,  -1.,  -1.,  -1.,  -1.],
       [ -1., -10.,  -1., -10.,  -1.],
       [ -1.,  -1.,  -1.,  -1.,  -1.]])

In [4]:

3a. Bellman Update

def actionValue(initialPosition,action):
    if initialPosition in terminationStates:
        finalPosition = initialPosition
        reward=0
    else:
        #Compute final position
        finalPosition = np.array(initialPosition) + np.array(action)
        
        # If the action moves the finalPosition out of the grid, stay in same cell
        if -1 in finalPosition or gridSize in finalPosition:
                finalPosition = initialPosition
                reward= rewardValue[finalPosition[0],finalPosition[1]]
        else:
                reward= rewardValue[finalPosition[0],finalPosition[1]]
    
    #print(finalPosition)
    return finalPosition, reward

In [5]:

valueMap = np.zeros((gridSize, gridSize))
valueMap1 = np.zeros((gridSize, gridSize))
states = [[i, j] for i in range(gridSize) for j in range(gridSize)]

In [6]:

def policy_evaluation(numIterations,gamma,theta,valueMap):
    for i in range(numIterations):
        delta=0
        #print("iterations=",i)
        for state in states:
            weightedRewards=0
            for action in actions:
                finalPosition,reward = actionValue(state,action)
                #print("reward=",reward,"valueMap=",valueMap[finalPosition[0],finalPosition][1])
                weightedRewards += 1/4* (reward + gamma * valueMap[finalPosition[0],finalPosition][1])
            #print(weightedRewards)
            valueMap1[state[0],state[1]]=weightedRewards
            #print("wr=",weightedRewards,"va=",valueMap[state[0],state[1]]) 
            delta =max(delta,abs(weightedRewards-valueMap[state[0],state[1]]))
        valueMap = np.copy(valueMap1)
        #print(valueMap1)
        if(delta < 0.01):
            print(delta)                                                   
            print(valueMap)
            break

In [7]:

valueMap = np.zeros((gridSize, gridSize))
valueMap1 = np.zeros((gridSize, gridSize))
states = [[i, j] for i in range(gridSize) for j in range(gridSize)]
policy_evaluation(1000,1,0.0001,valueMap)

0.009697101372182715
[[-82.49768079 -80.51647225 -74.9345659  -80.51647225 -82.49768079]
 [-80.51647225 -71.15241689 -59.80375072 -71.15241689 -80.51647225]
 [-74.9345659  -59.80375072   0.         -59.80375072 -74.9345659 ]
 [-80.51647225 -71.15241689 -59.80375072 -71.15241689 -80.51647225]
 [-82.49768079 -80.51647225 -74.9345659  -80.51647225 -82.49768079]]

3b. Greedify

In [8]:

valueMap = np.zeros((gridSize, gridSize))
valueMap1 = np.zeros((gridSize, gridSize))
states = [[i, j] for i in range(gridSize) for j in range(gridSize)]
pi = np.ones((gridSize,gridSize))/4
pi1 = np.chararray((gridSize, gridSize))
pi1[:] = 'a'

In [9]:

# Compute the value state function for the Grid
def policy_evaluate(states,actions,gamma,valueMap):
    #print("iterations=",i)
    for state in states:
        weightedRewards=0
        for action in actions:
            finalPosition,reward = actionValue(state,action)
            weightedRewards += 1/4* (reward + gamma * valueMap[finalPosition[0],finalPosition][1])
        # Set the computed weighted rewards to valueMap1
        valueMap1[state[0],state[1]]=weightedRewards
    # Copy to original valueMap
    valueMap = np.copy(valueMap1)
    return(valueMap)

In [10]:

def argmax(q_values):
    idx=np.argmax(q_values)
    return(np.random.choice(np.where(a==a[idx])[0].tolist()))


# Compute the best action in each state
def greedify_policy(state,pi,pi1,gamma,valueMap):  
        q_values=np.zeros(len(actions))
        for idx,action in enumerate(actions):
            finalPosition,reward = actionValue(state,action)
            q_values[idx] += 1/4* (reward + gamma * valueMap[finalPosition[0],finalPosition][1])
        # Find the index of the action for which the q_value is 
        idx=q_values.argmax()
        pi[state[0],state[1]]=idx 
        if(idx == 0):
            pi1[state[0],state[1]]='u'
        elif(idx == 1):
            pi1[state[0],state[1]]='d'
        elif(idx == 2):
            pi1[state[0],state[1]]='r'
        elif(idx == 3):
            pi1[state[0],state[1]]='l'

In [11]:

def improve_policy(pi, pi1,gamma,valueMap):
    policy_stable = True
    for state in states:
        old = pi[state].copy()
        # Greedify policy for state
        greedify_policy(state,pi,pi1,gamma,valueMap)
        if not np.array_equal(pi[state], old):
            policy_stable = False
    print(pi)
    print(pi1)
    return pi, pi1, policy_stable

In [12]:

def policy_iteration(gamma, theta):
    valueMap = np.zeros((gridSize, gridSize))
    pi = np.ones((gridSize,gridSize))/4
    pi1 = np.chararray((gridSize, gridSize))
    pi1[:] = 'a'
    policy_stable = False
    print("here")
    while not policy_stable:
        valueMap = policy_evaluate(states,actions,gamma,valueMap)
        pi,pi1, policy_stable = improve_policy(pi,pi1,  gamma,valueMap)
    return valueMap, pi,pi1

In [13]:

theta=0.1
valueMap, pi,pi1 = policy_iteration(gamma, theta)

here
[[0. 2. 0. 2. 0.]
 [0. 0. 1. 0. 0.]
 [3. 2. 0. 3. 2.]
 [0. 1. 0. 1. 0.]
 [1. 2. 1. 2. 1.]]
[[b'u' b'r' b'u' b'r' b'u']
 [b'u' b'u' b'd' b'u' b'u']
 [b'l' b'r' b'u' b'l' b'r']
 [b'u' b'd' b'u' b'd' b'u']
 [b'd' b'r' b'd' b'r' b'd']]
[[0. 3. 0. 2. 0.]
 [0. 0. 1. 0. 0.]
 [3. 2. 0. 3. 2.]
 [1. 1. 0. 1. 1.]
 [1. 3. 1. 2. 1.]]
[[b'u' b'l' b'u' b'r' b'u']
 [b'u' b'u' b'd' b'u' b'u']
 [b'l' b'r' b'u' b'l' b'r']
 [b'd' b'd' b'u' b'd' b'd']
 [b'd' b'l' b'd' b'r' b'd']]
[[0. 3. 0. 2. 0.]
 [0. 0. 1. 0. 0.]
 [3. 2. 0. 3. 2.]
 [1. 1. 0. 1. 1.]
 [1. 3. 1. 2. 1.]]
[[b'u' b'l' b'u' b'r' b'u']
 [b'u' b'u' b'd' b'u' b'u']
 [b'l' b'r' b'u' b'l' b'r']
 [b'd' b'd' b'u' b'd' b'd']
 [b'd' b'l' b'd' b'r' b'd']]

In [14]:

gamma = 1 # discounting rate
gridSize=5
rewardValue = np.zeros((gridSize,gridSize)) -1
rewardValue = np.zeros((gridSize,gridSize)) -1
rewardValue[1]=np.array([-1,-10,-1,-10,-1])
rewardValue[3]=np.array([-1,-10,-1,-10,-1])
print(rewardValue)


terminationStates = [[2,2]]
actions = [[-1, 0], [1, 0], [0, 1], [0, -1]]
numIterations = 1000

[[ -1.  -1.  -1.  -1.  -1.]
 [ -1. -10.  -1. -10.  -1.]
 [ -1.  -1.  -1.  -1.  -1.]
 [ -1. -10.  -1. -10.  -1.]
 [ -1.  -1.  -1.  -1.  -1.]]

In [15]:

3c. Bellman Optimality Update

valueMap = np.zeros((gridSize, gridSize))
valueMap1 = np.zeros((gridSize, gridSize))
states = [[i, j] for i in range(gridSize) for j in range(gridSize)]
pi = np.ones((gridSize,gridSize))/4
pi1 = np.chararray((gridSize, gridSize))
pi1[:] = 'a'

In [16]:

def bellman_optimality_update(valueMap, state, gamma):

    q_values=np.zeros(len(actions))
    
    for idx,action in enumerate(actions):
        finalPosition,reward = actionValue(state,action)
        q_values[idx] += 1/4* (reward + gamma * valueMap[finalPosition[0],finalPosition][1])
    # Find the index of the action for which the q_value is 
    idx=q_values.argmax()
            
    max = np.argmax(q_values)
    valueMap[state[0],state[1]] = q_values[max]    
    #print(q_values[max])

In [17]:

def value_iteration(gamma, theta):
    valueMap = np.zeros((gridSize, gridSize))
    while True:
        delta = 0
        for state in states:
            v_old=valueMap[state[0],state[1]]
            bellman_optimality_update(valueMap, state, gamma)
            delta = max(delta, abs(v_old - valueMap[state[0],state[1]]))
        if delta < theta:
            break
    pi = np.ones((gridSize,gridSize))/4
    for state in states:
        greedify_policy(state,pi,pi1,gamma,valueMap)
    print(pi)
    print(pi1)
    return valueMap, pi,pi1

In [18]:

gamma = 1
theta = 0.000001
valueMap,pi,pi1=value_iteration(gamma, theta)
pi
pi1

[[1. 2. 1. 3. 1.]
 [1. 1. 1. 1. 1.]
 [2. 2. 0. 3. 3.]
 [0. 0. 0. 0. 0.]
 [0. 2. 0. 3. 0.]]
[[b'd' b'r' b'd' b'l' b'd']
 [b'd' b'd' b'd' b'd' b'd']
 [b'r' b'r' b'u' b'l' b'l']
 [b'u' b'u' b'u' b'u' b'u']
 [b'u' b'r' b'u' b'l' b'u']]

Out[18]:

chararray([[b'd', b'r', b'd', b'l', b'd'],
           [b'd', b'd', b'd', b'd', b'd'],
           [b'r', b'r', b'u', b'l', b'l'],
           [b'u', b'u', b'u', b'u', b'u'],
           [b'u', b'r', b'u', b'l', b'u']], dtype='|S1')

Findings

We can see that the Bellman Optimality Update correctly finds the path the to end node which we can see from the valueMap1 above which is

Conclusion:

We can see how with the Bellman equations implemented iteratively with dynamic programming we can solve mazes of arbitrary shapes and complexities as long as we correctly choose the reward for the transitions

References
1. Reinforcement learning – An introduction by Richard S. Sutton and Andrew G Barto
2. Reinforcement learning (RL) 101 with Python Blog by Gerard Martinez
3. Reinforcement Learning Demystified: Solving MDPs with Dynamic Programming Blog by Mohammed Ashraf

You may also like

1. My book ‘Deep Learning from first principles:Second Edition’ now on Amazon
2. Big Data-4: Webserver log analysis with RDDs, Pyspark, SparkR and SparklyR
3. Practical Machine Learning with R and Python – Part 3
3. Pitching yorkpy…on the middle and outside off-stump to IPL – Part 2
4. Sixer – R package cricketr’s new Shiny avatar
5. Natural language processing: What would Shakespeare say?
6. Getting started with Tensorflow, Keras in Python and R

To see all posts click Index of posts

Cricpy performs granular analysis of players

“Gold medals aren’t really made of gold. They’re made of sweat, determination, & a hard-to-find alloy called guts.” Dan Gable

“It doesn’t matter whether you are pursuing success in business, sports, the arts, or life in general: The bridge between wishing and accomplishing is discipline” Harvey Mackay

“I won’t predict anything historic. But nothing is impossible.” Michael Phelps

Introduction

In this post, I introduce 2 new functions in my Python package ‘cricpy’ (cricpy v0.20) see Introducing cricpy:A python package to analyze performances of cricketers which enable granular analysis of batsmen and bowlers. They are

Step 1: getPlayerDataHA – This function is a wrapper around getPlayerData(), getPlayerDataOD() and getPlayerDataTT(), and adds an extra column ‘homeOrAway’ which says whether the match was played at home/away/neutral venues. A CSV file is created with this new column.
Step 2: getPlayerDataOppnHA – This function allows you to slice & dice the data for batsmen and bowlers against specific oppositions, at home/away/neutral venues and between certain periods. This reducedsubset of data can be used to perform analyses. A CSV file is created as an output based on the parameters of opposition, home or away and the interval of time

Note All the existing cricpy functions can be used on this smaller fine-grained data set for a closer analysis of players

This post has been published in Rpubs and can be accessed at Cricpy performs granular analysis of players

You can download a PDF version of this post at Cricpy performs granular analysis of players

I have also updated the cricpy template with these lastest changes. See cricpy-template

1. Analyzing Rahul Dravid at 3 different stages of his career

The following functions analyze Rahul Dravid during 3 different periods of his illustrious career. a) 1st Jan 2001-1st Jan 2002 b) 1st Jan 2004-1st Jan 2005 c) 1st Jan 2009-1st Jan 2010

import cricpy.analytics as ca
# Get the homeOrAway dataset for Dravid in matches
# Note:Since I have already got the data I reuse the CSV file
#df=ca.getPlayerDataHA(28114,tfile="dravidTestHA.csv",matchType="Test")

# Get Dravid's data for 2001-02

df1=ca.getPlayerDataOppnHA(infile="dravidTestHA.csv",outfile="dravidTest2001.csv",startDate="2001-01-01",endDate="2002-01-01")

# Get Dravid's data for 2004-05
df2=ca.getPlayerDataOppnHA(infile="dravidTestHA.csv",outfile="dravidTest2004.csv", startDate="2004-01-01",endDate="2005-01-01")

# Get Dravid's data for 2009-10
df3=ca.getPlayerDataOppnHA(infile="dravidTestHA.csv",outfile="dravidTest2009.csv",startDate="2009-01-01",endDate="2010-01-01")

1a. Plot the performance of Dravid at venues during 2001,2004,2009

Note: Any of the cricpy functions can be used on the fine-grained subset of data as below.

import cricpy.analytics as ca
ca.batsmanAvgRunsGround("dravidTest2001.csv","Dravid-2001")

ca.batsmanAvgRunsGround("dravidTest2004.csv","Dravid-2004")

ca.batsmanAvgRunsGround("dravidTest2009.csv","Dravid-2009")

1b. Plot the performance of Dravid against different oppositions during 2001,2004,2009

import cricpy.analytics as ca
ca.batsmanAvgRunsOpposition("dravidTest2001.csv","Dravid-2001")

ca.batsmanAvgRunsOpposition("dravidTest2004.csv","Dravid-2004")

ca.batsmanAvgRunsOpposition("dravidTest2009.csv","Dravid-2009")

1c. Plot the relative cumulative average and relative strike rate of Dravid in 2001,2004,2009

The plot below compares Dravid’s cumulative strike rate and cumulative average during 3 different stages of his career

import cricpy.analytics as ca
frames=["dravidTest2001.csv","dravidTest2004.csv","dravidTest2009.csv"]
names=["Dravid-2001","Dravid-2004","Dravid-2009"]
ca.relativeBatsmanCumulativeAvgRuns(frames,names)

ca.relativeBatsmanCumulativeStrikeRate(frames,names)

2. Analyzing Virat Kohli’s performance against England in England in 2014 and 2018

The analysis below looks at Kohli’s performance against England in ‘away’ venues (England) in 2014 and 2018

import cricpy.analytics as ca
# Get the homeOrAway data for Kohli in Test matches
#df=ca.getPlayerDataHA(253802,tfile="kohliTestHA.csv",type="batting",matchType="Test")

# Get the homeOrAway data for Kohli in Test matches
df=ca.getPlayerDataHA(253802,tfile="kohliTestHA.csv",type="batting",matchType="Test")

# Get the subset if data of Kohli's performance against England in England in 2014

df=ca.getPlayerDataOppnHA(infile="kohliTestHA.csv",outfile="kohliTestEng2014.csv",  opposition=["England"],homeOrAway=["away"],startDate="2014-01-01",endDate="2015-01-01")

# Get the subset if data of Kohli's performance against England in England in 2018
df1=ca.getPlayerDataOppnHA(infile="kohliTestHA.csv",outfile="kohliTestEng2018.csv",
   opposition=["England"],homeOrAway=["away"],startDate="2018-01-01",endDate="2019-01-01")

2a. Kohli’s performance at England grounds in 2014 & 2018

Kohli had a miserable outing to England in 2014 with a string of low scores. In 2018 Kohli pulls himself out of the morass

import cricpy.analytics as ca
ca.batsmanAvgRunsGround("kohliTestEng2014.csv","Kohli-Eng-2014")

ca.batsmanAvgRunsGround("kohliTestEng2018.csv","Kohli-Eng-2018")

2a. Kohli’s cumulative average runs in 2014 & 2018

Kohli’s cumulative average runs in 2014 is in the low 15s, while in 2018 it is 70+. Kohli stamps his class back again and undoes the bad memories of 2014

import cricpy.analytics as ca
ca.batsmanCumulativeAverageRuns("kohliTestEng2014.csv", "Kohli-Eng-2014")

ca.batsmanCumulativeAverageRuns("kohliTestEng2018.csv", "Kohli-Eng-2018")

3a. Compare the performances of Ganguly, Dravid and VVS Laxman against opposition in ‘away’ matches in Tests

The analyses below compares the performances of Sourav Ganguly, Rahul Dravid and VVS Laxman against Australia, South Africa, and England in ‘away’ venues between 01 Jan 2002 to 01 Jan 2008

import cricpy.analytics as ca
#Get the HA data for Ganguly, Dravid and Laxman
#df=ca.getPlayerDataHA(28779,tfile="gangulyTestHA.csv",type="batting",matchType="Test")
#df=ca.getPlayerDataHA(28114,tfile="dravidTestHA.csv",type="batting",matchType="Test")
#df=ca.getPlayerDataHA(30750,tfile="laxmanTestHA.csv",type="batting",matchType="Test")

# Slice the data 
df=ca.getPlayerDataOppnHA(infile="gangulyTestHA.csv",outfile="gangulyTestAES2002-08.csv" ,opposition=["Australia", "England", "South Africa"],                        homeOrAway=["away"],startDate="2002-01-01",endDate="2008-01-01")
df=ca.getPlayerDataOppnHA(infile="dravidTestHA.csv",outfile="dravidTestAES2002-08.csv" ,opposition=["Australia", "England", "South Africa"],                        homeOrAway=["away"],startDate="2002-01-01",endDate="2008-01-01")
df=ca.getPlayerDataOppnHA(infile="laxmanTestHA.csv",outfile="laxmanTestAES2002-08.csv",opposition=["Australia", "England", "South Africa"],                       homeOrAway=["away"],startDate="2002-01-01",endDate="2008-01-01")

3b Plot the relative cumulative average runs and relative cumative strike rate

Plot the relative cumulative average runs and relative cumative strike rate of Ganguly, Dravid and Laxman

-Dravid towers over Laxman and Ganguly with respect to cumulative average runs. – Ganguly has a superior strike rate followed by Laxman and then Dravid

import cricpy.analytics as ca
frames=["gangulyTestAES2002-08.csv","dravidTestAES2002-08.csv","laxmanTestAES2002-08.csv"]
names=["GangulyAusEngSA2002-08","DravidAusEngSA2002-08","LaxmanAusEngSA2002-08"]
ca.relativeBatsmanCumulativeAvgRuns(frames,names)

ca.relativeBatsmanCumulativeStrikeRate(frames,names)

4. Compare the ODI performances of Rohit Sharma, Joe Root and Kane Williamson against opposition

Compare the performances of Rohit Sharma, Joe Root and Kane williamson in away & neutral venues against Australia, West Indies and Soouth Africa

Joe Root piles us the runs in about 15 matches. Rohit has played far more ODIs than the other two and averages a steady 35+

import cricpy.analytics as ca
# Get the ODI HA data for Rohit, Root and Williamson
#df=ca.getPlayerDataHA(34102,tfile="rohitODIHA.csv",type="batting",matchType="ODI")
#df=ca.getPlayerDataHA(303669,tfile="joerootODIHA.csv",type="batting",matchType="ODI")
#df=ca.getPlayerDataHA(277906,tfile="williamsonODIHA.csv",type="batting",matchType="ODI")

# Subset the data for specific opposition in away and neutral venues

## C:\Users\Ganesh\ANACON~1\lib\site-packages\statsmodels\compat\pandas.py:56: FutureWarning: The pandas.core.datetools module is deprecated and will be removed in a future version. Please use the pandas.tseries module instead.
##   from pandas.core import datetools

df=ca.getPlayerDataOppnHA(infile="rohitODIHA.csv",outfile="rohitODIAusWISA.csv"
                       ,opposition=["Australia", "West Indies", "South Africa"],
                      homeOrAway=["away","neutral"])
df=ca.getPlayerDataOppnHA(infile="joerootODIHA.csv",outfile="joerootODIAusWISA.csv"
                       ,opposition=["Australia", "West Indies", "South Africa"],
                       homeOrAway=["away","neutral"])
df=ca.getPlayerDataOppnHA(infile="williamsonODIHA.csv",outfile="williamsonODIAusWiSA.csv",opposition=["Australia", "West Indies", "South Africa"],                    homeOrAway=["away","neutral"])

4a. Compare cumulative strike rates and cumulative average runs of Rohit, Root and Williamson

The relative cumulative strike rate of all 3 are comparable

import cricpy.analytics as ca
frames=["rohitODIAusWISA.csv","joerootODIAusWISA.csv","williamsonODIAusWiSA.csv"]
names=["Rohit-ODI-AusWISA","Joe Root-ODI-AusWISA","Williamson-ODI-AusWISA"]
ca.relativeBatsmanCumulativeAvgRuns(frames,names)

ca.relativeBatsmanCumulativeStrikeRate(frames,names)

5. Plot the performance of Dhoni in T20s against specific opposition at all venues

Plot the performances of Dhoni against Australia, West Indies, South Africa and England

import cricpy.analytics as ca
# Get the HA T20 data for Dhoni
#df=ca.getPlayerDataHA(28081,tfile="dhoniT20HA.csv",type="batting",matchType="T20")
#Subset the data
df=ca.getPlayerDataOppnHA(infile="dhoniT20HA.csv",outfile="dhoniT20AusWISAEng.csv",opposition=["Australia", "West Indies", "South Africa","England"],                homeOrAway=["all"])

5a. Plot Dhoni’s performances in T20

Note You can use any of cricpy’s functions against the fine grained data

import cricpy.analytics as ca
ca.batsmanAvgRunsOpposition("dhoniT20AusWISAEng.csv","Dhoni")

ca.batsmanAvgRunsGround("dhoniT20AusWISAEng.csv","Dhoni")

ca.batsmanCumulativeStrikeRate("dhoniT20AusWISAEng.csv","Dhoni")

ca.batsmanCumulativeAverageRuns("dhoniT20AusWISAEng.csv","Dhoni")

6. Compute and performances of Anil Kumble, Muralitharan and Warne in ‘away’ test matches

Compute the performances of Kumble, Warne and Maralitharan against New Zealand, West Indies, South Africa and England in pitches that are not ‘home’ pithes

import cricpy.analytics as ca
# Get the bowling data for Kumble, Warne and Muralitharan in Test matches
#df=ca.getPlayerDataHA(30176,tfile="kumbleTestHA.csv",type="bowling",matchType="Test")
#df=ca.getPlayerDataHA(8166,tfile="warneTestHA.csv",type="bowling",matchType="Test")
#df=ca.getPlayerDataHA(49636,tfile="muraliTestHA.csv",type="bowling",matchType="Test")

# Subset the data
df=ca.getPlayerDataOppnHA(infile="kumbleTestHA.csv",outfile="kumbleTest-NZWISAEng.csv",opposition=["New Zealand", "West Indies", "South Africa","England"],
                       homeOrAway=["away"])

df=ca.getPlayerDataOppnHA(infile="warneTestHA.csv",outfile="warneTest-NZWISAEng.csv"
                       ,opposition=["New Zealand", "West Indies", "South Africa","England"], homeOrAway=["away"])

df=ca.getPlayerDataOppnHA(infile="muraliTestHA.csv",outfile="muraliTest-NZWISAEng.csv"
                       ,opposition=["New Zealand", "West Indies", "South Africa","England"], homeOrAway=["away"])

6a. Plot the average wickets of Kumble, Warne and Murali

import cricpy.analytics as ca
ca.bowlerAvgWktsOpposition("kumbleTest-NZWISAEng.csv","Kumble-NZWISAEng-AN")

ca.bowlerAvgWktsOpposition("warneTest-NZWISAEng.csv","Warne-NZWISAEng-AN")

ca.bowlerAvgWktsOpposition("muraliTest-NZWISAEng.csv","Murali-NZWISAEng-AN")

6b. Plot the average wickets in different grounds of Kumble, Warne and Murali

import cricpy.analytics as ca
ca.bowlerAvgWktsGround("kumbleTest-NZWISAEng.csv","Kumble")

ca.bowlerAvgWktsGround("warneTest-NZWISAEng.csv","Warne")

ca.bowlerAvgWktsGround("muraliTest-NZWISAEng.csv","Murali")

6c. Plot the cumulative average wickets and cumulative economy rate of Kumble, Warne and Murali

Murali has the best economy rate followed by Kumble and then Warne
Again Murali has the best cumulative average wickets followed by Warne and then Kumble

import cricpy.analytics as ca
frames=["kumbleTest-NZWISAEng.csv","warneTest-NZWISAEng.csv","muraliTest-NZWISAEng.csv"]
names=["Kumble","Warne","Murali"]
ca.relativeBowlerCumulativeAvgEconRate(frames,names)

ca.relativeBowlerCumulativeAvgWickets(frames,names)

7. Compute and plot the performances of Bumrah in 2016, 2017 and 2018 in ODIs

import cricpy.analytics as ca
# Get the HA data for Bumrah in ODI in bowling
#df=ca.getPlayerDataHA(625383,tfile="bumrahODIHA.csv",type="bowling",matchType="ODI")

# Slice the data for periods 2016, 2017 and 2018
df=ca.getPlayerDataOppnHA(infile="bumrahODIHA.csv",outfile="bumrahODI2016.csv",
                       startDate="2016-01-01",endDate="2017-01-01")

df=ca.getPlayerDataOppnHA(infile="bumrahODIHA.csv",outfile="bumrahODI2017.csv",
                       startDate="2017-01-01",endDate="2018-01-01")

df=ca.getPlayerDataOppnHA(infile="bumrahODIHA.csv",outfile="bumrahODI2018.csv",
                       startDate="2018-01-01",endDate="2019-01-01")

7a. Compute the performances of Bumrah in 2016, 2017 and 2018

Very clearly Bumrah is getting better at his art. His economy rate in 2018 is the best!!!
Bumrah has had a very prolific year in 2017. However all the years he seems to be quite effective

import cricpy.analytics as ca
frames=["bumrahODI2016.csv","bumrahODI2017.csv","bumrahODI2018.csv"]
names=["Bumrah-2016","Bumrah-2017","Bumrah-2018"]
ca.relativeBowlerCumulativeAvgEconRate(frames,names)

ca.relativeBowlerCumulativeAvgWickets(frames,names)

8. Compute and plot the performances of Shakib, Bumrah and Jadeja in T20 matches for bowling

import cricpy.analytics as ca
# Get the HA bowling data for Shakib, Bumrah and Jadeja
#df=ca.getPlayerDataHA(56143,tfile="shakibT20HA.csv",type="bowling",matchType="T20")
#df=ca.getPlayerDataHA(625383,tfile="bumrahT20HA.csv",type="bowling",matchType="T20")
#df=ca.getPlayerDataHA(234675,tfile="jadejaT20HA.csv",type="bowling",matchType="T20")

# Slice the data for performances against Sri Lanka, Australia, South Africa and England
df=ca.getPlayerDataOppnHA(infile="shakibT20HA.csv",outfile="shakibT20-SLAusSAEng.csv" ,opposition=["Sri Lanka","Australia", "South Africa","England"],
                       homeOrAway=["all"])
df=ca.getPlayerDataOppnHA(infile="bumrahT20HA.csv",outfile="bumrahT20-SLAusSAEng.csv",opposition=["Sri Lanka","Australia", "South Africa","England"],
                       homeOrAway=["all"])

df=ca.getPlayerDataOppnHA(infile="jadejaT20HA.csv",outfile="jadejaT20-SLAusSAEng.csv"                      ,opposition=["Sri Lanka","Australia", "South Africa","England"],   homeOrAway=["all"])

8a. Compare the relative performances of Shakib, Bumrah and Jadeja

Jadeja and Bumrah have comparable economy rates. Shakib is more expensive
Shakib pips Bumrah in number of cumulative wickets, though Bumrah is close behind

import cricpy.analytics as ca
frames=["shakibT20-SLAusSAEng.csv","bumrahT20-SLAusSAEng.csv","jadejaT20-SLAusSAEng.csv"]
names=["Shakib-SLAusSAEng","Bumrah-SLAusSAEng","Jadeja-SLAusSAEng"]
ca.relativeBowlerCumulativeAvgEconRate(frames,names)

ca.relativeBowlerCumulativeAvgWickets(frames,names)

Conclusion

By getting the homeOrAway data for players using the profileNo, you can slice and dice the data based on your choice of opposition, whether you want matches that were played at home/away/neutral venues. Finally by specifying the period for which the data has to be subsetted you can create fine grained analysis.

Hope you have a great time with cricpy!!!

Also see
1. My book ‘Cricket analytics with cricketr and cricpy’ is now on Amazon
2. The 3rd paperback & kindle editions of my books on Cricket, now on Amazon
3. Exploring Quantum Gate operations with QCSimulator
4. Deep Learning from first principles in Python, R and Octave – Part 6
5. Natural selection of database technology through the years
6. Pitching yorkpy … short of good length to IPL – Part 1
7. Using Linear Programming (LP) for optimizing bowling change or batting lineup in T20 cricket
8. Practical Machine Learning with R and Python – Part 3

To see all posts click Index of posts

Introduction

ODI Functions

1. Analysis of women’s ODI matches

Save all matches between 2 teams

Save all matches played by an ODI team against all other ODI teams

1.Scorecard

2.Batting Partnerships

Analyze bowling in the women’s ODI England-New Zealand match on 15 Feb 2013

3.Wicket kind

4.Match worm graph

Analysis of team in all matches against another team (Class 2)

5. Team Batsmen partnerships

6. Team bowler wicketkind

Performance of women ODI teams against all other teams in all matches (Class 3)

7. Overall batting scorecard

Individual batsman and bowler performances (Class 4)

8. Batsmen performances

Plot Runs vs Strike Rate

Plot the moving average

Analyze ODI bowler performances

9. Bowler performances

2. Analysis of women’s International Twenty 20 matches

T20 Functions

Save all matches between teams

Save all T20 matches played by a team against all other teams

T20 Match Analysis (Class 1)

10. Batting scorecard

11. Bowling scorecard

Head to head between 2 women’s T20 teams (Class 2)

12. Team batting partnerships

13. Team batting partnerships (plot)

14. Team Wicketkind

Performance of teams against all other teams in all T20 matches (Class 3)

15. Overall team scorecard

15. Team batting partnerships

16. Team bowling wicketkind

Analyze women T20 batsmen & bowlers (Class 4)

17. T20 batsmen performances

Analyze women’s T20 bowlers.

18. T20 bowler performances

3a. Rank women ODI batsmen

3b. Rank women ODI bowlers

4a. Rank women T20 batsman

4b. Rank women T20 bowlers

Conclusion

Rate this:

Share:

Like this:

Introduction

1a. Rank ODI Batsmen

1b. Rank ODI Bowlers

2a. Rank T20 Batsmen

2b. Rank T20 Bowlers

3a. Rank IPL Batsmen

3a. Rank IPL Bowlers

Rate this:

Share:

Like this:

Display data

Create a Convolutional Neural Network

Summary of CNN

Perform Gradient descent and validate with the validation data

Generate the filter activations at the intermediate CNN layers

Display the activations at the intermediate layers

Visualize the pattern that the filters respond to maximally

Visualizing Class Outputs

Conclusion:

References

Also see

Rate this:

Share:

Like this:

A) Run 1

Run 3

Run 4

References

See also

Rate this:

Share:

Like this: