2023 05 18 automated briefings

2018-09-05 Random forests/Sitrep 16-17.csv

@@ -0,0 +1,110 @@
+trust_code,GAbeds,Norovirus,diverts,ACC,attendance,performancesy
+RXP,89.21,0,0,79.7,6.32,0.895299
+RTR,96.19,0,0,86.99,5.39,0.931424
+RLN,87.33,0.01,0,70.7,3.1,0.890376
+RTD,87.56,0.44,0,80.44,8.92,0.919367
+RR7,94.17,0,0,88.75,1.4,0.946481333
+RTF,94.15,0.08,0,95.03,4.92,0.893035333
+RE9,93.35,0.02,0,79.08,0.59,0.909940667
+RWD,96.72,0.01,0,77.39,4.09,0.761819667
+RX1,93.43,0.34,0,84.63,8.38,0.774517
+RCX,97.46,0.04,0,90.95,0.64,0.899322667
+RC9,94.91,0,0,77.02,2.31,0.984394
+RK5,93.28,0.03,0,88.69,2.49,0.936054667
+RTG,87.07,0.04,0,81.67,5.7,0.860491333
+RFS,94.46,0.01,0,87.07,1.15,0.828021
+RWJ,90.51,0.03,0,82.56,1.84,0.717097333
+RJN,96.98,0,0,70.92,0.34,0.765705333
+RHQ,95.49,0.37,0,80.28,8.42,0.819239333
+RJF,98.07,0.03,0,71.86,1.05,0.894674667
+R1F,97.53,0,0,80.23,0.39,0.842843667
+RYJ,93.71,0,0,97.36,6.74,0.862219667
+R1K,94.91,0.06,0,76.52,9.47,0.803925333
+RWG,91.69,0,0,88.95,2.57,0.771880333
+RQW,99.65,0.02,0,86.47,0.97,0.666211667
+RRV,77.39,0.02,0,82.78,3.87,0.871944
+RAP,99.64,0,0,87.96,2.11,0.792240667
+RM1,99.58,0.07,0,81.51,3.91,0.777558333
+RGP,96.26,0.06,0,77.3,0.81,0.882759333
+RHW,94.05,0.01,0,90.04,2.2,0.904404333
+RTH,93.73,0,0,72.96,4.48,0.860474333
+RDE,97.96,0.01,0,76.89,1.89,0.831785667
+R1H,94.77,0.05,0,87.72,22.53,0.819526667
+RDD,99.59,0,0,88.24,2.69,0.894008333
+RAJ,94.95,0,0,89.76,1.51,0.799368667
+RF4,94.49,0,0,85.76,8.44,0.832420333
+RQQ,97.44,0.02,0,90.46,0.26,0.747513
+RNQ,98.89,0,0,88.97,1.45,0.756472
+RKB,96.5,0.03,0,95.14,5.44,0.788562
+RWE,90.79,0.07,0,73.34,10.66,0.791167667
+RNS,99.46,0,0,82.78,2.09,0.809344
+RD8,96.28,0,0,79.33,0,0.875429333
+RLT,96.44,0,0,71.84,0.61,0.808379
+RRK,98.39,0,0,96.85,3.27,0.770758667
+RQ3,88.75,0,0,55.71,0.4,0.947426667
+RJC,94.39,0.05,0,75.91,0.83,0.939936
+RL4,90.71,0.06,0,64.5,4.85,0.892209667
+RNA,92.6,0.02,0,94.61,3.61,0.902644333
+RLQ,93,0.01,0,83,0.59,0.844752333
+RWP,99.05,0.15,0.01,92.12,3.11,0.768087667
+RBK,96.97,0.07,0,89.26,1.53,0.785969
+RXW,93.67,0,0,88.69,2.72,0.758751667
+RTE,93.16,0,0,88.15,3.25,0.752257333
+RTK,98.16,0.01,0,83.11,1.36,0.829655333
+RKE,97.4,0,0,84.22,0.6,0.850969333
+RN7,94.97,0,0,92.25,1.67,0.806435
+RJ2,97.87,0.04,0,91.7,6.42,0.822741667
+RJZ,98.84,0.29,0,96.2,10.49,0.783680667
+RJ7,93.46,0.08,0,83.68,4.06,0.887711667
+RCF,94.59,0.01,0,62.89,0.59,0.899442667
+RAE,94.83,0.06,0,82.9,2.55,0.863046667
+RR8,95.96,0.17,0,72.3,13.34,0.812935
+RFR,93.98,0.09,0,69.8,2.08,0.821598
+RP5,92.55,0.02,0,89.85,3.46,0.867936333
+RFF,96.93,0,0,84.05,0.83,0.852187667
+RWY,91.1,0.07,0,72.51,2.95,0.927092
+RCD,91.79,0.03,0,77.77,0.45,0.937470667
+RCB,91.63,0.19,0,87.65,4.49,0.802483667
+RWA,86.5,0.11,0,74.99,3.98,0.840269
+RN5,86.16,0.02,0,74.7,2.62,0.849109
+RHU,95.77,0,0,85.24,3.78,0.740002667
+RA2,94.86,0.02,0,96.13,0.73,0.874913
+RN3,93.47,0.02,0,79.13,1.36,0.813659667
+RBD,90.04,0,0,74.88,0.33,0.976974667
+RDZ,90.95,0.01,0,67.4,1.38,0.928473333
+RD3,95.08,0.02,0,80.75,1.07,0.914599
+RD1,94.68,0.07,0,80.77,1.44,0.797075
+RVJ,97.3,0,0,88.47,2,0.783172333
+RA7,85.46,0,0,84.78,2.41,0.802455
+RA3,97.81,0.02,0,89.55,0.33,0.666578333
+RH8,82.6,0.08,0,80.09,2.01,0.917079333
+RBA,88.68,0.02,0,76.2,0.94,0.919198667
+REF,91.29,0.01,0,73.86,1.5,0.804558667
+RK9,98.07,0.01,0,70.21,2.26,0.830004667
+RC1,93.91,0,0,68.36,0.77,0.909231333
+RJE,98.31,0.19,0,87.15,7.8,0.755551
+RXR,96.21,0,0,75.54,5.42,0.774975333
+RXN,94.33,0,0,80.43,0,0.814520667
+RXL,95.89,0.03,0,75.39,4.03,0.838749667
+RW3,92.43,0.13,0,91.18,10.02,0.896629667
+RMC,92.18,0.01,0,86.08,1.69,0.814044667
+RW6,94.41,0.05,0.01,89.49,8.49,0.775423
+RM2,90.86,0.05,0,91.15,2.41,0.849619667
+RMP,98.02,0,0,88.12,1.06,0.799364
+RBN,94.4,0,0,91.68,3.09,0.830070667
+RWW,88.5,0.01,0,86.85,1.61,0.851599
+RNL,91.68,0.16,0,74.25,2.7,0.841071667
+REP,73.34,0,0,41.58,0.03,0.969193667
+REM,96.08,0.1,0,83.94,3.16,0.815655
+RQ6,95.51,0,0,92.82,0,0.863516333
+RBT,92.73,0.08,0,72.72,1.12,0.890214
+RVY,87.75,0.01,0,87.77,1.41,0.8916
+RBL,98.3,0.11,0,89.05,2.56,0.799897333
+RJR,93.92,0.04,0,74.64,0.92,0.842648333
+RVR,98.97,0.01,0,88.4,2.61,0.946300667
+RJ6,98.78,0,0,92.58,1.49,0.847868667
+RXH,96.85,0.05,0,87.52,3.44,0.792876333
+RXC,95.28,0.06,0,81.17,1.94,0.756502333
+RTP,96.19,0.03,0,93.75,1.65,0.892421667
+RPA,96.68,0,0,96.65,1.89,0.739128333
+RWF,96.55,0,0,88.47,3.31,0.804745667

2018-09-05 Random forests/random forest-demo-09-18.R

@@ -0,0 +1,87 @@
+if (!require(rpart)) install.packages('rpart') # DECISION TREE PACKAGE
+library(rpart)
+if (!require(rpart.plot)) install.packages('rpart.plot') # PLOTTING PACKAGE
+library(rpart.plot)
+if (!require(rattle)) install.packages('rattle') # PLOTTING PACKAGE
+library(rattle)
+if (!require(RColorBrewer)) install.packages('RColorBrewer') # PLOTTING PACKAGE
+library(RColorBrewer)
+if (!require(randomForest)) install.packages('randomForest') # RANDOM FOREST PACKAGE
+library(randomForest)
+if (!require(caret)) install.packages('caret') # ML SUPPORT PACKAGE
+library(caret)
+if (!require(tidyverse)) install.packages('tidyverse') # DATA MANIPULATION AND PLOTTING PACKAGE
+library(tidyverse)
+
+#set random seed for reproducability
+set.seed(111)
+
+#read in input data
+setwd("~/spotlight seminar 06.09.18")
+
+Sitrep_12_16 <- read_csv('Sitrep 12-16.csv') # 4 WINTERS OF SITREP DATA
+
+Sitrep_16_17 <- read_csv('Sitrep 16-17.csv') # SITREP DATA FROM WITER 16-17
+
+
+#creat (stratified) test/train splits
+# ========================================================================== 
+
+#  create index for test/train split
+index <- createDataPartition(Sitrep_12_16$performanceny, # stratify according to performance in the next year
+                             p = 0.8, # split 80:20
+                             list = FALSE) 
+
+Sitrep_12_16_train <- Sitrep_12_16[index,] # create training set
+
+Sitrep_12_16_test <- Sitrep_12_16[-index,] # create test set
+
+# confirm stratification
+mean(Sitrep_12_16_train$performanceny)
+mean(Sitrep_12_16_test$performanceny)
+
+#USE WINTERS 12-16 TO TRAIN AND EVALUATE TREE AND FOREST
+# ==========================================================================
+
+tree <- rpart(performanceny~., # model formula
+                Sitrep_12_16_train, # training data
+                method = 'anova') # regesion tree 
+
+fancyRpartPlot(tree) # visulize tree
+
+forest <- randomForest(performanceny~., # model formula
+                         Sitrep_12_16_train, # training data
+                         ntree=1000, # 1000 tree forest
+                         importance=TRUE)  # calculate variable importance
+
+varImpPlot(forest) # visulize variable importance
+importance(forest) # list variable importance
+
+tree_p <- predict(tree,Sitrep_12_16_test) # make tree predictions
+tree_error <- tree_p - Sitrep_12_16_test$performanceny # calculate tree error
+rmse_tree <- sqrt(mean(tree_error^2)) # calculate tree RMSE
+
+forest_p <- predict(forest,Sitrep_12_16_test) # make forest predictions
+forest_error <- forest_p - Sitrep_12_16_test$performanceny # calculate forest error
+rmse_forest <- sqrt(mean(forest_error^2)) # calculate forest RMSE
+
+
+
+#TRAIN FINAL MODEL ON FULL DATASET (TRAIN + TEST) AND MAKE PREDICTIONS + PLOT SUMMARY
+# ============================================================================================
+
+forest_final <- randomForest(performanceny~.,
+                               Sitrep_12_16,
+                               ntree=1000,
+                               importance=TRUE) # grow forest on all data
+
+
+
+#MAKE FINAL PREDITIONS 
+# =========================================================================================
+
+predictions <- data.frame(trust_code = Sitrep_16_17$trust_code,
+                          performance_16_17 = Sitrep_16_17$performancesy,
+                          predictions_17_18 = predict(forest_final,Sitrep_16_17)) # make final predictions
+
+write_csv(predictions,'predictions.csv')

2018-09-05 Random forests/readme.md

@@ -0,0 +1,3 @@
+# spotlight-seminar-random-forest
+
+the input files and r code for a basic random forest

2019-09-20 Iteration/readme.md

@@ -1,5 +1 @@
 Material from the R coding club session led br Matt Malcher on 20th September 2019.
-
-Any other materials for the R coding club session on iteration can be found here:
-
-## [R Coding Club - Iteration](https://github.com/DataS-DH/RCC-Iteration)

2019-10-04 TidyR LearnR/RCC_tidyr_learnr.Rproj

@@ -0,0 +1,13 @@
+Version: 1.0
+
+RestoreWorkspace: Default
+SaveWorkspace: Default
+AlwaysSaveHistory: Default
+
+EnableCodeIndexing: Yes
+UseSpacesForTab: Yes
+NumSpacesForTab: 2
+Encoding: UTF-8
+
+RnwWeave: Sweave
+LaTeX: pdfLaTeX

2019-10-04 TidyR LearnR/readme.md

@@ -0,0 +1,3 @@
+A quick tutorial for the R Coding Club demonstrating how to use `learnr`
+
+App deployed at: https://matthew-malcher.shinyapps.io/tidyr_learnr/

2019-10-04 TidyR LearnR/tidyr_learnr/data/preg.csv

@@ -0,0 +1,4 @@
+"name","treatmenta","treatmentb"
+"John Smith",NA,18
+"Jane Doe",4,1
+"Mary Johnson",6,7

2019-10-04 TidyR LearnR/tidyr_learnr/data/preg2.csv

@@ -0,0 +1,3 @@
+"treatment","John Smith","Jane Doe","Mary Johnson"
+"a",NA,4,6
+"b",18,1,7