ML R packages (my focus: supervised learning) by CRAN downloads

ML_pkgs.R

##install.packages("cranlogs")

library(data.table)
library(cranlogs)


##caret/models/file
## grep "library =" * | sed 's/.*=//' | sed 's/c(//' | sed 's/),/,/' | grep -v NULL | sed 's/,.*$/,/' | sort | uniq | tr -d '\n'

caret_pkgs <- c("rpart", "C50", "CHAID", "Cubist", "FCNN4R", "HDclassif", "HiDimDA", "KRLS", "LiblineaR", 
                "LogicReg", "MASS", "RRF", "RSNNS", "RWeka", "Rborist", "VGAM", "ada", "adabag", "adaptDA", 
                "arm", "bartMachine", "binda", "bnclassify", "brnn", "bst", "caTools", "caret", "class", 
                "deepboost", "deepnet", "e1071", "earth", "elasticnet", "elmNN", "evtree", "extraTrees", 
                "fastAdaboost", "fastICA", "foba", "frbs", "gam", "gbm", "glmnet", "gpls", "h2o", "hda",
                "ipred", "keras", "kerndwd", "kernlab", "kknn", "klaR", "kohonen", "lars", "leaps", 
                "logicFS", "mboost", "mda", "mgcv", "monmlp", "monomvn", "msaenet", "mxnet", "naivebayes", 
                "neuralnet", "nnet", "nnls", "nodeHarvest", "obliqueRF", "ordinalNet", "pamr", "partDSA",
                "party", "penalized", "penalizedLDA", "pls", "plsRglm", "plyr", "proxy", "qrnn", 
                "quantregForest", "rFerns", "randomForest", "randomGLM", "relaxo", "robustDA", "rocc", 
                "rotationForest", "rpart", "rpartScore", "rqPen", "rrcov", "rrcovHD", "rrlda", "sda", 
                "sdwd", "snn", "sparseLDA", "sparsediscrim", "spikeslab", "spls", "stepPlr", "superpc", 
                "supervisedPRIM", "vbmp", "wsrf", "xgboost")
caret_pkgs <- setdiff(caret_pkgs,c("plyr","caTools"))


## cat cran_taskviews_ML.txt | sed 's/ (core)//' |sort| uniq | sed 's/^/"/' | sed 's/$/",/' | tr -d '\n'

taskview_pkgs <- c("BART","BDgraph","BayesTree","Boruta","C50","CORElearn","CoxBoost","Cubist","GAMBoost",
                   "GMMBoost","ICEbox","LTRCtrees","LiblineaR","OneR","REEMtree","RGF","RLT","ROCR","RPMM",
                   "RSNNS","RWeka","RXshrink","Rborist","RcppDL","Rmalschains","RoughSets","SIS",
                   "SuperLearner","ahaz","arules","bartMachine","biglasso","bmrm","bst","caret","deepnet",
                   "e1071","earth","effects","elasticnet","evclass","evtree","frbs","gamboostLSS","gbm",
                   "ggRandomForests","glmnet","glmpath","gradDescent","grf","grplasso","grpreg","h2o","hda",
                   "hdi","hdm","ipred","kernlab","klaR","lars","lasso2","maptree","mboost","mlr3",
                   "mlr3proba","naivebayes","ncvreg","nnet","opusminer","pamr","party","partykit","pdp",
                   "penalized","penalizedLDA","picasso","plotmo","quantregForest","randomForest",
                   "randomForestSRC","ranger","rattle","rdetools","relaxo","rgenoud","rminer","rpart",
                   "sda","ssgraph","stabs","svmpath","tensorflow","tgp","tree","trtf","varSelRF","wsrf",
                   "xgboost")


extra_pkgs <- c("lightgbm")


ML_pkgs <- unique(c(caret_pkgs,taskview_pkgs,extra_pkgs))

d <- as.data.table(cran_downloads(when="last-month", packages=ML_pkgs))
as.data.frame(d[,.(n=sum(count)),by=package][order(-n)])

        package      n

1 e1071 288132
2 MASS 204587
3 caret 155748
4 kernlab 145719
5 mgcv 131765
6 leaps 117709
7 ipred 87420
8 ranger 85792
9 ROCR 79862
10 randomForest 74759
11 xgboost 73768
12 glmnet 66530
13 arules 55555
14 nnet 49769
15 class 48601
16 pls 46767
17 arm 45597
18 rpart 45557
19 effects 44705
20 Boruta 42253
21 tensorflow 37488
22 keras 32485
23 proxy 32111
24 partykit 31814
25 klaR 31703
26 gbm 30026
27 party 28596
28 rrcov 26591
29 VGAM 26467
30 h2o 21845
31 grpreg 19392
32 fastICA 18890
33 gam 17197
34 earth 16990
35 plotmo 16049
36 lars 15092
37 rattle 14883
38 mlr3 14658
39 neuralnet 13689
40 Cubist 13678
41 ada 13466
42 RWeka 13178
43 mda 12001
44 BDgraph 11611
45 tree 11106
46 C50 10744
47 kknn 10249
48 nnls 8998
49 pamr 8339
50 pdp 8152
51 randomForestSRC 7201
52 rgenoud 7169
53 spls 6975
54 mboost 6530
55 elasticnet 5851
56 tgp 5667
57 LiblineaR 5555
58 adabag 5139
59 naivebayes 4763
60 RSNNS 4492
61 penalized 4397
62 rFerns 4235
63 stabs 4088
64 mlr3proba 4042
65 lasso2 3884
66 extraTrees 3757
67 bartMachine 3595
68 ncvreg 3434
69 CORElearn 3401
70 bst 3301
71 maptree 3233
72 deepnet 3070
73 kohonen 2967
74 CoxBoost 2950
75 spikeslab 2928
76 sda 2834
77 grf 2817
78 RRF 2748
79 evtree 2744
80 SuperLearner 2723
81 brnn 2709
82 frbs 2610
83 gamboostLSS 2482
84 rminer 2474
85 stepPlr 2327
86 foba 2296
87 LogicReg 2280
88 biglasso 2265
89 sparseLDA 2246
90 nodeHarvest 2134
91 rrlda 2104
92 rotationForest 2075
93 hdi 2071
94 SIS 2068
95 binda 1977
96 GAMBoost 1933
97 grplasso 1742
98 fastAdaboost 1717
99 OneR 1716
100 Rborist 1590
101 trtf 1558
102 qrnn 1509
103 BART 1501
104 RPMM 1421
105 BayesTree 1417
106 monomvn 1411
107 quantregForest 1406
108 RXshrink 1331
109 ggRandomForests 1275
110 svmpath 1258
111 ahaz 1225
112 hdm 1208
113 varSelRF 1158
114 kerndwd 1143
115 HDclassif 1121
116 gradDescent 1108
117 bnclassify 1063
118 bmrm 1041
119 glmpath 1016
120 rqPen 1013
121 ordinalNet 974
122 rdetools 965
123 GMMBoost 952
124 hda 945
125 LTRCtrees 924
126 ICEbox 921
127 rrcovHD 919
128 superpc 918
129 RoughSets 918
130 Rmalschains 911
131 relaxo 904
132 KRLS 885
133 picasso 883
134 msaenet 877
135 evclass 873
136 plsRglm 853
137 wsrf 847
138 deepboost 845
139 opusminer 845
140 REEMtree 844
141 RcppDL 821
142 penalizedLDA 812
143 monmlp 797
144 HiDimDA 792
145 ssgraph 786
146 RLT 768
147 snn 746
148 sdwd 718
149 RGF 704
150 randomGLM 692
151 obliqueRF 690
152 rocc 680
153 partDSA 678
154 rpartScore 672
155 lightgbm 655
156 robustDA 627
157 supervisedPRIM 590
158 FCNN4R 111
159 gpls 81
160 vbmp 49
161 sparsediscrim 29
162 adaptDA 15
163 elmNN 9
164 CHAID 0
165 logicFS 0
166 mxnet 0