pierreqi/r_code_50k · Datasets at Hugging Face

6f62c788844806ab2bf127f6500c1e993461ca9c

3a153e714918af3684d9341d250089576e727b21

/workflow/Code/resamp_analysis.R

f357a0fd5678922da89a57e56d94407992b65352

[]

no_license

SarahVal/geb_text_class_2020

434a6ded390805a70682616667111b08d5fafa1c

64206d7759df7af7b1bc37ff65e56e78f8391438

refs/heads/master

2023-03-23T23:51:33.990376

2020-10-09T12:39:43

null

0

null

UTF-8

R

false

12,068

r

resamp_analysis.R

#### # R script for analysing the AUC scores etc of the resampled models #### # Clear environment rm(list = ls()) graphics.off() # Load packages # cvAUC require(cvAUC) # H measure require(hmeasure) # ROCR require(ROCR) # ggplot require(ggplot2) # Functions # Function to convert string of numbers to vector str_to_vector <- function(input_str){ # Remove [ and ] and , str <- gsub("\\[|\\]|,", "", input_str) # Separate by ' ' str <- strsplit(str, ' ')[[1]] # Create vector to store output vect <- rep(99, length(str)) # Convert characters to floats and store in vect for (i in 1:length(str)){ vect[i] <- as.numeric(str[i]) } return (vect) } # Function to calculate metrics for cv predictions given an input df cv_metrics_calc_resamp <- function(input_df){ # Convert prob/class cols to character input_df$CV_Predicted_Probs <- as.character(input_df$CV_Predicted_Probs) input_df$CV_True_Class <- as.character(input_df$CV_True_Class) input_df['UniqueID'] <- paste(as.character(input_df$Attribute), as.character(input_df$Seed_number), as.character(input_df$Stop_words), sep = '_') # Create list to store model metrics model_metrics_list <- list() # Go through each model for (i in unique(input_df$UniqueID)){ sub_df <- subset(input_df, input_df$UniqueID == i) # Create list to store stats for this model model_metrics_list[[i]] <- list() # Create lists/vectors to store values from each fold for cvAUC y_pred_list <- list() y_true_list <- list() fold_id_list <- list() y_pred_vect <- c() y_true_vect <- c() fold_id_vect <- c() h_vect <- c() # Loop through each cv fold for (j in 1:dim(sub_df)[1]){ # Create list to store fold metrics model_metrics_list[[i]][[paste('Fold', j, sep = '_')]] <- list() # Extract useful values from row of interes tmp_pred <- str_to_vector(sub_df$CV_Predicted_Probs[j]) tmp_true <- str_to_vector(sub_df$CV_True_Class[j]) tmp_fold <- rep(j, length(tmp_true)) # Add values to lists/ vectors for later analysis y_pred_list[[j]] <- tmp_pred y_true_list[[j]] <- tmp_true fold_id_list[[j]] <- tmp_fold y_pred_vect <- c(y_pred_vect, tmp_pred) y_true_vect <- c(y_true_vect, tmp_true) fold_id_vect <- c(fold_id_vect, tmp_fold) model_metrics_list[[i]][[paste('Fold', j, sep = '_')]][['Predictions']] <- tmp_pred model_metrics_list[[i]][[paste('Fold', j, sep = '_')]][['True_Labels']] <- tmp_true if (j == 1){ # Extract predictions and true lables tmp_pred_test <- str_to_vector(sub_df$Test_Predicted_Probs[j]) tmp_true_test <- str_to_vector(sub_df$Test_True_Class[j]) pred_test <- prediction(tmp_pred_test, tmp_true_test) model_metrics_list[[i]][['Test_AUC']] <- AUC(tmp_pred, tmp_true) } } # Calculate cv AUC metrics and add to list model_metrics_list[[i]][['cvAUC']] <- cvAUC(y_pred_list, y_true_list, folds = fold_id_list) model_metrics_list[[i]][['cicvAUC']] <- ci.cvAUC(y_pred_vect, y_true_vect, folds = fold_id_vect) # Calculate av ROC cv_pred <- prediction(y_pred_list, y_true_list) model_metrics_list[[i]][['cv_ROC']] <- performance(cv_pred, 'tpr', 'fpr') } return (model_metrics_list) } # Function to create AUC/H df create_resamp_AUC_H_df <- function(input_list, input_df){ # List columns in input_df cols <- colnames(input_df) # Remove unnecessary columns # "N_training_docs", "Classifier", "Cost_function", "CV_folds", "Fold", "CV_Predicted_Probs", "CV_True_Class", "N_test_docs", "Test_Predicted_Probs" "Test_True_Class" indices_rm <- which(cols == 'N_training_docs' | cols == 'Cost_function' | cols == 'CV_folds' | cols == 'Fold' | cols == 'CV_Predicted_Probs' | cols == 'CV_True_Class' | cols == 'N_test_docs' | cols == 'Test_Predicted_Probs' | cols == 'Test_True_Class') cols <- cols[-indices_rm] # Add avAUC and avH cols cols <- c(cols, c("avAUC")) # Create df AUC_H_df <- data.frame(matrix(nrow = length(input_list), ncol = length(cols))) names(AUC_H_df) <- cols df_indices <- seq(1, dim(input_df)[1], 10) # Extract relevant info from input_list/input_df for (i in 1:length(input_list)){ for (j in 1:length(cols)){ if (cols[j] == 'avAUC'){ AUC_H_df[i,j] <- input_list[[i]]$cicvAUC$cvAUC next } else{ AUC_H_df[i,j] <- as.character(input_df[df_indices[i], which(colnames(input_df) == cols[j])]) } } } return (AUC_H_df) } min_mean_se_max_resamp <- function(x) { df <- data.frame('ymin' = min(x$avAUC), 'q2_5' = as.numeric(quantile(x$avAUC, 0.025)), 'lower' = mean(x$avAUC) - sd(x$avAUC)/sqrt(length(x$avAUC)), 'lo_CI' = mean(x$avAUC) - 1.96*(sd(x$avAUC)/sqrt(length(x$avAUC))), 'middle' = mean(x$avAUC), 'q50' = as.numeric(quantile(x$avAUC, 0.5)), 'up_CI' = mean(x$avAUC) + 1.96*(sd(x$avAUC)/sqrt(length(x$avAUC))), 'upper' = mean(x$avAUC) + sd(x$avAUC)/sqrt(length(x$avAUC)), 'q97_5' = as.numeric(quantile(x$avAUC, 0.975)), 'ymax' = max(x$avAUC)) return(df) } #### # Main code #### # Load resampled data lr_lpi_resamp_df <- read.csv('../Results/Model_metrics/LR/lpi_resample_metrics.csv') lr_predicts_resamp_df <- read.csv('../Results/Model_metrics/LR/predicts_resample_metrics.csv') nn_lpi_resamp_df <- read.csv('../Results/Model_metrics/NN/lpi_resample_metrics.csv') nn_predicts_resamp_df <- read.csv('../Results/Model_metrics/NN/predicts_resample_metrics.csv') # Calculate metrics lr_lpi_resamp_metr_list <- cv_metrics_calc_resamp(input_df = lr_lpi_resamp_df) lr_predicts_resamp_metr_list <- cv_metrics_calc_resamp(input_df = lr_predicts_resamp_df) nn_lpi_resamp_metr_list <- cv_metrics_calc_resamp(input_df = nn_lpi_resamp_df) nn_predicts_resamp_metr_list <- cv_metrics_calc_resamp(input_df = nn_predicts_resamp_df) # Export to df lr_lpi_resamp_auc_df <- create_resamp_AUC_H_df(input_list = lr_lpi_resamp_metr_list, input_df = lr_lpi_resamp_df) lr_predicts_resamp_auc_df <- create_resamp_AUC_H_df(input_list = lr_predicts_resamp_metr_list, input_df = lr_predicts_resamp_df) nn_lpi_resamp_auc_df <- create_resamp_AUC_H_df(input_list = nn_lpi_resamp_metr_list, input_df = nn_lpi_resamp_df) nn_predicts_resamp_auc_df <- create_resamp_AUC_H_df(input_list = nn_predicts_resamp_metr_list, input_df = nn_predicts_resamp_df) # Add dataset col lr_lpi_resamp_auc_df['Dataset'] <- 'LPD' lr_predicts_resamp_auc_df['Dataset'] <- 'PREDICTS' nn_lpi_resamp_auc_df['Dataset'] <- 'LPD' nn_predicts_resamp_auc_df['Dataset'] <- 'PREDICTS' lr_lpi_resamp_auc_df['Model'] <- 'LR A' lr_predicts_resamp_auc_df['Model'] <- 'LR A' nn_lpi_resamp_auc_df['Model'] <- 'CNN A' nn_predicts_resamp_auc_df['Model'] <- 'CNN A' # Rbind resamp_auc_df <- rbind(lr_lpi_resamp_auc_df[c('Dataset', 'Model', 'avAUC')], lr_predicts_resamp_auc_df[c('Dataset', 'Model', 'avAUC')], nn_lpi_resamp_auc_df[c('Dataset', 'Model', 'avAUC')], nn_predicts_resamp_auc_df[c('Dataset', 'Model', 'avAUC')]) resamp_bp_df <- ddply(resamp_auc_df, .(Dataset, Model), min_mean_se_max_resamp) # resamp_bp_df$ymin # 0.9696655 0.9815266 0.9734725 0.9880587 # Load original model scores lr_lpi_orig_mod_df <- read.csv('../Results/Model_metrics/LR/lpi_models_to_use.csv')[1,] lr_predicts_orig_mod_df <- read.csv('../Results/Model_metrics/LR/predicts_models_to_use.csv')[1,] nn_lpi_orig_mod_df <- read.csv('../Results/Model_metrics/NN/lpi_models_to_use.csv')[1,] nn_predicts_orig_mod_df <- read.csv('../Results/Model_metrics/NN/predicts_models_to_use.csv')[1,] lr_lpi_orig_mod_df['Dataset'] <- 'LPD' lr_predicts_orig_mod_df['Dataset'] <- 'PREDICTS' nn_lpi_orig_mod_df['Dataset'] <- 'LPD' nn_predicts_orig_mod_df['Dataset'] <- 'PREDICTS' lr_lpi_orig_mod_df['Model'] <- 'LR A' lr_predicts_orig_mod_df['Model'] <- 'LR A' nn_lpi_orig_mod_df['Model'] <- 'CNN A' nn_predicts_orig_mod_df['Model'] <- 'CNN A' orig_mod_df <- rbind(lr_lpi_orig_mod_df[c('Dataset', 'Model', 'avAUC')], lr_predicts_orig_mod_df[c('Dataset', 'Model', 'avAUC')], nn_lpi_orig_mod_df[c('Dataset', 'Model', 'avAUC')], nn_predicts_orig_mod_df[c('Dataset', 'Model', 'avAUC')]) # Merge original and resampled dfs resamp_bp_df <- merge(x = resamp_bp_df, y = orig_mod_df[c('Model', 'Dataset', 'avAUC')], by = c('Model', 'Dataset')) resamp_bp_df$Model <- factor(resamp_bp_df$Model, levels = c("LR A", "CNN A")) # Plot variation in avAUC resamp_plt <- ggplot(data = resamp_bp_df) + geom_errorbar(aes(x = Model, ymin = q2_5, ymax = q97_5, width = 0, colour = Dataset), size = 1, position = position_dodge(width = 0.3), show.legend = F) + geom_point(aes(x = Model, y = q50, colour = Dataset), pch = 16, size = 3.5, position = position_dodge(width = 0.3)) + geom_point(aes(x = Model, y = avAUC, group = Dataset), pch = 18, colour = 'black', size = 3.5, alpha = 0.7, position = position_dodge(width = 0.3)) + ylab('Average AUC') + xlab('Model') + geom_vline(xintercept = 1.5, lty = 'dashed', colour = 'grey50') + scale_y_continuous(breaks = c(0.95, 1.00), limits = c(0.95, 1.00)) + scale_color_manual(name = 'Indicator dataset', values = c("#E69F00", "#009E73"), breaks = c('LPD', 'PREDICTS')) + theme_bw() + theme(axis.text = element_text(size = 16), axis.title = element_text(size = 20), legend.text = element_text(size = 16), legend.title = element_text(size = 18), panel.grid.major = element_blank(), panel.grid.minor = element_blank()) resamp_plt # ggsave(plot = resamp_plt, filename = 'resamp_plt.pdf', # path = '../Results/Figs', # width = 8, height = 5, dpi = 300, device = "pdf") #### # Standard stats #### range(resamp_auc_df[which(resamp_auc_df$Model == 'LR A' & resamp_auc_df$Dataset == 'LPD'),]$avAUC) mean(resamp_auc_df[which(resamp_auc_df$Model == 'LR A' & resamp_auc_df$Dataset == 'LPD'),]$avAUC) sd(resamp_auc_df[which(resamp_auc_df$Model == 'LR A' & resamp_auc_df$Dataset == 'LPD'),]$avAUC) range(resamp_auc_df[which(resamp_auc_df$Model == 'LR A' & resamp_auc_df$Dataset == 'PREDICTS'),]$avAUC) mean(resamp_auc_df[which(resamp_auc_df$Model == 'LR A' & resamp_auc_df$Dataset == 'PREDICTS'),]$avAUC) sd(resamp_auc_df[which(resamp_auc_df$Model == 'LR A' & resamp_auc_df$Dataset == 'PREDICTS'),]$avAUC) range(resamp_auc_df[which(resamp_auc_df$Model == 'CNN A' & resamp_auc_df$Dataset == 'LPD'),]$avAUC) mean(resamp_auc_df[which(resamp_auc_df$Model == 'CNN A' & resamp_auc_df$Dataset == 'LPD'),]$avAUC) sd(resamp_auc_df[which(resamp_auc_df$Model == 'CNN A' & resamp_auc_df$Dataset == 'LPD'),]$avAUC) range(resamp_auc_df[which(resamp_auc_df$Model == 'CNN A' & resamp_auc_df$Dataset == 'PREDICTS'),]$avAUC) mean(resamp_auc_df[which(resamp_auc_df$Model == 'CNN A' & resamp_auc_df$Dataset == 'PREDICTS'),]$avAUC) sd(resamp_auc_df[which(resamp_auc_df$Model == 'CNN A' & resamp_auc_df$Dataset == 'PREDICTS'),]$avAUC)

20c7677981f121ec8649f65816289b6c799064f1

63d65462159caec758526256b242350653bc6c95

/7_1_21_voltage.R

0c70fc5d8de1ce14b3d798a5f8c7de1928f6f8e2

[]

no_license

foliva3/buckthorn_sap_flux

a2db03cbf939a990ea1d417a9b2d12ab23f0a763

ee734385708b818cebc891277af85d2b521da83f

refs/heads/main

2023-06-28T07:41:18.480466

2021-07-20T13:55:13

381,122,923

0

null

UTF-8

R

false

808

r

7_1_21_voltage.R

library(lubridate) library(ggplot2) datav <- read.csv("K:\\Environmental_Studies\\hkropp\\Data\\campus\\buckthorn\\sapflux\\campbell\\07_01_2021\\Sapflow_TableTC.dat", skip = 4, header = FALSE, na.strings = "NAN") tablev <- datav[,c(1,165:166)] colnames(tablev) <- c("date","Htr1", "Htr2") datav$date <- ymd_hms(datav$date) #changes format of dates to POSIXct Datev <- as.POSIXct(tablev$date, format = "%Y-%m-%d %H:%M") #Heater 1 ggplot(data = tablev, aes(Datev, Htr1, group = 1))+ geom_line()+ scale_x_datetime(date_breaks = "1 day", date_labels = "%m/%d")+ ggtitle("Heater 1") #Heater 2 ggplot(data = tablev, aes(Datev, Htr2, group = 1))+ geom_line()+ scale_x_datetime(date_breaks = "1 day", date_labels = "%m/%d")+ ggtitle("Heater 2")

028452f4846650245c578305367c123753f4fd01

679ad602f16cfb52ff7ca24264c51c19a063eb3c

/man/setTest.Rd

5418cac1846e204961d2b0bc38d3a2bee2d84fbf

[]

no_license

mitra-ep/rSEA

cd7cde7f0691507a8a3e8b30eb78f3120a2faf36

20c9a545781daef94c119fccb1db66bdb37dbbbc

refs/heads/master

2021-10-25T05:27:09.743107

2021-10-17T19:59:20

213,402,082

0

1

null

2021-10-17T19:59:20

2019-10-07T14:14:43

R

UTF-8

R

false

true

2,661

rd

setTest.Rd

% Generated by roxygen2: do not edit by hand % Please edit documentation in R/settest.R \name{setTest} \alias{setTest} \title{setTest} \usage{ setTest(pvalue, featureIDs, data, set, testype, testvalue) } \arguments{ \item{pvalue}{The vector of p-values. It can be the name of the covariate representing the Vector of raw p-values in the \code{data} or a single vector but in the latter case it should match the \code{featureIDs} vector} \item{featureIDs}{The vector of feature IDs. It can be the name of the covariate representing the IDs in the \code{data} or a single vector but in the latter case it should match the \code{pvalue} vector} \item{data}{Optional data frame or matrix containing the variables in \code{pvalue} and \code{featureIDs}} \item{set}{The selection of features defining the feature-set based on the the \code{featureIDs}. If missing, the set of all features is selected} \item{testype}{Character, type of the test: "selfcontained" or "competitive". Choosing the self-contained option will automatically set the threshold to zero and the \code{testvalue} is ignored. Choosing the competitive option without a \code{testvalue} will set the threshold to the overall estimated proportion of true hypotheses} \item{testvalue}{Optional value to test against. Setting this value to c along with \code{testype=="competitive"} will lead to testing the null hypothesis against a threshold c. Note: this value needs to be a proportion} } \value{ The adjusted p-value of the specified test for the feature-set is returned. } \description{ calculates the adjusted p-value for the local hypothesis as defined by \code{testtype} and \code{testvalue}. } \examples{ \dontrun{ #Generate a vector of pvalues set.seed(159) m<- 100 pvalues <- runif(m,0,1)^5 featureIDs <- as.character(1:m) # perform a self-contained test for all features settest(pvalues, featureIDs, testype = "selfcontained") # create a random pathway of size 60 randset=as.character(c(sample(1:m, 60))) # perform a competitive test for the random pathway settest(pvalues, featureIDs, set=randset, testype = "competitive") # perform a unified null hypothesis test against 0.2 for a set of size 50 settest(pvalues, featureIDs, set=randset, testype = "competitive", testvalue = 0.2 ) } } \references{ Mitra Ebrahimpoor, Pietro Spitali, Kristina Hettne, Roula Tsonaka, Jelle Goeman, Simultaneous Enrichment Analysis of all Possible Gene-sets: Unifying Self-Contained and Competitive Methods, Briefings in Bioinformatics, , bbz074, https://doi.org/10.1093/bib/bbz074 } \seealso{ \code{\link{setTDP}} \code{\link{SEA}} } \author{ Mitra Ebrahimpoor \email{m.ebrahimpoor@lumc.nl} }

71ae15663b6a6f00506e94b503f7e8b06d550dfe

8dbe9cebc5603e7a05de12dc997986468c431563

/create_human.R

0c96b3c3a3f512530d3e8663694afcab18e36627

[]

no_license

ottoy91/IODS-project

ac1bbb68abcd7aee6ef1246180632c3cb83bd9d4

4994bc33d7fc2851cfc86ac26bb879b736098774

refs/heads/master

2020-04-05T15:46:10.330083

2018-12-09T16:58:19

156,983,096

0

null

2018-11-10T13:19:54

null

UTF-8

R

false

2,065

r

create_human.R

### 25.11.2018/Otto Ylöstalo/IODS-project Week 4 ### library(dplyr) #1 hd <- read.csv("http://s3.amazonaws.com/assets.datacamp.com/production/course_2218/datasets/human_development.csv", stringsAsFactors = F) gii <- read.csv("http://s3.amazonaws.com/assets.datacamp.com/production/course_2218/datasets/gender_inequality.csv", stringsAsFactors = F, na.strings = "..") #2 #structure, dimensions and summary of hd str(hd) dim(hd) summary(hd) #structure, dimensions and summary of gii str(gii) dim(gii) summary(gii) #3 names(hd) <- c("hd_rank","country","HDI","LE_at_birth","exp_edu_years","mean_edu_years","GNI","GNI_rank") names(gii) <- c("gii_rank","country","gii","mmr","abr","prp","eduF","eduM","labF","labM") #4 gii <- mutate(gii, edu_ratio = eduF/eduM) gii <- mutate(gii, lab_ratio = labF/labM) #5 human <- inner_join(hd, gii, by = "country", suffix = c(".hd", ".gii")) #save file write.csv(human,"human.csv",row.names = F) ##2.12.2018/Otto Ylöstalo/IODS-project Week 5 ##Data contains different variables that are related to human development reports from different countries ##and how these and not the economic growth alone are a criteria for a countrys development. ##Soure: ##http://s3.amazonaws.com/assets.datacamp.com/production/course_2218/datasets/human_development.csv ##http://s3.amazonaws.com/assets.datacamp.com/production/course_2218/datasets/gender_inequality.csv str(human) dim(human) #dataset combines 19 indicators from 195 countries in the world related to HDI library(stringr) str(human$GNI) #1 str_replace(human$GNI, pattern=",", replace =".") %>% as.numeric #2 keep <- c("country", "edu_ratio", "lab_ratio", "LE_at_birth", "exp_edu_years", "GNI", "mmr", "abr", "prp") human <- select(human, one_of(keep)) #3 complete.cases(human) data.frame(human[-1], comp = complete.cases(human)) human <- filter(human, complete.cases(human)) complete.cases(human) #4 print(human$country) last <- nrow(human) - 7 human <- human[1:last, ] #5 rownames(human) <- human$country human <- human[,2:9] write.csv(human,"human.csv",row.names = F)

944ee9133e4bc4503c704ff22fcadaa70778fe7e

f0c9e167c8ceae9388986d22e89eb7293c664343

/data/FGClimatology/R/LV_windrose.R

7c0c9add832bf6f2e761878681a2870469105ed9

[]

no_license

gavin971/r_dev_messkonzept

1a2c91e51d45f18df65476bf04b7918c809f0503

af6073c2b18f1f036a528bf8df6143bf1b51a561

refs/heads/master

2020-03-19T03:13:34.235435

2015-05-22T08:07:06

null

0

null

UTF-8

R

false

3,027

r

LV_windrose.R

windrose<-function(windspeed, winddir, r=5,p=10){ # Stand: 03.12.2013 # # windrose ist eine Funktion um Winddaten aus den Campbell Datenloggern in Windrosen zu plotten. # Voraussetzung ist entsprechende Vorprozessierung (keine NAN) und Installation des climatol-packages # Da das climatol package von der CRAN repository entfernt wurde (warum auch immer), muss es einmalig aus dem Archiv heruntergeladen werden. # <http://cran.r-project.org/src/contrib/Archive/climatol/> da die neuste Version herunterladen und mit #! install.packages(file, repos = NULL, type = "source") installieren (file: character vektor of directory/download path) # # Autor des Skripts: Carsten Vick # Code im Skript: Britta Jaenicke # # windspeed: a numeric vector containing windspeed data # winddir: a numeric vector containing winddirection data # # Anmerkung des Autors: Ich habe den folgenden Code fast genauso von Britta übernommen und einfach eine Funktion daraus gemacht. # Die Variablennamen sind an einigen Stellen uneindeutig bezeichnet und sind oft Hilfsvariablen. # Falls es Fragen zum Code gibt, bitte an <britta.jaenicke@yahoo.de> wenden. library(climatol) # laden des packages data(windfr) # laden des data.frames aus dem Beispiel (ist bereits gut vordefiniert) windv_class<- windfr*0 # löschen der Daten aus dem Beispiel dirup <- c(11.25,33.75,56.25,78.75,101.25,123.75,146.25,168.75,191.25,213.75,236.25,258.75,281.25,303.75,326.25,348.75) #festlegen der Gruppenobergrenzen dirlow <-c(348.75,11.25,33.75,56.25,78.75,101.25,123.75,146.25,168.75,191.25,213.75,236.25,258.75,281.25,303.75,326.25) #festlegen der Grupenuntergrenzen speedup <- c(0.5,1.0,1.5,2.0) #festlegen der Geschwindigkeitsobergrenzen speedlow <- (c(0.0,0.5,1.0,1.5)) #festlegen der Geschwindigkeitsuntergrenzen rownames(windv_class)<- c(paste(toString(speedlow[1]),"-",toString(speedup[1])), paste(toString(speedlow[2]),"-",toString(speedup[2])), paste(toString(speedlow[3]),"-",toString(speedup[3])), paste(">",toString(speedup[3]))) # ändern der rownames des Beispiels # die folgende Schleife ordnet die Windgeschwindigkeiten in entsprechende Geschwindigkeitsgruppen for (a in seq_along(speedup)){ idx <- which(windspeed >=speedlow[a] & windspeed < speedup[a]) # die jetzt innerhalb folgende Schleife ordnet die Windrichtung in eine Richtungsgruppe for (b in seq_along(dirlow)){ if (b == 1) cnt1 <- length(which(winddir[idx] >=dirlow[b])) if (b == 1) cnt2 <- length(which(winddir[idx] < dirup[b])) if (b == 1) cnt <- cnt1+cnt2 if (b > 1) cnt <- length(which(winddir[idx] >=dirlow[b] & winddir[idx] < dirup[b])) windv_class[a,b]<-cnt } } # die so entstandene windv_class hat genau die Dimensionen die von der rosavent-Funktion benötigt wird. rosavent(windv_class,r,p,ang=-3*pi/16,main="Windrose der Station") }

4371e6e834eb711da1ca59d1f0097936c5cf7260

26021ab16e74ecfdca657c2ba3fbad6ef3f92ed5

/ShipTracker/tests/testthat/test-map_server.R

65d94b85cd4378d279bf31cdc80a49e7604a1f8b

[]

no_license

radbasa/shiptracker

ed034b33b7d0326db599e3841ab5e93039d5c6a8

5a4dfbe7bbccb113d1d00b2cfd6bc88d7c2fc281

refs/heads/master

2023-07-28T20:47:45.363191

2021-09-30T10:53:19

411,481,093

0

null

UTF-8

R

false

572

r

test-map_server.R

setwd("../..") source("global.R") sdm <- ShipData$new(global$data_file_path) test_that("Map Server outputs a leaflet map", { testthat::local_edition(3) selected_ship <- list( ship_legs = reactive({sdm$get_ship_legs(316100)}), ship_info = reactive({sdm$get_ship_info(316100)}) ) testServer(mapServer, args = list(selected_ship = selected_ship), { expect_s3_class(output$ship_map, "json") # This saves the Mapbox access token in the snapshot. Gitignore this. expect_snapshot(output$ship_map) }) })

ab06f351354ffea2fb307cffc314d96bf44d372d

6589453f2dfec010434963966841d0de183d3a15

/project_scripts/orthogroups.R

a077db740a6363b67bf96e75377e92a1688a27d3

[]

no_license

Werner0/anomura

c3256da9630e4db54dac3386dd5e4734adc3b91b

6fc9f5eb205bac7ec55ad4c503f9cb34b0b5496b

refs/heads/main

2023-03-09T09:58:05.142920

2021-02-28T05:38:23

333,056,367

0

null

UTF-8

R

false

4,250

r

orthogroups.R

#SNIPPETS (USE BEFORE SCRIPT) library(data.table) library(tibble) library(plyr) library(gridExtra) if (!getwd()=="/Users/wernerveldsman/Desktop/Rsources/Orthogroups/") { setwd("/Users/wernerveldsman/Desktop/Rsources/Orthogroups/")} #perspecies <- fread("Statistics_PerSpecies.tsv") #Cut file manually before loading #colSums(perspecies[,2:10]) orthonumbers <- fread("Orthogroups.GeneCount.tsv") #orthonumbers <- orthonumbers[apply(orthonumbers!=0, 1, all),] #orthonumbers[, "sum"] <- orthonumbers$kingcrab.aa+orthonumbers$bluekingcrab-(orthonumbers$lobster.aa)-(orthonumbers$coconutcrab.aa) #orthonumbers <- orthonumbers[apply(orthonumbers[,c(2:10)],1,function(x) all(x>0)),] orthonumbers <- orthonumbers[apply(orthonumbers[,c(2:10)],1,function(x) all(x==1)),] #orthonumbers <- orthonumbers[orthonumbers$amphipod.aa<3&orthonumbers$isopod.aa<3,] orthonumbers <- orthonumbers[order(-orthonumbers$coconutcrab.aa),] orthos <- fread("Orthogroups.tsv", sep="\t") #singleorthos <- orthonumbers$Orthogroup alltables_reduced <- fread("alltables_reduced.txt") #singletable <- data.table() #for (i in singleorthos) { #query <- strsplit(gsub(" ","",orthos$bluekingcrab[orthos$Orthogroup=="OG000001"]),",") #query <- query[[1]] #subject <- alltables_reduced[alltables_reduced$query_name %in% query,] #OGdata <- count(subject[subject$organism=="bluekingcrab",c("Preferred_name","eggNOG free text desc.")]) # OGplaceholder <- data.table(Preferred_name = NA,eggNOG.free.text.desc.=NA) # if (nrow(OGdata)<1) {singletable <- rbind(singletable, OGplaceholder, fill = T)} # singletable <- rbind(singletable, OGdata[,1:2], fill = T) } #highcopygenes <- count(alltables_reduced[,c("Preferred_name","organism")]) #highcopygenes <- as.data.table(highcopygenes[order(-highcopygenes$freq),]) #Choose a gene from here for use in next line #highcopyidentifiers <- alltables_reduced[alltables_reduced$Preferred_name %in% "KIF22"&alltables_reduced$organism=="coconutcrab",c("query_name")] #orthogroupsofinterest <- orthos[orthos$coconutcrab.aa %in% highcopyidentifiers$query_name, c("Orthogroup")] library("ape") library("Biostrings") library("ggplot2") library("ggtree") #library("flextable") tree <- read.tree("SpeciesTree_rooted_node_labels.txt") nodedata <- as.data.frame(tree$node.label) nodedata$duplications <- c(758,846,740,1449,227,581,2935,11042) nodedata$duplicationsl <- c("","","","","","","","") colnames(nodedata) <- c("newick_label","duplications","duplicationsl") groupInfo <- split(tree$tip.label, gsub("_\\w+", "", tree$tip.label)) names(groupInfo) <- gsub("\\..*","",names(groupInfo)) names(groupInfo) <- gsub("\\-.*","",names(groupInfo)) names(groupInfo) <- c("A. vulgare", "P. hawaiensis", "Achelata", "P. virginalis", "P. trituberculatus", "Anomura", "Anomura", "Anomura", "L. vannamei") tree$tip.label <- c("Armadillidium vulgare", "Parhyale hawaiensis", "Panulirus ornatus", "Procambarus virginalis", "Portunus trituberculatus", "Birgus latro", "Paralithodes camtschaticus", "Paralithodes platypus", "Litopenaeus vannamei") tree <- groupOTU(tree, groupInfo) #tree <- root(tree, node = 011, edgelabel = TRUE) g <- ggtree(tree, size=1.5) %<+% nodedata + geom_treescale() + xlim(NA, 1.1) g <- rotate(g,14) #Lobster g <- rotate(g,16) #Anomura g2 <- g + geom_label(aes(label = duplicationsl, fill = duplications), show.legend = FALSE) + theme(legend.position = NULL) + scale_fill_gradientn(colors = RColorBrewer::brewer.pal(3, "YlGnBu")) + #geom_cladelabel(node=12, label="Pleocyemata", color="red2", offset=0.6, align=TRUE, angle = 90, offset.text = 0.05, hjust = 0.5, barsize = 1) + #geom_cladelabel(node=17, label="Anomura", color="red2", offset=0.8, align=TRUE, angle = 90, offset.text = 0.05, hjust = 0.5, barsize = 1) + theme_tree2() #geom_text(aes(subset=(node==508), label = italic('Acetobacter spp.')), parse=TRUE, colour="blue", hjust=-.02) #(isopod.aa_6304:0.253617,(amphipod.aa_8566:0.638071,(((lobster.aa_49883:0.202489,marbled.aa_11462:0.19829)N4_227:0.0536186,(swimmingcrab.aa_5873:0.346618,(coconutcrab.aa_85938:0.181323,(kingcrab.aa_78264:0.00540666,bluekingcrab_84676:0.0272019)N7_11042:0.215072)N6_2935:0.188058)N5_581:0.0669401)N3_1449:0.0654401,whiteshrimp.aa_4397:0.293433)N2_740:0.234702)N1_846:0.253617)N0_758;

f6e008d9db02fdf46e7eab70fb78de1a54575dd1

21051e5f5923f2f88fe0f869201e55a16c848434

/man/ev_surveillance.Rd

79d28eb3e023922487854ddcbe67f84c02c96b96

[]

no_license

XiangdongGu/hkdata

4bb4011f2c3f22f7ae9f20ece7e354d1ba70c70f

453261388c8e8b30d8e7912c0d20ba719afeb185

refs/heads/master

2020-03-23T02:06:55.540330

2019-08-02T08:55:08

140,956,671

3

0

null

UTF-8

R

false

true

1,611

rd

ev_surveillance.Rd

% Generated by roxygen2: do not edit by hand % Please edit documentation in R/hk-health.R \name{ev_surveillance} \alias{ev_surveillance} \title{Get Data for "Department of Health: EV Scan's Figures Data"} \format{A data frame with 11 variables.\cr * `year`: Year\cr * `week`: Week\cr * `from`: From (Date)\cr * `to`: To (Date)\cr * `n_ev71`: Number of EV71 cases by week\cr * `n_hfmd_inst`: Number of HFMD institutional outbreaks by week\cr * `n_hfmd_hosp_adm`: Number of hospital admission episodes of HFMD by week\cr * `rate_hfmd_aed`: Accident & Emergency Department surveillance of HFMD syndrome group (per 1000 coded cases)\cr * `prop_hfmd_ccc_kg`: Proportion of child care centres/kindergartens (CCC/KG) with HFMD cases based on HFMD sentinel surveillance at CCC/KG by week\cr * `rate_hfmd_pvt_med`: Consultation rate for HFMD based on HFMD sentinel surveillance among private medical practitioner clinics by week (per 1000 consultations) \cr * `rate_hfmd_gopc`: Consultation rate for HFMD based on HFMD sentinel surveillance among General Out-patient Clinics by week (per 1000 consultations)} \source{ <https://data.gov.hk/en-data/dataset/hk-dh-chpsebcdde-ev-scan> } \usage{ ev_surveillance(path = ".", keep = FALSE) } \arguments{ \item{path}{path to save the file} \item{keep}{whether to keep the file after read} } \description{ Hand, foot and mouth disease surveillance data including number of EV71 cases, institutional outbreaks, hospital surveillance and sentinel surveillance. \cr \cr UPDATE FREQUENCY: WEEKLY } \details{ * Recent data are provisional figures and subject to further revision. }

c429b7985b557ba1196feefd4eba78e3c3983905

15d1e3d8f2f3dc5dda60e264be1153a202934100

/man/iterate_umap.Rd

6e80e7ef53522097c14292860e38fe1c13772f22

[ "MIT" ]

permissive

Ryan-Laird/PhenoClustR

547bcad7c327e4e9567b54eead77a354734ed94b

c294c3b04449503e726a89c205d8800805c029af

refs/heads/master

2022-04-25T09:16:41.941153

2020-04-28T22:39:33

254,729,594

2

0

null

UTF-8

R

false

true

693

rd

iterate_umap.Rd

% Generated by roxygen2: do not edit by hand % Please edit documentation in R/iterate_umap.R \name{iterate_umap} \alias{iterate_umap} \title{Run UMAP with many combinations of hyperparameters.} \usage{ iterate_umap(dat, dat.labels, params) } \arguments{ \item{dat}{Unlabeled dataframe for UMAP input} \item{dat.labels}{Labels for dat} \item{params}{Sqaure dataframe of UMAP hyperparameter values (all required): \itemize{ \item n_neighbors \item min_dist \item n_components \item metric \item method \item seed }} } \value{ Nested tibble::tibble() containing UMAP objs, output layout, labeled output, and 2D/3D plot obj. } \description{ Run UMAP with many combinations of hyperparameters. }

d65e2ff124f576de743ab2cdece81f8ce6b82c3d

43ee56af5973642c4ee4a8044ea5aef8ef4b6477

/lib/XGBoost_model_fitting_WHOLE_DATA.R

e330f19b0cfb7a73abf8614188d5d99bc8af7dc9

[]

no_license

TZstatsADS/Fall2016-proj5-grp3

865651c1ca1afd6a2306c5bdefa273d29cb111b8

a0b4a4c8e55e9cf8e2aa87af2de378f5c5a4bbd9

refs/heads/master

2021-05-01T02:41:01.497237

2016-12-13T21:03:28

74,604,825

0

1

null

UTF-8

R

false

1,295

r

XGBoost_model_fitting_WHOLE_DATA.R

library(dplyr) library(xgboost) setwd('C:\\Users\\LENOVO\\Desktop\\Academic\\ADS\\project_5') load('df(final).RData') train_df <- df for (i in c(3, 4, 5, 12, 13, 14, 15, 17, 18, 21, 24)){ train_df[, i] <- as.factor(train_df[ , i]) } ###Train the model train_df_X <- data.matrix(train_df[ , c(3:6, 8:15, 17:24)]) cv_number <- nrow(train_df_X) k_folds <- cut(1:cv_number, breaks = 5, labels = FALSE) cv_errors <- matrix(0, nrow = 24, ncol = 5) for (k in 1:5){ models <- list() cv_indices <- which(k_folds == k, arr.ind = TRUE) cv_train <- train_df_X[cv_indices, ] cv_test <- train_df_X[-cv_indices, ] for (i in 25:48){ cv_y <- train_df[cv_indices , i] temp_model <- xgboost(data = cv_train, label = cv_y, nrounds = 50, missing = NaN ) models[[i - 24]] <- temp_model print(i) } pred_matrix <- matrix(0, ncol = 24, nrow = nrow(cv_test)) error_rate <- rep(0, 24) for (j in 1:24){ pred <- round(predict(models[[j]], cv_test, missing = NaN)) pred_matrix[ , j] <- pred error_rate[j] <- sum(pred != train_df[-cv_indices , j + 24]) / length(pred) print(j) } cv_errors[, k] <- error_rate }

adcabf5313e44d783089b27e3aa529473fc0965e

d5a14ba66821cab667def0c8730dbbef1551b762

/man/IDW.Rd

6181b1ea527228ba23ddd7ef73a658cc418a4130

[]

no_license

overeem11/RAINLINK

5f964a6bcfae67e3f32c42a55c5dcd6fcca20f6d

5e38a76f8b99ccbb444d6486b96c7ce4c3cb9954

refs/heads/master

2023-07-11T00:15:34.423165

2023-06-20T13:41:26

48,035,739

12

20

null

2018-06-07T13:40:54

2015-12-15T10:22:57

R

UTF-8

R

false

true

1,050

rd

IDW.Rd

% Generated by roxygen2: do not edit by hand % Please edit documentation in R/IDW.R \name{IDW} \alias{IDW} \title{Subfunction for inverse distance weighted interpolation on point data.} \usage{ IDW(idp, rain.grid, Rainlink) } \arguments{ \item{idp}{The inverse distance weighting power.} \item{rain.grid}{Interpolation grid in a Cartesian coordinate system.} \item{Rainlink}{Coordinates of links in a Cartesian coordinate system and rainfall intensity (latitude in km, longitude in km, intensity in mm h\eqn{^{-1}}).} } \value{ Interpolated field of rainfall intensities. } \description{ Subfunction for inverse distance weighted interpolation on point data. } \examples{ IDW(idp=idp,rain.grid=rain.grid,Rainlink=Rainlink) } \references{ ''ManualRAINLINK.pdf'' Overeem, A., Leijnse, H., and Uijlenhoet, R., 2016: Retrieval algorithm for rainfall mapping from microwave links in a cellular communication network, Atmospheric Measurement Techniques, 9, 2425-2444, https://doi.org/10.5194/amt-9-2425-2016. } \author{ Aart Overeem & Hidde Leijnse }

955ecbada9a0b4ef8f29e1232259911b65081a14

e525513e27156b29a12a0aa585327faa6241ed53

/R/cea_policy_tree.R

46df2f8bc906909cc38452d3c8a2f2416d2633ef

[]

no_license

bonander/CEAforests

b9c340305a5004d06c94802fed90bebee08d0643

111db3c0518a1379d2635aaafe524270244e7d40

refs/heads/master

2023-03-26T03:57:52.073071

2021-03-25T10:40:52

2021-03-25T10:41:31

228,402,995

7

0

null

UTF-8

R

false

11,944

r

cea_policy_tree.R

#' @title Train a policy tree after a CEA forest. #' @description \code{cea_policy_tree} Trains an efficient policy decision tree given a CEA forest (a wrapper for policytree::policy_tree). #' #' @param forest A trained CEA forest. #' @param X A covariate matrix containing variables that are to be used in the policy tree. #' @param WTP Willingness to pay for a one unit increase in the outcome. If NULL, the WTP supplied to the CEA forest is used. #' @param depth The desired depth for the decision tree. #' @param ci.level Desired significance level (for confidence intervals). #' @param robust.se Whether or not robust (sandwich) standard errors are desired. Defaults to FALSE. #' #' #' @references Athey, S., & Wager, S. (2017). Efficient policy learning. arXiv preprint arXiv:1702.02896. #' #' @return Returns a trained policy tree. #' @examples #' \dontrun{ #' To be added... #' } #' @importFrom utils installed.packages #' @import stats #' @export cea_policy_tree = function(forest, X, WTP=NULL, depth=2, ci.level=0.95, robust.se=FALSE) { if (isTRUE("policytree" %in% rownames(installed.packages())==FALSE)) { stop("Package \"policytree\" must be installed to estimate policy trees.") } if (isTRUE(any(class(forest) %in% c("CEAforests")))) { if (is.null(WTP)==TRUE) { WTP = forest[["WTP"]] warning(paste("WTP not specified, assuming WTP = ", WTP, ", as supplied to the CEAforests object.", sep="")) } gamma1 = cate.prepare(forest[["outcome.forest"]])*WTP gamma2 = cate.prepare(forest[["cost.forest"]]) gamma = gamma1-gamma2 Gamma = cbind(control=-gamma, treated=gamma) treefit = policytree::policy_tree(X, Gamma, depth=depth) } else { stop("Unrecognized or unsupported forest object. Please supply a CEAforests object.")} results = list() treefit$n.sample = nrow(Gamma) results[["tree"]] = treefit results[["X"]] = X class(results) = c("cea_policy_tree", "CEAforests") return(results) } #' @title Conduct inference for a personalized treatment policy. #' @description Conduct inference for a personalized treatment policy, either using a manually specified policy or a learned policy. #' #' @param forest A trained CEA forest. #' @param treat.policy A logical vector or cea policy tree defining the subset covered by the policy. #' @param WTP Willingness to pay for a one unit increase in the outcome. If NULL, the WTP supplied to the CEA forest is used. #' @param ci.level Desired significance level (for confidence intervals). #' @param robust.se Whether or not robust (sandwich) standard errors are desired. Defaults to FALSE. Ignored when boot.ci=TRUE. #' @param boot.ci Whether or not bootstrapped confidence intervals are desired. Defaults to FALSE. #' @param R The number of bootstrap replications. Defaults to 999. Ignored when boot.ci=FALSE. #' #' @return Returns a matrix containing estimates for the average welfare gain per population member under various treatment policies (treat everyone vs. treat no one; treat suggested subset vs. treat no one; treat suggested subset vs. treat everyone). Also outputs the share of the popuation covered by the policy. #' @examples #' \dontrun{ #' To be added... #' } #' @import stats #' @import boot #' @export infer_policy = function(forest, treat.policy, WTP=NULL, ci.level=0.95, robust.se=FALSE, boot.ci=FALSE, R=999) { subset = treat.policy if (isTRUE(any(class(forest) %in% c("CEAforests")))) {#Check if forest class OK if (is.null(WTP)==TRUE) { WTP = forest[["WTP"]] warning(paste("WTP not specified, assuming WTP = ", WTP, ", as supplied to the CEAforests object.", sep="")) } #Estimate doubly robust scores gamma1 = cate.prepare(forest[["outcome.forest"]])*WTP gamma2 = cate.prepare(forest[["cost.forest"]]) gamma = gamma1-gamma2 if (isTRUE("cea_policy_tree" %in% class(subset))) { #Predict the suggested policy using the supplied X X = subset[["X"]] predicted.action = predict(subset[["tree"]], newdata=X) P = as.numeric(predicted.action==2) } else { if (class(subset) == "logical" & length(subset) == length(forest[["outcome.forest"]]$Y.hat)) { subset <- which(subset) } if (!all(subset %in% 1:length(forest[["outcome.forest"]]$Y.hat))) { stop(paste("treat.policy must be a vector contained in 1:n,", "a boolean vector of length n or a trained CEA policy tree.")) } P = rep(0, length(gamma)); P[subset] = 1 } tau_tr = gamma #Scores for treat everyone policy_suggested = tau_tr*P #Scores for suggested policy policy_diff = policy_suggested-tau_tr #Scores for difference between suggested and treat everyone #Fit intercept only models to get mean and variance all.m = lm(tau_tr~1) sugg.m = lm(policy_suggested~1) vs.m = lm(policy_diff~1) est.tr.all = as.vector(coef(all.m)) est.suggested = as.vector(coef(sugg.m)) est.diff = as.vector(coef(vs.m)) ests = c(est.tr.all, est.suggested, est.diff) if (!isTRUE(boot.ci)) {#Asymptotic variance estimates #Extract estimates and standard errors if (isTRUE(robust.se)) { se.tr.all = as.vector(sqrt(diag(sandwich::vcovHC(all.m)))) se.suggested = as.vector(sqrt(diag(sandwich::vcovHC(sugg.m)))) se.diff = as.vector(sqrt(diag(sandwich::vcovHC(vs.m)))) } else { se.tr.all = as.vector(sqrt(diag(vcov(all.m)))) se.suggested = as.vector(sqrt(diag(vcov(sugg.m)))) se.diff = as.vector(sqrt(diag(vcov(vs.m)))) } ses = c(se.tr.all, se.suggested, se.diff) #Get confidence intervals lowers = ests-ses*qt(1-(1-ci.level)/2, df=length(gamma)-1) uppers = ests+ses*qt(1-(1-ci.level)/2, df=length(gamma)-1) } else {#Else bootstrap scores bootres=boot.policy_scores(tau_tr, policy_suggested, R, 1-ci.level)[[1]] ses = bootres[,1] lowers = bootres[,2] uppers = bootres[,3] } #Share of population who are treated (in suggested policy) tr.sugg.share = mean(P) #Tidy up and print results res = as.data.frame(cbind(ests,ses,lowers,uppers)) res = rbind(res, c(tr.sugg.share,NA,NA,NA)) colnames(res) = c("Estimate", "Std.Err", "Lower.CI", "Upper.CI") rownames(res) = c("Average NMB, new-for-all vs control-for-all", "Average NMB, suggested policy vs control-for-all", "Difference in NMB, suggested vs. new-for-all", "Prop. who gets new treatment, suggested policy") return(res) } else (stop("Unrecognized forest object.")) } #' Writes each node information #' If it is a leaf node: show it in different color, show number of samples, show leaf id #' If it is a non-leaf node: show its splitting variable and splitting value #' @param tree the tree to convert #' @param index the index of the current node #' @param group.names names of the treatment and control states (defaults to c("Do not reimburse", "Reimburse")) #' @keywords internal cea_create_dot_body <- function(tree, index = 1, group.names=c("Do not reimburse", "Reimburse")) { nam1 <- group.names[1] nam2 <- group.names[2] #n = tree$n.sample node <- tree$nodes[[index]] # Leaf case: print label only if (node$is_leaf) { action <- node$action action <- ifelse(action==1, nam1, nam2) line_label <- paste(index - 1, ' [shape=box,style=filled,color="White", height=0.2, label="', action, "\n", '"];', sep="") return(line_label) } # Non-leaf case: print label, child edges if (!is.null(node$left_child)) { edge <- paste(index - 1, "->", node$left_child - 1) if (index == 1) { edge_info_left <- paste(edge, '[labeldistance=2.5, labelangle=45, headlabel="Yes"];') } else { edge_info_left <- paste(edge, " ;") } } else { edge_info_right <- NULL } if (!is.null(node$right_child)) { edge <- paste(index - 1, "->", node$right_child - 1) if (index == 1) { edge_info_right <- paste(edge, '[labeldistance=2.5, labelangle=-45, headlabel="No"]') } else { edge_info_right <- paste(edge, " ;") } } else { edge_info_right <- NULL } variable_name <- tree$columns[node$split_variable] node_info <- paste(index - 1, '[label="', variable_name, "<=", round(node$split_value, 2), '"] ;') this_lines <- paste(node_info, edge_info_left, edge_info_right, sep = "\n" ) left_child_lines <- ifelse(!is.null(node$left_child), cea_create_dot_body(tree, index = node$left_child), NULL ) right_child_lines <- ifelse(!is.null(node$right_child), cea_create_dot_body(tree, index = node$right_child), NULL ) lines <- paste(this_lines, left_child_lines, right_child_lines, sep = "\n") return(lines) } #' Export a tree in DOT format. #' This function generates a GraphViz representation of the tree, #' which is then written into `dot_string`. #' @param tree the tree to convert #' @param group.names names of the treatment and control states (defaults to c("Do not reimburse", "Reimburse")) #' @keywords internal cea_export_graphviz <- function(tree,group.names=c("Do not reimburse", "Reimburse")) { header <- "digraph nodes { \n node [shape=box] ;" footer <- "}" body <- cea_create_dot_body(tree,group.names=group.names) dot_string <- paste(header, body, footer, sep = "\n") return(dot_string) } #' Plot a cea_policy_tree tree object. #' @param x The tree to plot #' @param group.names names of the treatment and control states (defaults to "Control treatment", "New treatment") #' @param ... Additional options (currently ignored). #' #' @method plot cea_policy_tree #' @export plot.cea_policy_tree <- function(x,group.names=c("Do not reimburse", "Reimburse"), ...) { if (!requireNamespace("DiagrammeR", quietly = TRUE)) { stop("Package \"DiagrammeR\" must be installed to plot trees.") } dot_file <- cea_export_graphviz(x[["tree"]],group.names=group.names) DiagrammeR::grViz(dot_file) } #' @title Bootstrap average effects #' @description \code{boot.dr_scores} Bootstraps doubly robust scores and obtains accelerated bootstrap confidence intervals (BCa). #' #' @param Gamma_all Scores for treating everyone vs treating no-one. #' @param Gamma_policy Scores for suggested policy vs treating no-one. #' @param R Number of bootstrap replicates. #' @param alpha Desired confidence level. #' @keywords internal #' @return Returns a matrix with estimated standard errors and BCa confidence intervals. #' @export #' boot.policy_scores <- function(Gamma_all, Gamma_policy, R, alpha) { df = as.data.frame(cbind(Gamma_all, Gamma_policy)) n = nrow(df) bfun = function(data, indices, Gamma_all, Gamma_policy) { d=data[indices,] tr_all=mean(d[,Gamma_all]) tr_policy=mean(d[,Gamma_policy]) tr_diff=tr_policy-tr_all res=c(tr_all,tr_policy,tr_diff) return(res)} b=boot::boot(data=df, bfun, R=R, Gamma_all="Gamma_all", Gamma_policy="Gamma_policy") all_se=sd(b$t[,1]) policy_se=sd(b$t[,2]) diff_se=sd(b$t[,3]) res = list() if (R<=n) { warning("Number of bootstrap replicates R is smaller than the number of rows in the data. BCa confidence intervals cannot not be computed. Please increase R.") res[[1]] = cbind(c(NA,NA,NA), c(NA,NA,NA), c(NA,NA,NA))} if (R>n) { bci_all=boot::boot.ci(b, index=1, conf=1-alpha, type="bca") bci_policy=boot::boot.ci(b, index=2, conf=1-alpha, type="bca") bci_diff=boot::boot.ci(b, index=3, conf=1-alpha, type="bca") lower_all = bci_all$bca[,4] upper_all = bci_all$bca[,5] lower_policy = bci_policy$bca[,4] upper_policy = bci_policy$bca[,5] lower_diff = bci_diff$bca[,4] upper_diff = bci_diff$bca[,5] ses = c(all_se,policy_se,diff_se) lowers = c(lower_all,lower_policy,lower_diff) uppers = c(upper_all, upper_policy, upper_diff) res = list() res[[1]] = cbind(ses, lowers, uppers) } res[[2]] = b$t return(res) }

ad5261aa816f0510f566509388a75e16cb9a5031

7b74f00cd80694634e6925067aaeb6572b09aef8

/2020/notes-2020/session_files/session_6_fe8828_r_blockchain-master/fe8828_r_blockchain-master/node_server/blockchain-node-server.R

2c825347df94ee8f21d5f9f7ef7c6a894f6e6487

[]

no_license

leafyoung/fe8828

64c3c52f1587a8e55ef404e8cedacbb28dd10f3f

ccd569c1caed8baae8680731d4ff89699405b0f9

refs/heads/master

2023-01-13T00:08:13.213027

2020-11-08T14:08:10

107,782,106

1

0

null

UTF-8

R

false

6,462

r

blockchain-node-server.R

list.of.packages <- c("uuid") new.packages <- list.of.packages[!(list.of.packages %in% installed.packages()[,"Package"])] if(length(new.packages)) install.packages(new.packages) require(uuid) # make sure you put the path of your blockchain.R file source("blockchain.R") # Generate a globally unique address for this node node_identifier = gsub('-','',UUIDgenerate()) # Instantiate the Blockchain blockchain = Blockchain() # genesis block blockchain$nextBlock(previousHash=1, nonce=100) #* @get /chain/show #* @html function(req) { render.html <- "" render.html <- paste0(render.html, '<div>') render.html <- paste0(render.html, '<h1>Current nodes:</h1>') if (length(blockchain$nodes) > 0) { for (i in 1:length(blockchain$nodes)) { render.html <- paste0(render.html, '<b>Node:</b>' , i ,'<br>') render.html <- paste0(render.html, 'name:', blockchain$nodes[i][1]) render.html <- paste0(render.html, '<br>') } } render.html <- paste0(render.html, '<br>') render.html <- paste0(render.html, '</div>') render.html <- paste0(render.html, '<div>') render.html <- paste0(render.html, '<h1>Current transactions:</h1>') if (length(blockchain$currentTransactions) > 0) { for (i in 1:length(blockchain$currentTransactions)) { render.html <- paste0(render.html, '<b>Transaction:</b>', i ,'<br>') render.html <- paste0(render.html, 'sender:', blockchain$currentTransactions[i]$transaction$sender) render.html <- paste0(render.html, '<br>') render.html <- paste0(render.html, 'recipient:', blockchain$currentTransactions[i]$transaction$recipient) render.html <- paste0(render.html, '<br>') render.html <- paste0(render.html, 'amount:', blockchain$currentTransactions[i]$transaction$amount) render.html <- paste0(render.html, '<br>') } } render.html <- paste0(render.html, '<br>') render.html <- paste0(render.html, '</div>') render.html <- paste0(render.html, '<div>') render.html <- paste0(render.html, '<h1>Current block:</h1>') for (i in 1:blockchain$lastBlock()$block$index) { render.html <- paste0(render.html, '<br>') render.html <- paste0(render.html, '<b>Block nr:</b>', blockchain$chain[i]$block$index) render.html <- paste0(render.html, '<br>') render.html <- paste0(render.html, '<b>Transactions:</b>') render.html <- paste0(render.html, '<br>') if (length(blockchain$chain[i]$block$transactions) > 0 ) { for (j in 1:length(blockchain$chain[i]$block$transactions)) { render.html <- paste0(render.html, blockchain$chain[i]$block$transactions[j]) render.html <- paste0(render.html, '<br>') } } render.html <- paste0(render.html, '<b>Nonce:</b>') render.html <- paste0(render.html, '<br>') render.html <- paste0(render.html,blockchain$chain[i]$block$nonce) render.html <- paste0(render.html, '<br>') if (i > 1) { render.html <- paste0(render.html, "<b>Proof guess:</b>") render.html <- paste0(render.html, '<br>') render.html <- paste0(render.html, blockchain$guessProof(blockchain$chain[i-1]$block$nonce, blockchain$chain[i]$block$nonce)) render.html <- paste0(render.html, '<br>') } render.html <- paste0(render.html, '<hr>') } render.html <- paste0(render.html, '<br>') render.html <- paste0(render.html, '</div>') render.html } #* @serializer custom_json #* @get /chain function(req) { list('length'= length(blockchain$chain), 'chain' = blockchain$chain) } #* @serializer custom_json #* @get /transactions/new #* @post /transactions/new function(req, sender, recipient, amount) { # eg req_json <- '{"sender": "my address", "recipient": "someone else address", "amount": 5}' # values <- jsonlite::fromJSON(req_json) if (req$REQUEST_METHOD == "GET") { values <- list(sender = sender, recipient = recipient, amount = amount) } else if (req$REQUEST_METHOD == "POST") { values <- jsonlite::fromJSON(req$postBody) } # Check that the required fields are in the POST'ed data required = c('sender','recipient', 'amount') if (!all(required %in% names(values))) { return ('Missing Values - sender, recipient and amount are required') } index <- blockchain$addTransaction(values$sender, values$recipient, values$amount) list('message' = paste('Transaction will be added to Block', index)) } #* @serializer custom_json #* @get /mine function(req) { # We run the proof of work algorithm to get the next nonce lastBlock <- blockchain$lastBlock() lastNonce <- lastBlock$block$nonce nonce <- blockchain$proofOfWork(lastNonce) # We must receive a reward for finding the Nonce. # The sender is "0" to signify that this node has mined a new coin. blockchain$addTransaction(sender="0", recipient = node_identifier, amount=1) # Forge the new block by adding it to the chain previousHash <- blockchain$hashBlock(lastBlock) block <- blockchain$nextBlock(nonce, previousHash) list('message'='New block forged', 'index'= block$block$index, 'transactions'= block$block$transactions, 'nonce'=block$block$nonce, 'previousHash'=block$block$previousHash) # list('message'='New block forged', c('index'= block$block$index, 'transactions'= block$block$transactions, 'nonce'=block$block$nonce,'previousHash'=block$block$previousHash)) } #* @serializer custom_json #* @post /nodes/register #* @get /nodes/register function(req, nodes) { # req_json <- '{"sender": "my address", "recipient": "someone else address", "amount": 5}' if (req$REQUEST_METHOD == "GET") { } else if (req$REQUEST_METHOD == "POST") { values <- jsonlite::fromJSON(req$postBody) nodes <- values$nodes } cat(paste0(nodes, "\n")) if (is.null(nodes)) { return("Error: the list of nodes is not valid") } blockchain$registerNode(nodes) } #* @serializer custom_json #* @get /nodes/resolve function (req) { replaced = blockchain$handleConflicts() if (replaced) { list('message'='Replaced', 'chain' = blockchain$chain) } else { list('message'='Authoritative block chain - not replaceable ', 'chain'=blockchain$chain) } } #* Log some information about the incoming request #* @filter logger function(req){ cat(as.character(Sys.time()), "-", req$REQUEST_METHOD, req$PATH_INFO, "-", req$HTTP_USER_AGENT, "@", req$REMOTE_ADDR, "\n") plumber::forward() }

149d0405865f49b98f9e558ee0b27bab3efd2b60

94aed2117dfdb2227eea8b019fed82d5b6193e4e

/TextMine.R

061168a7ed7e0a19ea513d100e952e0fe985cb56

[]

no_license

elliott828/Working

c2199454360508eac7f7a21c302abf71908a7530

e5caeba56871c7c1c4a890e2b08737452a5cb013

refs/heads/master

2020-12-07T00:46:05.930959

2015-06-04T08:30:30

36,857,779

0

null

2015-06-04T08:24:17

2015-06-04T08:24:16

R

UTF-8

R

false

5,430

r

TextMine.R

#-------------------------------- req.pcg <- function(pcg){ # packages to be installed tbinst <- pcg[(!(pcg %in% installed.packages()[, "Package"]))| (pcg %in% old.packages()[, "Package"])] if (sum(tbinst %in% c("tmcn", "Rwordseg", "Rweibo"))>0){ cntm <- tbinst[tbinst %in% c("tmcn", "Rwordseg", "Rweibo")] install.packages(cntm, repos = "http://R-Forge.R-project.org", type = "source") }else if(sum(tbinst == "Rgraphviz")>0){ source("http://bioconductor.org/biocLite.R") biocLite("Rgraphviz") }else if (length(tbinst)){ install.packages(tbinst, dependencies = T) } sapply(pcg, require, warn.conflicts = FALSE, character.only = TRUE, quietly = TRUE) } all.pcg <- c("tm", "SnowballC", "qdap", "qdapDictionaries", "dplyr", "RColorBrewer", "ggplot2", "scales", "wordcloud", "igraph", "Rweibo", "Rwordseg", "RWeka", "ggdendro") req.pcg(all.pcg) # ERROR: compilation failed for package 'tmcn' # Warning in install.packages : package 'tmcn' is not available (for R version 3.2.0) #-------------------------------- df <- read.csv("FO_Increased.csv") df <- read.csv("FO_Dropped.csv") df <- read.csv("FO_Same.csv") df <- read.csv("FO_Total.csv") df <- read.csv("Australia.csv") i <- 1 i <- 2 sub_cont <- Corpus(VectorSource(df[complete.cases(df[, i]), i])) sub_cont <- sub_cont %>% tm_map(content_transformer(tolower)) %>% tm_map(removeWords, stopwords("english")[c(-81:-98, -160, -165:-167)]) # [160] "more" # others: not change <- content_transformer(function(x, from, to) gsub(from, to, x)) for(j in c(81:98, 166)) { sub_cont <- tm_map(sub_cont, change, stopwords("english")[j], "not") } sub_cont <- sub_cont %>% tm_map(change, "blue oval", "blueoval") %>% tm_map(change, "loyal followers", "loyalfollower") sub_cont <- sub_cont %>% tm_map(removePunctuation) %>% tm_map(stripWhitespace) %>% tm_map(stemDocument) %>% tm_map(removeNumbers) # change words into original form mat <- matrix(c(c("releas", "purchas", "websit", "territori", "specif", "peopl", "futur", "decid", "brochur", "pictur"), c("release", "purchase", "website", "territory", "specify", "people", "future", "decide", "brochure", "picture")), nrow = 2, byrow = TRUE) for(k in 1:ncol(mat)){ sub_cont <- tm_map(sub_cont, change, mat[1, k], mat[2, k]) } sub_cont <- tm_map(sub_cont, removeWords, c("even", "still", "just", "will", "yet", "can", "much", "car", "ford", "also", "one", "vehicl")) dtm <- DocumentTermMatrix(sub_cont) tdm <- TermDocumentMatrix(sub_cont) dim(dtm) # inspect(dtm[1:5, 1:5]) freq <- colSums(as.matrix(dtm)) # length(freq) ord <- order(freq, decreasing = TRUE) # table(freq) freq <- freq[ord] wf <- data.frame(word=names(freq), freq=freq) head(wf) # Histogram of Frequency subset(wf, freq > 2) %>% ggplot(aes(word, freq)) + geom_bar(stat="identity") + theme(axis.text.x=element_text(angle=45, hjust=1)) comp <- function(words, mat){ for(i in 1:length(words)){ if(any(words[i] == mat[1, ])){ words[i] <- mat[2, ][which(words[i] == mat[1, ])] } } return(words) } wf$word <- comp(wf$word, mat) wf$word <- factor(wf$word, levels = wf[order(wf[,2], decreasing = FALSE), 1], ordered=T) ggplot(subset(wf, freq > 25),aes(x= word, freq)) + geom_bar(stat = "identity") + coord_flip() + ggtitle("Word Frequency > 25") + ylab("Frequency") + xlab("Word") # png("Dendrogram_db.png", width=12, height=8, units="in", res=300) set.seed(123) wordcloud(names(freq), freq, min.freq = 4, scale = c(5, .8), random.order = FALSE, colors=brewer.pal(6, "Dark2")) # Association plot Attrs <- list(node = list(shape = "ellipse", fixedsize = FALSE, style = "invis", fontcolor = "white", fillcolor = "red"), edge = list(dir = "both", color = "darkblue", weight = 1.2)) plot(dtm, terms = findFreqTerms(dtm, lowfreq = 4), corThreshold = 0.2, attrs = Attrs, weighting = TRUE) # Cluster Dendrogram: # DistMat <- dist(scale(as.matrix(tdm))) DistMat <- dist(scale(as.matrix(tdm)[order(rowSums(as.matrix(tdm)), decreasing = TRUE), ][1:35, ])) fit <- hclust(DistMat) # method = "ward.D", "ward.D2", "single", "complete", "average"... plot(fit) ggdendrogram(fit) # cut tree into k clusters rect.hclust(fit, k = 6) # rect.hclust(tree, k = NULL, which = NULL, x = NULL, h = NULL, # border = 2, cluster = NULL) # kmeans # findAssocs(dtm, "not", corlimit = 0.3) #------------------------------------ df <- read.csv("FO_Compare.csv", head = FALSE) df <- read.csv("FO_Compare2.csv", head = FALSE) df <- sapply(df, as.character) df <- df[, -1] sub_cont <- Corpus(DataframeSource(df)) tdm6 <- as.matrix(tdm) tdm6 <- tdm6[!rownames(tdm6) %in% c("new", "vehicl"), ] colnames(tdm6) <- c("Drop1", "Drop2", "Increase1", "Increase2", "Same1", "Same2") colnames(tdm6) <- c("Drop2", "Increase2", "Same2") comparison.cloud(tdm6, random.order = F, max.words = Inf, title.size = 1.5) commonality.cloud(tdm6, random.order=FALSE, colors = brewer.pal(8, "Dark2"), title.size=1.5)

5d2b30df4efb30376a04def5834f064b0187c8bb

6e7af9b27cf18bb4633ad9d0b63a7e8ed9a887fb

/man/ranges_df-pHSensor-method.Rd

8b049209ed639a484ebec5d26df2b9c9b25d13bd

[ "MIT" ]

permissive

ApfeldLab/SensorOverlord

0fc62dd3c11b702cd477d0692085ea7be46911a7

2fbe7e0d0963561241d5c1e78dd131211e1b31a0

refs/heads/master

2022-12-27T15:20:27.343783

2020-10-13T23:28:48

176,821,341

2

0

null

2020-06-14T15:37:09

2019-03-20T21:40:17

R

UTF-8

R

false

true

1,588

rd

ranges_df-pHSensor-method.Rd

% Generated by roxygen2: do not edit by hand % Please edit documentation in R/Sensor_Methods.R \name{ranges_df,pHSensor-method} \alias{ranges_df,pHSensor-method} \title{Finds the ranges df of this pH sensor at given inaccuracies} \usage{ \S4method{ranges_df}{pHSensor}( object, inaccuracies = c(0.02), pHmin = 1, pHmax = 14, by = 0.001, name = "Sensor", thresholds = c(0.01, 0.05, 0.1, 0.15, 0.2) ) } \arguments{ \item{object}{A pHSensor object} \item{inaccuracies}{(optional, default: c(0.02)) A vector of inaccuracies (e.g. 0.02 for 2\% error), always relative} \item{pHmin}{(optional, default: 1) The minimum pH for which to record error} \item{pHmax}{(optional, default: 14) The maximum pH for which to record error} \item{by}{(optional, default: 0.001) The granularity of the error table--e.g., by = 0.01 would record 7 and 7.01, etc.} \item{name}{(optional, default: "Sensor") A name for this sensor} \item{thresholds}{A vector of error thresholds (e.g. c(0.5, 1) for 0.5 and 1)} } \value{ A dataframe of suited ranges with these columns: 'Sensor_Name': the name of the sensor 'Minimum': the minimum pH measurable at the given inaccuracy 'Maximum': the maximum pH measurable at the given inaccuracy 'Inaccuracy': the inaccuracy associated with this row (relative) 'error_thresh': the error threshold associated with this row } \description{ Adding this method on 31 May 2020, hoping this style will depreciate getErrorTable in the future. } \examples{ my_sensor <- new("pHSensor", new("Sensor", Rmin = 1, Rmax = 5, delta = 0.2), pKa = 7) ranges_df(my_sensor) }

0689eefd9f38ef5b43a36025c1440dabbd1d0f29

2d1f4a315a1b6fda16341144a95e62da2898c9d9

/workflow/daily_update.R

b6476aa746615e8adb5020a64c3d57a451329585

[ "LicenseRef-scancode-public-domain", "CC0-1.0" ]

permissive

sonar98/covid-19

913893c15a5ff1f765b7c55bcd16b9ce185849a3

cd56342803266f83850be4170d54ad520f0966bf

refs/heads/master

2023-01-10T00:10:00.817896

2020-11-02T13:59:58

null

0

null

UTF-8

R

false

9,458

r

daily_update.R

pull(repo) # Generate Banner source("workflow/generate_banner.R") # Parse RIVM, NICE and corrections data source("workflow/parse_nice-data.R") source("workflow/parse_rivm-data.R") source("workflow/parse_lcps-data.R") source("workflow/parse_municipalities.R") source("workflow/parse_corrections.R") Sys.setlocale("LC_TIME", "nl_NL") ## Merge RIVM, NICE and corrections data rivm.by_day <- read.csv("data/rivm_by_day.csv") nice.by_day <- read.csv("data-nice/nice-today.csv") lcps.by_day <- read.csv("data/lcps_by_day.csv") corr.by_day <- read.csv("corrections/corrections_perday.csv") daily_datalist <- list(rivm.by_day,nice.by_day,lcps.by_day,corr.by_day) all.data <- Reduce( function(x, y, ...) merge(x, y, by="date",all.x = TRUE, ...), daily_datalist ) all.data$date <- as.Date(all.data$date) all.data <- all.data[order(all.data$date),] write.csv(all.data, file = "data/all_data.csv",row.names = F) source("plot_scripts/daily_plots.R") #source("plot_scripts/daily_maps_plots.R") all.data <- read.csv("data/all_data.csv") nice_by_day <- read.csv("data/nice_by_day.csv") ## Corrections or not? text.deaths.corrections <- paste0(ifelse(last(all.data$net.deaths)>=0," (+"," (-"),abs(last(all.data$net.deaths))," ivm ",last(all.data$corrections.deaths)," corr.)") # get tokens source("workflow/twitter/token_mzelst.R") source("workflow/twitter/token_edwinveldhuizen.R") ## Build tweets tweet.main <- paste0("#COVID19NL statistieken t.o.v. gisteren: Positief getest: ",last(all.data$new.infection)," Totaal: ",last(all.data$cases)," (+",last(all.data$net.infection)," ivm ",last(all.data$corrections.cases)," corr.) Opgenomen*: ",last(all.data$Kliniek_Nieuwe_Opnames_COVID)," Huidig*: ",last(all.data$Kliniek_Bedden)," Opgenomen op IC*: ",last(all.data$IC_Nieuwe_Opnames_COVID)," Huidig*: ",last(all.data$IC_Bedden_COVID)," * LCPS cijfers - www.lcps.nu Overleden: ",last(all.data$new.deaths)," Totaal: ",last(all.data$deaths),"") posted_tweet <- post_tweet ( tweet.main, token = token.mzelst, media = (paste0("banners/",Sys.Date(),".png")) ) ## Post tweet posted_tweet <- fromJSON(rawToChar(posted_tweet$content)) tweet.main.id <- posted_tweet$id_str tweet.last_id <- tweet.main.id # Retweet for @edwinveldhuizen post_tweet (token = token.edwinveldhuizen, retweet_id = tweet.main.id) # Tweet for hospital numbers - Data NICE #### temp = tail(list.files(path = "data-nice/data-nice-json/",pattern="*.csv", full.names = T),2) myfiles = lapply(temp, read.csv) dat.today <- as.data.frame(myfiles[2]) dat.yesterday <- as.data.frame(myfiles[1]) Verpleeg_Opname_Bevestigd <- sum(dat.today$Hospital_Intake_Proven) - sum(dat.yesterday$Hospital_Intake_Proven) Verpleeg_Opname_Verdacht <- sum(dat.today$Hospital_Intake_Suspected) - sum(dat.yesterday$Hospital_Intake_Suspected) IC_Opname_Bevestigd <- sum(dat.today$IC_Intake_Proven) - sum(dat.yesterday$IC_Intake_Proven) IC_Opname_Verdacht <- sum(dat.today$IC_Intake_Suspected) - sum(dat.yesterday$IC_Intake_Suspected) Verpleeg_Huidig_Toename <- last(dat.today$Hospital_Currently) - last(dat.yesterday$Hospital_Currently) IC_Huidig_Toename <- last(dat.today$IC_Current) - last(dat.yesterday$IC_Current) hospital.cumulative <- rjson::fromJSON(file = "https://www.stichting-nice.nl/covid-19/public/zkh/intake-cumulative/",simplify = TRUE) %>% map(as.data.table) %>% rbindlist(fill = TRUE) sign.hosp.nice <- paste0(ifelse(Verpleeg_Huidig_Toename>=0," (+"," (")) sign.ic.nice <- paste0(ifelse(IC_Huidig_Toename>=0," (+"," (")) tweet.nice <- paste0("#COVID19NL statistieken t.o.v. gisteren (data NICE): Patiënten verpleegafdeling Bevestigd: ",Verpleeg_Opname_Bevestigd," Verdacht: ",Verpleeg_Opname_Verdacht," Huidig: ",last(dat.today$Hospital_Currently),sign.hosp.nice,Verpleeg_Huidig_Toename,") Totaal: ",last(hospital.cumulative$value)," Patiënten IC Bevestigd: ",IC_Opname_Bevestigd," Verdacht: ",IC_Opname_Verdacht," Huidig: ",last(dat.today$IC_Current),sign.ic.nice,IC_Huidig_Toename,") Totaal: ",last(dat.today$IC_Cumulative)) # Tweet for report #### posted_tweet <- post_tweet ( tweet.nice, token = token.mzelst, media = c("plots/positieve_tests_per_dag.png", "plots/overview_aanwezig_zkh.png", "plots/overview_opnames_zkh.png" ), in_reply_to_status_id = tweet.last_id, auto_populate_reply_metadata = TRUE ) posted_tweet <- fromJSON(rawToChar(posted_tweet$content)) tweet.last_id <- posted_tweet$id_str ######## # report ######## tweet.report = "Ik heb een start gemaakt met een dagelijks epidemiologisch rapport (work in progress). Hierin vindt u kaarten en tabellen met gegevens per leeftijdsgroep, provincie, en GGD: https://github.com/mzelst/covid-19/raw/master/reports/daily_report.pdf" posted_tweet <- post_tweet ( tweet.report, token = token.mzelst, in_reply_to_status_id = tweet.last_id, auto_populate_reply_metadata = TRUE ) posted_tweet <- fromJSON(rawToChar(posted_tweet$content)) tweet.last_id <- posted_tweet$id_str ##### Generate municipality images source("workflow/generate_municipality_images.R") ##### ######## # Municipality tweet - cases ######## tweet.municipality.date <- Sys.Date() %>% format('%d %b') %>% str_to_title() %>% str_replace( '^0', '') tweet.municipality.colors <- read.csv("data/municipality-totals-color.csv", fileEncoding = "UTF-8") tweet.municipality.cases <- "Geconstateerde besmettingen per gemeente %s %s %d / 355 gemeentes %s %d / 355 gemeentes %s %d / 355 gemeentes Zie de eerste afbeelding voor een uitgebreide legenda [%s] %s" tweet.municipality.cases <- sprintf(tweet.municipality.cases, intToUtf8(0x1F447), intToUtf8(0x1F6D1), tweet.municipality.colors$d0[[4]], intToUtf8(0x1F7E3), tweet.municipality.colors$d0[[5]], intToUtf8(0x26A1), tweet.municipality.colors$d0[[6]], tweet.municipality.date, 'https://raw.githack.com/mzelst/covid-19/master/workflow/daily_municipality.html' ) Encoding(tweet.municipality.cases) <- "UTF-8" posted_tweet <- post_tweet ( tweet.municipality.cases, token = token.edwinveldhuizen, media = c("plots/list-cases-head.png", "plots/list-cases-all-part1.png", "plots/list-cases-all-part2.png", "plots/list-cases-all-part3.png"), in_reply_to_status_id = tweet.main.id, auto_populate_reply_metadata = TRUE ) posted_tweet <- fromJSON(rawToChar(posted_tweet$content)) tweet.last_id <- posted_tweet$id_str rm(tweet.municipality.cases, tweet.municipality.colors, posted_tweet) post_tweet ( token.mzelst, retweet_id = tweet.last_id) ######## # Municipality tweet - hospital admissions ######## tweet.municipality.hosp <- "Positief geteste patiënten per gemeente die zijn opgenomen met specifiek COVID-19 als reden v. opname [%s]" tweet.municipality.hosp <- sprintf(tweet.municipality.hosp, tweet.municipality.date ) posted_tweet <- post_tweet ( tweet.municipality.hosp, token = token.edwinveldhuizen, media = c("plots/list-hosp-head.png", "plots/list-hosp-all-part1.png", "plots/list-hosp-all-part2.png", "plots/list-hosp-all-part3.png"), in_reply_to_status_id = tweet.last_id, auto_populate_reply_metadata = TRUE ) posted_tweet <- fromJSON(rawToChar(posted_tweet$content)) tweet.last_id <- posted_tweet$id_str rm(tweet.municipality.hosp, posted_tweet) ######## # Municipality tweet - deaths ######## tweet.municipality.deaths <- "Patiënten per gemeente die positief getest zijn op COVID-19 en helaas zijn overleden [%s] Onze condoleance en veel sterkte aan alle nabestaanden. %s" tweet.municipality.deaths <- sprintf(tweet.municipality.deaths, tweet.municipality.date, intToUtf8(0x1F339) ) Encoding(tweet.municipality.deaths) <- "UTF-8" posted_tweet <- post_tweet ( tweet.municipality.deaths, token = token.edwinveldhuizen, media = c("plots/list-deaths-head.png", "plots/list-deaths-all-part1.png", "plots/list-deaths-all-part2.png", "plots/list-deaths-all-part3.png"), in_reply_to_status_id = tweet.last_id, auto_populate_reply_metadata = TRUE ) posted_tweet <- fromJSON(rawToChar(posted_tweet$content)) tweet.last_id <- posted_tweet$id_str rm(tweet.municipality.deaths, tweet.municipality.date, posted_tweet) ##### Download case file rivm.data <- utils::read.csv("https://data.rivm.nl/covid-19/COVID-19_casus_landelijk.csv", sep =";") ## Read in data with all cases until today filename <- paste0("data-rivm/casus-datasets/COVID-19_casus_landelijk_",Sys.Date(),".csv") write.csv(rivm.data, file=filename,row.names = F) ## Write file with all cases until today ##### Sys.setenv(RSTUDIO_PANDOC="C:/Program Files/RStudio/bin/pandoc"); rmarkdown::render('reports/daily_report.Rmd') ## Render daily report file.copy(from = list.files('reports', pattern="*.pdf",full.names = TRUE), to = paste0("reports/daily_reports/Epidemiologische situatie COVID-19 in Nederland - ", format((Sys.Date()),'%d')," ",format((Sys.Date()),'%B'),".pdf")) ## Save daily file in archive git.credentials <- read_lines("git_auth.txt") git.auth <- cred_user_pass(git.credentials[1],git.credentials[2]) ## Push to git repo <- init() add(repo, path = "*") commit(repo, all = T, paste0("Daily (automated) update RIVM and NICE data ",Sys.Date())) push(repo, credentials = git.auth) ## Workflows for databases source("workflow/dashboards/cases_ggd_agegroups.R") source("workflow/dashboards/date_statistics_mutations.R") source("workflow/parse_age-data.R") source("workflow/dashboards/heatmap-age-week.R") source("workflow/dashboards/rivm-date-corrections.R")

b56b6cc6d76b6ec989ada4e3dbde3c44e0f2cb89

2a7e77565c33e6b5d92ce6702b4a5fd96f80d7d0

/fuzzedpackages/RPresto/R/utility_functions.R

8ec4fe75864e09691d51c27e57e53c942f06b0ee

[ "BSD-3-Clause" ]

permissive

akhikolla/testpackages

62ccaeed866e2194652b65e7360987b3b20df7e7

01259c3543febc89955ea5b79f3a08d3afe57e95

refs/heads/master

2023-02-18T03:50:28.288006

2021-01-18T13:23:32

329,981,898

7

1

null

UTF-8

R

false

1,266

r

utility_functions.R

# Copyright (c) Facebook, Inc. and its affiliates. # All rights reserved. # # This source code is licensed under the BSD-style license found in the # LICENSE file in the root directory of this source tree. check.status.code <- function(response) { status <- httr::status_code(response) if (status >= 400 && status < 500) { text.content <- httr::content(response, as = "text", encoding='UTF-8') if (is.null(text.content) || !nzchar(text.content)) { httr::stop_for_status(status) } stop('Received error response (HTTP ', status, '): ', text.content) } } response.to.content <- function (response) { text.content <- httr::content(response, as = "text", encoding='UTF-8') return(jsonlite::fromJSON(text.content, simplifyVector = FALSE)) } wait <- function () { # sleep 50 - 100 ms Sys.sleep(stats::runif(n = 1, min = 50, max = 100) / 1000) } get.state <- function (content) { if (is.null(content[['stats']]) || is.null(content[['stats']][['state']]) ) { stop('No state information in content') } return(content[['stats']][['state']]) } stop.with.error.message <- function (content) { query.id <- content[['id']] message <- content[['error']][['message']] stop("Query ", query.id, " failed: ", message) }

4125bacb354cf965e09500193d7b47218a7e8024

76a4c4703c9c23f43e5f9ffcf6e800c36f5c92df

/IR_DS_filtering_workflow.R

9f6735461bc8bbe2280e6a7208c9eb7e0810711f

[]

no_license

danjst/PDAC_2020

f00bd18dbef619e1f4a08f9a9b771a3d3b7efd20

4ebaf0a1b2231c12b34913ec8cbf113a9e4381f6

refs/heads/master

2023-04-27T10:44:18.702881

2020-09-23T11:11:49

295,741,952

1

0

null

UTF-8

R

false

2,237

r

IR_DS_filtering_workflow.R

# first filter out events with more than 10% NA pdac_less_10_percent_NA <- all_pdac_events[rowSums(is.na(all_pdac_events)) < 0.1 * ncol(all_pdac_events), ] ###Next filtering for less than 0.05 abs NA difference between 2 clusters #make pdac_cluster1 and pdac_cluster2 first pdac_10na_2<-pdac_less_10_percent_NA[,pdac_cluster2] pdac_10na_1<-pdac_less_10_percent_NA[,pdac_cluster1] pdac_1_NAs <- apply(pdac_10na_1, 1, function(x) sum(is.na(x))) pdac_1_NAs_percent<-pdac_2_NAs/40 pdac_2_NAs <- apply(pdac_10na_2, 1, function(x) sum(is.na(x))) pdac_2_NAs_percent<-pdac_2_NAs/36 pdac_1_minus_2_percent<-pdac_1_NAs_percent-pdac_2_NAs_percent pdac_1_minus_2_percent<-abs(pdac_1_minus_2_percent) pdac_over_0.05<-c() for (i in 1:nrow(pdac_less_10_percent_NA)){ if (pdac_1_minus_2_percent[[i]]>0.05){pdac_over_0.05<-c(pdac_over_0.05,names(pdac_1_minus_2_percent[i]))} } pdac_less_10_percent_NA_also_0.05 <- pdac_less_10_percent_NA[!row.names(pdac_less_10_percent_NA)%in%pdac_over_0.05,] #join with DS list #re-adjust p-values here #check for significance here, filter padj<0.05 ###doing the 0.1 mean psi difference pdac_10na_0.05_1<-pdac_less_10_percent_NA_also_0.05[,pdac_cluster1] pdac_10na_0.05_2<-pdac_less_10_percent_NA_also_0.05[,pdac_cluster2] pdac_rowmeans_1<-rowMeans(pdac_10na_0.05_1,na.rm = T) pdac_rowmeans_2<-rowMeans(pdac_10na_0.05_2,na.rm = T) pdac_rowmeans_difference<-pdac_rowmeans_1-pdac_rowmeans_2 pdac_rowmeans_difference_greater_0.1<-pdac_rowmeans_difference[abs(pdac_rowmeans_difference)>0.1] pdac_final_event_rownames<-names(pdac_rowmeans_difference_greater_0.1) pdac_final_ds_events <- pdac_less_10_percent_NA_also_0.05[row.names(pdac_less_10_percent_NA_also_0.05)%in%pdac_final_event_rownames,] pdac_final_event_meanpsi_values<-as.numeric(paste(unlist(pdac_rowmeans_difference_greater_0.1))) pdac_final_ds_events<-cbind(pdac_final_ds_events,pdac_final_event_meanpsi_values) pdac_final_ds_events_x<-cbind(pdac_final_ds_events[1:6],pdac_final_event_meanpsi_values) pdac_final_ds_events_x$gene_no_version<-substr(pdac_final_ds_events_x$gene_old_ensembl,start=1,stop=15) pdac_final_ds_events_x<-join_all(list(pdac_final_ds_events_x,gencodev22_bed_2020), by = c('gene_no_version'), type = "left", match = "all")

eff61bacdf25705d855f0d8080b4cfac3e5f7a3f

ffdea92d4315e4363dd4ae673a1a6adf82a761b5

/data/genthat_extracted_code/syuzhet/examples/get_sentences.Rd.R

9f9c80ec3869e03a2b5ebfda657c357b540a322d

[]

no_license

surayaaramli/typeRrh

d257ac8905c49123f4ccd4e377ee3dfc84d1636c

66e6996f31961bc8b9aafe1a6a6098327b66bf71

refs/heads/master

2023-05-05T04:05:31.617869

2019-04-25T22:10:06

null

0

null

UTF-8

R

false

500

r

get_sentences.Rd.R

library(syuzhet) ### Name: get_sentences ### Title: Sentence Tokenization ### Aliases: get_sentences ### ** Examples (x <- c(paste0( "Mr. Brown comes! He says hello. i give him coffee. i will ", "go at 5 p. m. eastern time. Or somewhere in between!go there" ), paste0( "Marvin K. Mooney Will You Please Go Now!", "The time has come.", "The time has come. The time is now. Just go. Go. GO!", "I don't care how." ))) get_sentences(x) get_sentences(x, as_vector = FALSE)

9e0c85ef630f54dbc7065fb389a81c43d9929aaf

1811c5e994ab0d62884a02639b425f1da7b7bde4

/R/graphics/ggplot2/ggplot2_tut.R

5207a7101449897f1bd46e8b934aee8da806cea1

[]

no_license

aufrank/tutorials

62aa2fb8d3e5eb061fb6693e58597252fc970dc7

12b06fdfffee7bde6127dc437cdaa0b58a49bcdd

refs/heads/master

2021-01-19T03:13:56.736438

2009-02-25T20:17:43

137,469

1

0

null

UTF-8

R

false

5,224

r

ggplot2_tut.R

library(ggplot2) library(plyr) ## get an interactive window quartz() ## x11() ## lexical decision data data(english, package="languageR") ## default scatterplot qplot(x = WrittenFrequency, y = RTlexdec, data = english, main = "Frequency and Reaction Time") ## transform the dependent variable qplot(x = WrittenFrequency, y = exp(RTlexdec), data = english) ## color for groups qplot(x = WrittenFrequency, y = RTlexdec, data = english, colour=AgeSubject) ## color for continuous variable qplot(WrittenFrequency, RTlexdec, data=english, colour=WrittenSpokenFrequencyRatio) ## shape for groups qplot(x = WrittenFrequency, y = RTlexdec, data = english, shape=AgeSubject) ## colour and shape for groups qplot(x = WrittenFrequency, y = RTlexdec, data = english, colour=AgeSubject, shape=WordCategory) ## colour, shape, and size qplot(x = WrittenFrequency, y = RTlexdec, data = english, colour=AgeSubject, shape=WordCategory, size=FamilySize) ## panels ## one conditioning variable qplot(x = WrittenFrequency, y = RTlexdec, data = english, colour=AgeSubject, shape=Frication, facets = . ~ WordCategory) ## panels with totals shown qplot(x = WrittenFrequency, y = RTlexdec, data = english, colour=AgeSubject, shape=AgeSubject, facets = . ~ WordCategory, margins=TRUE) ## two conditioning variables qplot(x = WrittenFrequency, y = RTlexdec, data = english, colour=AgeSubject, shape=Frication, facets = CV ~ WordCategory) ## panels with totals shown qplot(x = WrittenFrequency, y = RTlexdec, data = english, colour=AgeSubject, shape=Frication, facets = CV ~ WordCategory, margins=TRUE) ## changing type of plot requires changing the "geom" qplot(x = WrittenFrequency, data = english, geom = "density", fill=rev(AgeSubject)) ## histograms don't have a sensible default for bin size qplot(x = RTlexdec, data = english, geom = "histogram", fill=AgeSubject) ## stacked bar charts based on counts are the default qplot(x=WordCategory, data = english, geom = "bar", fill=Voice) ## use "fill" positioning to convert them to proportions qplot(x=WordCategory, data = english, geom = "bar", fill=Voice, position="fill") ## use "dodge" positioning to unstack them qplot(x=WordCategory, data = english, geom = "bar", fill=Voice, position="dodge") ## use stat="identity" if you've already aggregated your data and want ## to plot the results d <- with(english, aggregate(RTlexdec, by=list(Age=AgeSubject, Category=WordCategory), FUN=mean)) qplot(x=Age, y=x, data=d, fill=Category, geom="bar", stat="identity", position="dodge") ## plots from scratch, without qplot ## boxplot bxp <- ggplot(data = english, aes(x=AgeSubject, y=RTlexdec)) bxp <- bxp + stat_boxplot(aes(fill=WordCategory)) (bxp) ## barplot of means bpm <- ggplot(data=english, aes(x=AgeSubject, y=RTlexdec, fill=WordCategory)) bpm <- bpm + stat_summary(fun.y=mean, geom="bar", pos="dodge") (bpm) ## change the scale on the y axis to zoom in on the relevant region of ## the data ## NB: If you don't want to show the entire range starting ## from 0, you should consider using a dot plot instead of a bar plot bpm <- bpm + scale_y_continuous(limits=c(5,7)) (bpm) ## bar plot with automatically calculated normal error bars bpe <- ggplot(data=english, aes(x=AgeSubject, y=RTlexdec, fill=WordCategory)) bpe <- bpe + stat_summary(fun.y="mean", geom="bar", pos="dodge") bpe <- bpe + stat_summary(fun.data="mean_cl_normal", geom="errorbar", width=0.2) (bpe) ## dot plot of means dpm <- ggplot(data=english, aes(x=AgeSubject, y=RTlexdec, colour=WordCategory)) dpm <- dpm + stat_summary(fun.y="mean", geom="point") (dpm) ## dot plot of means with error bars of 2*sd dpe <- ggplot(data=english, aes(x=AgeSubject, y=RTlexdec, colour=WordCategory)) dpe <- dpe + stat_summary(fun.data="mean_sdl", geom="pointrange") (dpe) ## horizontal dotplot, often used to display regression parameters dpe <- dpe + coord_flip() (dpe) ## xyplot with grouping by color xy.tr <- ggplot(data=english, aes(x=WrittenFrequency, y=RTlexdec, colour=AgeSubject)) xy.tr <- xy.tr + geom_point() ## now add a specific color scale where we define an alpha level xy.tr <- xy.tr + scale_colour_hue(alpha=1/3) (xy.tr) ## add a smoother to our scatterplot xy.tr <- xy.tr + stat_smooth() (xy.tr) ## or we can specify that the smooth be based on a linear model qplot(WrittenFrequency, RTlexdec, data=english, colour=AgeSubject) + scale_colour_hue(alpha=1/3) + stat_smooth(method=lm) ## and we can add a robust linear smooth if we want library(MASS) last_plot() + stat_smooth(method=rlm) ## apply the black and white theme to our plot last_plot() + theme_bw() ## and change font sizes. See page 123 and 124 of the ggplot2 book. last_plot() + opts(title = "Frequency Effects", plot.title=theme_text(size=24)) ## finally, let's make a hexbinplot hbp <- ggplot(english, aes(x=WrittenFrequency, y=RTlexdec)) + stat_binhex() + stat_smooth(method=rlm, colour=I("orange"), size=1.5) + opts(panel.background=theme_blank()) hbp

1b961282611fcf9061b095bfa0dfc3402805e1bf

a176626eb55b6525d5a41e2079537f2ef51d4dc7

/Uni/Projects/code/P050.Haifa.Mon/cn001_indiv_V4.r

9999ebb42fd519d4ed42f2e77b4ccbe600a4797f

[]

no_license

zeltak/org

82d696b30c7013e95262ad55f839998d0280b72b

d279a80198a1dbf7758c9dd56339e8a5b5555ff2

refs/heads/master

2021-01-21T04:27:34.752197

2016-04-16T04:27:57

18,008,592

0

1

null

UTF-8

R

false

3,431

r

cn001_indiv_V4.r

############### #LIBS ############### library(lme4) library(reshape) library(foreign) library(plyr) library(dplyr) library(data.table) library(reshape2) library(Hmisc) library(mgcv) library(gdata) library(readr) library(stargazer) library(splines) library(sjPlot) #ind <-read.csv("/home/zeltak/ZH_tmp/dat/indiv1415av.csv") ind <-read.dbf("/home/zeltak/ZH_tmp/dat/tipot_all_SPSS_6.02.16.dbf") names(ind) ind$month = as.numeric(format(ind$BIRTH_DATE,"%m")) #rename setnames(ind,old=c("WEIGHT1_VA", "HEAD1_VALU","PREGNANCYW"),new=c("birthw", "headc","ges")) # #subset data #ind<-ind[,c("Head1_Valu","X","Y","Gender","Weight1_Va","Mother_Nat","PregnancyW","month","TotalSibli","Education_","ApgarOneMi","ApgarFiveM","POPULATION","HOUSEHOLDS","AVERAGE_HH","DENSITY","OWNERSHIP","RENTALS","BAGRUT","BA","INCOME","N_AIRPORT","N_BAZAN","N_POWERSTA","N_OIL_L","N_OIL_S","N_ROAD","nox","day","Postal","Mother_Bir","pm25","so2","nox2014","Elevation","People_est","Pop_arnona"),with=FALSE] ind<-select (ind,birthw,headc,ges,X,Y,SEX,POPULATION,HOUSEHOLDS,AVERAGE_HH) #save for GIS # #clean data # #delete bad data # ind<-filter(ind,ges >= 20 ) # ind<-filter(ind,ges <= 44 ) # #clean all crap data # ind[ind == -999] <- NA # ind[ind == -9.99] <- NA # ind<-filter(ind,OWNERSHIP >= 0 ) # ind<-filter(ind,AVERAGE_HH >= 0 ) # ind<-filter(ind,BA >= 0 ) # ind<-filter(ind,birthw >= 2 ) # summary(ind$Mother_Nat) # ###recoding # #finer race # ind$mrn.n<-0 #for jews # ind<- ind[Mother_Nat == "יהודי" , mrn.n := 1] #jewish # #gender 01 # ind$sex<-1 #for male # ind<- ind[Gender == "נקבה" , sex := 0] #female # #only nox liner, dist linear, #nox was most significance # ind$N_BAZAN<-ind$N_BAZAN/1000 # ind$N_AIRPORT<-ind$N_AIRPORT/1000 # ind$N_POWERSTA<-ind$N_POWERSTA/1000 # ind$N_OIL_L<-ind$N_OIL_L/1000 # ind$N_OIL_S<-ind$N_OIL_S/1000 # ind$N_ROAD<-ind$N_ROAD/1000 # ind$headcWT<-ind$headc/ind$birthw # ind$birthwHC<-ind$birthw/ind$headc ind.pre<-filter(ind, ges < 37) ind.full<-filter(ind, ges >= 37) #write.dbf(ind,"/home/zeltak/ZH_tmp/dat/indiv1415_clean.dbf") #normal ##Headc regression nox m1.formula <- as.formula(headc.bc ~birthw+ges+as.factor(SEX)+as.factor(month)+MOTHER_COU+LNDIST_ROA+LNDIST_OIL) h1 <- lm(m1.formula,data=ind) summary(h1) ##Headc regression nox pre h1.pre <- lm(m1.formula,data=ind.pre) summary(h1.pre) ##Headc regression nox full h1.full <- lm(m1.formula,data=ind.full) summary(h1.full) #html stargazer(h1,h1.full,h1.pre, type="html", dep.var.labels=c("Head Circumference"), column.labels=c("Full model","pre term","Full term"), title="Factors affecting head circumference in the Haifa Bay area (dependent variables – head circumference (centimeters), Box-Cox transformed values (λ=1.752); method – OLS regression)", intercept.bottom = TRUE, omit.stat = c("ser"), report=('vct*'), model.numbers = FALSE, single.row = TRUE, #remove DF df = FALSE, #which variables to keep keep = c("birthw","ges","SEX","MOTHER_COU","LNDIST_ROA","LNDIST_OIL"), covariate.labels = c("Birth Weight (kg)", "pregnancy (weeks)","Gender (1=boy, 0=girl)", "2nd quarter", "3rd quarter", "Fourth quarter"), out="~/ZH_tmp/models_HC.htm") birthw+ges+as.factor(SEX)+as.factor(month)+MOTHER_COU+LNDIST_ROA+LNDIST_OIL

2e6797095be833550c203b81909d8b7d1e53230f

55cb4f0c690a409b41b5b6e8cd0eb9d322c115d6

/R/Bossa_Simi.R

a0423831d7f002ace16de5650671a8970f6d847b

[]

no_license

TinyOpen/OnGoing

d2b711233c08b59a7ef8abfe10395912092ebbab

776a2297267c7ed57aac77c41bd7380a917a4157

refs/heads/master

2021-01-23T15:28:19.375929

2017-11-12T07:30:45

102,712,117

0

null

UTF-8

R

false

2,867

r

Bossa_Simi.R

#' Bind two factors #' #' Create a new factor from two existing factors, where the new factor's levels #' are the union of the levels of the input factors. #' #' @param a factor #' @param b factor #' #' @return factor #' @export #' @examples #'#' fbind(iris$Species[c(1, 51, 101)], PlantGrowth$group[c(1, 11, 21)]) BossaClust <- function(data, alpha = 1, p = c(0.9, 0.8, 0.7, 0.5), lin = 0.2, is.pca = TRUE, pca.sum.prop = 0.95, fix.pca.comp = FALSE, n.comp = 50, cri = 1, lintype = "ward.D2") { require("psych") # Check input data -------------------------- data.pre <- BossaSimi(data, is.pca = is.pca, pca.sum.prop = pca.sum.prop, fix.pca.comp = fix.pca.comp, n.comp = n.comp, alpha = alpha) data.simi <- data.pre$bossa.simi n <- dim(data)[1] # Do overlap cluster with "SC" method ----------------------------- overlap.pre <- OverlapClust(data.simi, p = p, lin = lin) overlap.clu <- overlap.pre$overlap.clu clust.center <- overlap.pre$clust.center # Merge clusters----------------------------- sum.clu <- dim(overlap.clu)[2] - 2 colnames(overlap.clu) <- c("first.clu", "belong.layer", paste("clust.", 1:sum.clu, sep = "")) ori.clu <- overlap.clu[,-c(1,2)] shmat <- clush(overlap.clu[, -c(1, 2)]) sig.lev <- ifelse(cri < 1, cri, ifelse(cri == 1, 0.05/sum.clu, 0.05/sum.clu/(sum.clu-1))) if(sum.clu < 2) return(list(clust.center = clust.center, overlap.clu = overlap.clu, shmat = shmat, p = p)) clumatch<-keyfeat(ori.clu, sig.lev) scrit0<-clumatch$scrit0 scrit1<-clumatch$scrit1 clu.dis<-as.dist(clumatch$stat) merclu<-clumatch$kfp sepclu<-clumatch$kfn # Take charge of the left cells non.core.ind <- (1:n)[apply(overlap.clu[, -c(1,2)], 1, sum) == 0] k1<-dim(clust.center)[1] for (i in non.core.ind){ maxci<-rep(0,k1) ij<-0 for (j in 1:k1){ ij<-ij+1 maxci[ij]<-quantile(data.simi[i,overlap.clu[,1]==j], 0.5) } max.ind<-which.max(maxci) overlap.clu[i,1]<-max.ind overlap.clu[i,(max.ind+2)]<-1 } clu.hc <- hclust(clu.dis,lintype) tree.max <-max(cutree(clu.hc, h = scrit0)) tree.min <-max(cutree(clu.hc, h = scrit1)) clu.merge <- sapply(tree.min:tree.max, ClustMerge) cell.hc.clust <- sapply(tree.min:tree.max, function(x){ hc.clust <- hclust(as.dist(data.pre$bossa.disimi), lintype) hc.tree <- cutree(hc.clust, k = 13) }) return(list(overlap.clu = clu.merge, non.overlap.clu = cell.hc.clust, ori.overlap = overlap.clu, clust.center = clust.center, clu.dis = clu.dis, tree.max = tree.max, tree.min = tree.min, cell.simi = data.simi)) } plot.interactive <- function(object){ }

9e5cd4b877d4b28e7f0a75539626be3b3bed0530

c3f1366b81357f78e9d30988ef3770d4a253ec6e

/man/summary.stdf.Rd

79dd889f0c6ca608e8b449034ef2efb49586b5fe

[]

no_license

guiludwig/stdf

ae6716d48eb4c398d73b15c199f799d0d5f04cd1

76359cf39a96e94965a584e659789f669e1dbd12

refs/heads/master

2021-01-21T23:41:11.641890

2019-03-19T19:28:44

24,806,915

5

1

null

UTF-8

R

false

true

275

rd

summary.stdf.Rd

% Generated by roxygen2: do not edit by hand % Please edit documentation in R/summary.R \name{summary.stdf} \alias{summary.stdf} \title{Summarizing STDF Model Fits} \usage{ \method{summary}{stdf}(object, ...) } \description{ \code{summary} method for class \code{"stdf"}. }

f84fadb9ec9d6a13dedc5acf016f238523211ea8

d76b09f9edb1d7ce0b5f34573b57bf42c1e9d2e6

/intro/example_2.r

2e54e80f5ed3c88cd7f0e1999e1b7fc202a6de64

[]

no_license

rmporsch/automate-science

f32175f7862d902d92a53eecb1f03f54d026e85b

73d318e77a81bdb72769f8e6fcb06c92deab73a7

refs/heads/master

2020-07-15T08:56:32.183289

2017-07-06T02:51:00

94,306,081

0

null

UTF-8

R

false

269

r

example_2.r

# Import some data message("importing data") dat <- read.csv("data.csv") # convert Fahrenheit to Celsius message("convert Fahrenheit to Celsius") dat$Temp.C <- (dat$Temp - 32)/1.8 # write table back to disk message("writing to disk") write.csv(dat, "data_withC.csv")

5f708fafe3125772c7f9e5c65fa7d1a1b7dc06da

0f2405028a6d5a919445f405b449e50d176e3b92

/Code/Make Tidy.R

60d1e89d2310a4a27f74bdecaa125716df35b271

[]

no_license

Flashlightis/Capstone-Project

8451375aef4ccdf0f022c76db14cf7bab46c3d2d

f1dbe9cd0d985138a591cebd96971a7fd5c4dea7

refs/heads/master

2021-10-04T03:31:32.830578

2018-12-03T04:42:51

125,099,535

0

null

UTF-8

R

false

3,853

r

Make Tidy.R

# Make Tidy - All Data df3_clean <- gather(df3, Web_Metric, Number, c("Sessions", "Users", "New_Users", "Pageviews", "Number_Session", "Pages_Session", "Avg_S_Duration", "Bounce")) df4_m2 <- mutate(df3_2014, month = as.POSIXlt(Date)$mon + 1) df4_m2 <- group_by(df4_m2, month) df4_2014_month <- summarize(df4_m2, Sessions = sum(Sessions, na.rm = TRUE), Pageviews = sum(Pageviews, na.rm = TRUE), Users = sum(Users, na.rm = TRUE), New_Users = sum(New_Users, na.rm = TRUE), Number_Session = sum(Number_Session, na.rm = TRUE), Pages_Session = sum(Pages_Session, na.rm = TRUE), Avg_S_Duration = sum(Avg_S_Duration, na.rm = TRUE), Bounce = mean(Bounce, na.rm = TRUE)) df4_2014_month <- gather(df4_2014_month, Web_Metric, Number, c("Sessions", "Users", "New_Users")) # Make Tidy # 2015 df5_m2 <- mutate(df3_2015, month = as.POSIXlt(Date)$mon + 1) df5_m2 <- group_by(df5_m2, month) df5_2015_month <- summarize(df5_m2, Sessions = sum(Sessions, na.rm = TRUE), Pageviews = sum(Pageviews, na.rm = TRUE), Users = sum(Users, na.rm = TRUE), New_Users = sum(New_Users, na.rm = TRUE), Number_Session = sum(Number_Session, na.rm = TRUE), Pages_Session = sum(Pages_Session, na.rm = TRUE), Avg_S_Duration = sum(Avg_S_Duration, na.rm = TRUE), Bounce = mean(Bounce, na.rm = TRUE)) df5_2015_month <- gather(df5_2015_month, Web_Metric, Number, c("Sessions", "Users", "New_Users")) # Make Tidy # 2016 df6_m2 <- mutate(df3_2016, month = as.POSIXlt(Date)$mon + 1) df6_m2 <- group_by(df6_m2, month) df6_2016_month <- summarize(df6_m2, Sessions = sum(Sessions, na.rm = TRUE), Pageviews = sum(Pageviews, na.rm = TRUE), Users = sum(Users, na.rm = TRUE), New_Users = sum(New_Users, na.rm = TRUE), Number_Session = sum(Number_Session, na.rm = TRUE), Pages_Session = sum(Pages_Session, na.rm = TRUE), Avg_S_Duration = sum(Avg_S_Duration, na.rm = TRUE), Bounce = mean(Bounce, na.rm = TRUE)) df6_2016_month <- gather(df6_2016_month, Web_Metric, Number, c("Sessions", "Users", "New_Users")) # Make Tidy # 2017 df7_m2 <- mutate(df3_2017, month = as.POSIXlt(Date)$mon + 1) df7_m2 <- group_by(df7_m2, month) df7_2017_month <- summarize(df7_m2, Sessions = sum(Sessions, na.rm = TRUE), Pageviews = sum(Pageviews, na.rm = TRUE), Users = sum(Users, na.rm = TRUE), New_Users = sum(New_Users, na.rm = TRUE), Number_Session = sum(Number_Session, na.rm = TRUE), Pages_Session = sum(Pages_Session, na.rm = TRUE), Avg_S_Duration = sum(Avg_S_Duration, na.rm = TRUE), Bounce = mean(Bounce, na.rm = TRUE)) df7_2017_month <- gather(df7_2017_month, Web_Metric, Number, c("Sessions", "Users", "New_Users")) # Make Tidy # Convert to Tidy Dataset --- df3_ty2 <- gather(df3_Total_year, Web_Metric, Number, Sessions:Users) df3_ty3 <- gather(df3_Total_year, Web_Metric, Number, c("Sessions", "Users", "New_Users")) View(df3_ty) View(df3_ty2) View(df3_ty3)

a97ccdf5686c9ff1929e318c7504d536b32acdc7

c5f708e71aae6e56605e9f30a57e349ca1ced4a4

/server.R

51a392a928bb3b5f6f18a209747568c1c21b78e9

[]

no_license

xjlc/climit

977f6b3aa6644fa63fd4c344ad98a3e8e43275fb

07656c44691e7c6697f90c45c1218320e481210e

refs/heads/master

2021-01-10T21:21:07.621348

2015-01-22T06:33:22

29,587,583

0

null

UTF-8

R

false

4,184

r

server.R

# illustration of the CLT # based on a github gist by https://github.com/tgouhier/climit library(shiny) shinyServer(function(input, output) { # one simulation run: generate random numbers simdata <- function(input, n) { if (input$dist=="rpois") { vals <- do.call(input$dist, list(n=input$n, lambda=1)) } else if (input$dist=="rbinom") { vals <- do.call(input$dist, list(n=input$n, size=30, p=.25)) } else { vals <- do.call(input$dist, list(n=input$n)) } } data <- reactive({ vals <- simdata(input, n) return (list(fun=input$dist, vals=vals)) }) output$plot <- renderPlot({ # generate plot title based on user-chosen distribution distname <- switch(input$dist, runif = "Uniform distribution", # (n = ", rnorm = "Normal distribution", # (n = ", rlnorm = "Log-normal distribution", # (n = ", rexp = "Exponential distribution", # (n = ", rbinom = "Binomial distribution (n=30, p=.25)", rpois = "Poisson distribution", rcauchy = "Cauchy distribution") # (n = ") # extract parameters from user input n <- input$n N <- input$N pdist <- data()$vals # generate N samples x <- replicate(N, simdata(input, n)) # extract means of samples # note: this was rowMeans in the original code, but I think that was a mistake ndist <- colMeans(x) # expected values from the literature/formulary expect <- switch(input$dist, rexp = c(1^-1, 1^-2), rnorm = c(0, 1), rlnorm = c(exp(0+(1/2)*1^2), exp(0 + 1^2)*(exp(1^2)-1)), runif = c(0.5, (1/12)*1), rbinom = c(30*.25, 30*.25*.75), rpois =c(1, 1), rcauchy = rep(NA, 2)) obs <- data.frame(pdist=c(mean(pdist), var(pdist)), ndist=c(mean(ndist), var(ndist))) # TODO: better visualization, ggplot?, add means, samples, etc. nbreaks <- 10 par(mfrow=c(2,2)) # first panel: a single simulation pdens <- density(pdist) phist <- hist(pdist, plot=FALSE) hist(pdist, main=paste("A single sample of", n, "observations\nfrom the", distname), xlab="Values (X)", freq=FALSE, ylim=c(0, max(pdens$y, phist$density)), breaks=nbreaks) lines(pdens, col="black", lwd=2) abline(v=obs$pdist[1], col="blue", lwd=2, lty=2) abline(v=expect[1], col="red", lwd=2, lty=2) legend(x="topright", col=c("black", "red"), lwd=2, lty=2, legend=c("Observed", "Expected")) box() # second panel: add a plot showing the individual distributions # densities <- apply(x, 2, density, bw="SJ", adjust=2) if (input$dist=="rexp" | input$dist=="rlnorm" | input$dist=="rpois") { xl <- c(0, max(as.vector(x))) densities <- apply(x, 2, density, from=0.05) } else if (input$dist=="runif") { xl <- c(0, 1) densities <- apply(x, 2, density, n=512, from=0.02, to=.98) } else { xl <- range(as.vector(x)) densities <- apply(x, 2, density) } plot(densities[[1]], type="l", lwd=.5, xlim=xl, ylim=c(0, max(sapply(densities, "[[", "y"))), main="Individual samples (smoothed)\nwith sample means in red", xlab="Value") sapply(densities, lines, lwd=.5) abline(v=ndist, col="red", lty=1, lwd=.25) # third panel: histogram of sample means ndens <- density(ndist) nhist <- hist(ndist, plot=FALSE) hist(ndist, main=paste("Distribution of mean values from ", N, " random samples each\nconsisting of ", n, " observations from the ", distname, sep=""), col="red", xlab=expression(paste("Sample means (", bar(X), ")")), freq=FALSE, ylim=c(0, max(ndens$y, nhist$density)), breaks=nbreaks, xlim=range(phist$breaks)) lines(ndens, col="black", lwd=3) abline(v=obs$ndist[1], col="blue", lwd=2, lty=2) abline(v=expect[1], col="red", lwd=2, lty=2) legend(x="topright", col=c("blue", "red"), lwd=2, lty=2, legend=c("Observed", "Expected")) box() # fourth panel: compare sample means to normal distribution qqnorm(ndist, main=paste("Distribution of sample means\n from the", distname, "against Normal")) qqline(ndist) }) })

cffeeacdd4c0573e36ba456b331d949c86b249c1

f95b3720c2ff266261cfbbd2e06fc987b34db08a

/R/CopyToClipboard.R

fc21a7e2f8db3af6278ff0f63877ecaec3d88f6e

[]

no_license

selinaZitrone/Lessons_Alessio

674422516ff193c5834bca6a2abc46b2007448a5

1f816c104c7bc760e967669eb52cb1d56680d3d1

refs/heads/master

2023-01-19T09:28:07.413472

2020-10-05T17:48:51

300,846,108

0

2

null

UTF-8

R

false

349

r

CopyToClipboard.R

# file (string): file to be copied to the clipboard (with fileending) # filepath (string): path where file to be copied to clipboard is stored CopyToClipboard <- function(file, filepath = here::here("HTML")){ readLines(paste0(filepath, "/", file)) %>% clipr::write_clip() #xml2::read_html(paste0(filepath, "/", file)) %>% clipr::write_clip() }

69525749ec125d28e88c8998609e47b64e5f2e05

bac8d44dff959258dc7a706a35e1ebd1944c7641

/plot1.R

9a476505549687d818f137929ab8154ad02460b1

[]

no_license

subhashish7/ExData_Plotting1

889fefa4ca6d0511fd0f22e6335f4bc48e91286a

35b93a5192b2e5c79cca88c9b21c4bbcc69018b3

refs/heads/master

2020-04-10T14:00:01.145357

2018-12-28T15:00:45

161,064,479

0

null

2018-12-09T17:33:57

2018-12-09T17:33:56

null

UTF-8

R

false

494

r

plot1.R

# set working directory setwd('C:/Users/Desktop/Coursera/EDA') # read file data <- read.csv('household_power_consumption.txt', header = TRUE, sep=';', dec='.', stringsAsFactors = FALSE) # select data from relevant dates df <- data[data$Date %in% c("1/2/2007","2/2/2007"),] # open device png("plot1.png", width=480, height=480) # plot histogram hist(as.numeric(df$Global_active_power), col="red", main="Global Active Power", xlab="Global Active Power (kilowatts)") # close device dev.off()

d8eda88c55e5496d893150f3c5569fce02edac66

0d85667371d11e998a8526cbfea00eb919ef3e30

/man/convSiteData.Rd

db701e396efed1f9da9141581076ab481457ea5f

[ "MIT" ]

permissive

dinaIssakova/rgenesconverged

5ef5230af0cde0c61545cfed73f0bb5c2f0f1da3

e1b5bb82bfa8fe03d232f1aa19bb3ed785252d07

refs/heads/master

2020-07-28T06:07:59.193155

2020-01-06T20:48:17

209,332,677

0

null

UTF-8

R

false

true

937

rd

convSiteData.Rd

% Generated by roxygen2: do not edit by hand % Please edit documentation in R/getResults.R \name{convSiteData} \alias{convSiteData} \title{Get number of convergent sites} \usage{ convSiteData( tree, phydat, spe1, spe2, t, type = c("abs", "score"), m = getm(tree, phydat, spe1, spe2) ) } \arguments{ \item{tree}{A phylogenetic tree} \item{phydat}{An object of class phydat} \item{spe1}{The name of species 1} \item{spe2}{The name of species 2} \item{t}{threshold} \item{type}{Type of analysis: 'abs' for basic model or 'score' for by convergence score model} \item{m}{The length of each gene (Up to what is desired to be evaluated). Default is entire gene} } \value{ The number of potentially convergent sites } \description{ Get the number of potentially evolved convergent sites and print the probability that this occured by chance. } \examples{ \dontrun{ convSiteData(smallTree, primates, "Human", "Chimp", 5) } }

e3fde9575f98117e86373d41ba582037061a5358

3ab08891487e23f0a6bcf184649ba331a938070b

/NYC_Paking_Graded_Case_Study_V6.R

718c7d8bd2fc757a581cf6617ce281b2b309d30d

[]

no_license

Abhijit-Barik01/SPARK-R-SQL-NYC-PArking-Ticket-Analysis

b7a00f820ae3f7003ffaaa56c8a9e4db7e899c80

85f11875b4e8eda5a0faae56fdc9c9664d3d1d99

refs/heads/master

2023-03-18T08:17:05.755512

2019-01-01T07:03:43

null

0

null

UTF-8

R

false

77,588

r

NYC_Paking_Graded_Case_Study_V6.R

####################################################################################################################################################### #1.Problem Statement/Business Understanding #2.Broad Assumptions #3.Initialise Environment,load libraries & data #4.Understanding #5.Assignment Tasks - Examining Data #6.Assignment Tasks - Aggregation tasks #7.Closure ####################################################################################################################################################### ####################################################################################################################################################### #1.Problem Statement/Business Understanding ####################################################################################################################################################### #NYC Police Department has collected data for parking tickets.We have been provided with data files for 2015,2016 and 2017 #The purpose of this case study is to conduct an exploratory data analysis that helps to understand the data. #The scope of this analysis, we wish to compare the phenomenon related to parking tickets over three different years - 2015, 2016, 2017 #It's reccomended to do analysis over fiscal year however it's fine to use calendar year approach as well. # ####################################################################################################################################################### #2.Broad Assumptions #-We will be using calendar year instead of fiscal year- as permitted in the problem statement. #-We'll load all files together for analysis and perform required analysis.This will cause all data pertaining to calendar years 2015,2016, 2017 #-to be considered valid for case study.Some of this data would have been invalid in other approach/es. #-Since the purpose of this case study is EDA itself, we have performed EDA only as needed and cleanup is performed only in case it affects the analysis. #-We are assuming that all required libraries are installed in the environment prior to execution #-It was observed that 2017 file contained 8 less columns however names of the columns were same in all 3 years hence it was decided # to use combined data load instead of individual data frames per year to avoid repetative code. ####################################################################################################################################################### ####################################################################################################################################################### #3.Initialise Environment, load libraries & data ####################################################################################################################################################### #File Location #'/common_folder/nyc_parking/Parking_Violations_Issued_-_Fiscal_Year_201x.csv' # Load SparkR spark_path <- '/usr/local/spark' if (nchar(Sys.getenv("SPARK_HOME")) < 1) { Sys.setenv(SPARK_HOME = spark_path) } library(SparkR, lib.loc = c(file.path(Sys.getenv("SPARK_HOME"), "R", "lib"))) # Initialise the sparkR session sparkR.session(master = "yarn-client", sparkConfig = list(spark.driver.memory = "1g")) #Ensure that all libraries below are installed before loading #Sample install command #install.packages("sparklyr") library(sparklyr) library(dplyr) library(ggplot2) library(ggrepel) library(tidyr) #connect to spark session #sc <- spark_connect(master = "yarn-client") # Before executing any hive-sql query from RStudio, you need to add a jar file in RStudio sql("ADD JAR /opt/cloudera/parcels/CDH/lib/hive/lib/hive-hcatalog-core-1.1.0-cdh5.11.2.jar") #filePath <- "hdfs:///common_folder/nyc_parking/Parking_Violations_Issued_-_Fiscal_Year_201*.csv" #Load all data files at once - combined data NYC_Ticket_Base<-SparkR::read.df("hdfs:///common_folder/nyc_parking/Parking_Violations_Issued_-_Fiscal_Year_201*.csv", "CSV", header="true", inferSchema = "true") #Load individual year files NYC_Ticket_Base_2015<-SparkR::read.df("hdfs:///common_folder/nyc_parking/Parking_Violations_Issued_-_Fiscal_Year_2015.csv", "CSV", header="true", inferSchema = "true") NYC_Ticket_Base_2016<-SparkR::read.df("hdfs:///common_folder/nyc_parking/Parking_Violations_Issued_-_Fiscal_Year_2016.csv", "CSV", header="true", inferSchema = "true") NYC_Ticket_Base_2017<-SparkR::read.df("hdfs:///common_folder/nyc_parking/Parking_Violations_Issued_-_Fiscal_Year_2017.csv", "CSV", header="true", inferSchema = "true") ####################################################################################################################################################### #4.Understanding/Examining Data ####################################################################################################################################################### ncol(NYC_Ticket_Base_2015) #51 columns nrow(NYC_Ticket_Base_2015) #11809233 records/rows ncol(NYC_Ticket_Base_2016) #51 columns nrow(NYC_Ticket_Base_2016) #10626899 records/rows ncol(NYC_Ticket_Base_2017) #43 columns nrow(NYC_Ticket_Base_2017) #10803028 records/rows ncol(NYC_Ticket_Base)#51 columns nrow(NYC_Ticket_Base)#33239160 records/rows colnames(NYC_Ticket_Base_2015) #[1] "Summons Number" "Plate ID" "Registration State" #[4] "Plate Type" "Issue Date" "Violation Code" #[7] "Vehicle Body Type" "Vehicle Make" "Issuing Agency" #[10] "Street Code1" "Street Code2" "Street Code3" #[13] "Vehicle Expiration Date" "Violation Location" "Violation Precinct" #[16] "Issuer Precinct" "Issuer Code" "Issuer Command" #[19] "Issuer Squad" "Violation Time" "Time First Observed" #[22] "Violation County" "Violation In Front Of Or Opposite" "House Number" #[25] "Street Name" "Intersecting Street" "Date First Observed" #[28] "Law Section" "Sub Division" "Violation Legal Code" #[31] "Days Parking In Effect " "From Hours In Effect" "To Hours In Effect" #[34] "Vehicle Color" "Unregistered Vehicle?" "Vehicle Year" #[37] "Meter Number" "Feet From Curb" "Violation Post Code" #[40] "Violation Description" "No Standing or Stopping Violation" "Hydrant Violation" #[43] "Double Parking Violation" "Latitude" "Longitude" #[46] "Community Board" "Community Council " "Census Tract" #[49] "BIN" "BBL" "NTA" str(NYC_Ticket_Base_2015) # 'SparkDataFrame': 51 variables: # $ Summons Number : num 8002531292 8015318440 7611181981 7445908067 7037692864 7704791394 # $ Plate ID : chr "EPC5238" "5298MD" "FYW2775" "GWE1987" "T671196C" "JJF6834" # $ Registration State : chr "NY" "NY" "NY" "NY" "NY" "PA" # $ Plate Type : chr "PAS" "COM" "PAS" "PAS" "PAS" "PAS" # $ Issue Date : chr "10/01/2014" "03/06/2015" "07/28/2014" "04/13/2015" "05/19/2015" "11/20/2014" # $ Violation Code : int 21 14 46 19 19 21 # $ Vehicle Body Type : chr "SUBN" "VAN" "SUBN" "4DSD" "4DSD" "4DSD" # $ Vehicle Make : chr "CHEVR" "FRUEH" "SUBAR" "LEXUS" "CHRYS" "NISSA" # $ Issuing Agency : chr "T" "T" "T" "T" "T" "T" # $ Street Code1 : int 20390 27790 8130 59990 36090 74230 # $ Street Code2 : int 29890 19550 5430 16540 10410 37980 # $ Street Code3 : int 31490 19570 5580 16790 24690 38030 # $ Vehicle Expiration Date : chr "01/01/20150111 12:00:00 PM" "01/01/88888888 12:00:00 PM" "01/01/20160524 12:0 # $ Violation Location : int 7 25 72 102 28 67 # $ Violation Precinct : int 7 25 72 102 28 67 # $ Issuer Precinct : int 7 25 72 102 28 67 # $ Issuer Code : int 345454 333386 331845 355669 341248 357104 # $ Issuer Command : chr "T800" "T103" "T302" "T402" "T103" "T302" # $ Issuer Squad : chr "A2" "B" "L" "D" "X" "A" # $ Violation Time : chr "0011A" "0942A" "1020A" "0318P" "0410P" "0839A" # $ Time First Observed : chr "NA" "NA" "NA" "NA" "NA" "NA" # $ Violation County : chr "NY" "NY" "K" "Q" "NY" "K" # $ Violation In Front Of Or Opposite: chr "F" "F" "F" "F" "F" "F" # $ House Number : chr "133" "1916" "184" "120-20" "66" "1013" # $ Street Name : chr "Essex St" "Park Ave" "31st St" "Queens Blvd" "W 116th St" "Rutland Rd" # $ Intersecting Street : chr "NA" "NA" "NA" "NA" "NA" "NA" # $ Date First Observed : chr "01/05/0001 12:00:00 PM" "01/05/0001 12:00:00 PM" "01/05/0001 12:00:00 PM" "01 # $ Law Section : int 408 408 408 408 408 408 # $ Sub Division : chr "d1" "c" "f1" "c3" "c3" "d1" # $ Violation Legal Code : chr "NA" "NA" "NA" "NA" "NA" "NA" # $ Days Parking In Effect : chr "Y Y Y" "YYYYY" "NA" "YYYYY" "YYYYYYY" "Y" # $ From Hours In Effect : chr "1200A" "0700A" "NA" "0300P" "NA" "0830A" # $ To Hours In Effect : chr "0300A" "1000A" "NA" "1000P" "NA" "0900A" # $ Vehicle Color : chr "BL" "BROWN" "BLACK" "GY" "BLACK" "WHITE" # $ Unregistered Vehicle? : int NA NA NA NA NA NA # $ Vehicle Year : int 2005 0 2010 2015 0 0 # $ Meter Number : chr "NA" "NA" "NA" "NA" "NA" "NA" # $ Feet From Curb : int 0 0 0 0 0 0 # $ Violation Post Code : chr "A 77" "CC3" "J 32" "01 4" "19 7" "C 32" # $ Violation Description : chr "21-No Parking (street clean)" "14-No Standing" "46A-Double Parking (Non-COM)" # $ No Standing or Stopping Violation: chr "NA" "NA" "NA" "NA" "NA" "NA" # $ Hydrant Violation : chr "NA" "NA" "NA" "NA" "NA" "NA" # $ Double Parking Violation : chr "NA" "NA" "NA" "NA" "NA" "NA" # $ Latitude : chr "NA" "NA" "NA" "NA" "NA" "NA" # $ Longitude : chr "NA" "NA" "NA" "NA" "NA" "NA" # $ Community Board : chr "NA" "NA" "NA" "NA" "NA" "NA" # $ Community Council : chr "NA" "NA" "NA" "NA" "NA" "NA" # $ Census Tract : chr "NA" "NA" "NA" "NA" "NA" "NA" # $ BIN : chr "NA" "NA" "NA" "NA" "NA" "NA" # $ BBL : chr "NA" "NA" "NA" "NA" "NA" "NA" # $ NTA : chr "NA" "NA" "NA" "NA" "NA" "NA" setdiff(colnames(NYC_Ticket_Base_2015),colnames(NYC_Ticket_Base_2016)) #Both data sets have same columns setdiff(colnames(NYC_Ticket_Base_2015),colnames(NYC_Ticket_Base_2017)) #NYC_Ticket_Base_2017 does not have below columns # [1] "Latitude" "Longitude" "Community Board" "Community Council " "Census Tract" "BIN" # [7] "BBL" "NTA" #This shows that all data files contain same columns- 2017 has 8 less columns however those are not requrired for analysis #Hence we will use the combined data set that contains data for all years for further analysis. # Finding number of null values in each column nullcount <- SparkR::select(NYC_Ticket_Base, lapply(columns(NYC_Ticket_Base), function(c) alias(sum(cast(isNull(NYC_Ticket_Base[[c]]), "integer")), c))) nullcount %>% as.data.frame %>% gather(NYC_Ticket_Base, sum_null) # NYC_Ticket_Base sum_null # 1 Summons Number 0 # 2 Plate ID 4 # 3 Registration State 0 # 4 Plate Type 0 # 5 Issue Date 0 # 6 Violation Code 0 # 7 Vehicle Body Type 127700 # 8 Vehicle Make 212135 # 9 Issuing Agency 0 # 10 Street Code1 0 # 11 Street Code2 0 # 12 Street Code3 0 # 13 Vehicle Expiration Date 1 # 14 Violation Location 5740226 # 15 Violation Precinct 1 # 16 Issuer Precinct 1 # 17 Issuer Code 1 # 18 Issuer Command 5704460 # 19 Issuer Squad 5706033 # 20 Violation Time 6058 # 21 Time First Observed 30042504 # 22 Violation County 3594319 # 23 Violation In Front Of Or Opposite 5985339 # 24 House Number 6312797 # 25 Street Name 18338 # 26 Intersecting Street 23509505 # 27 Date First Observed 2 # 28 Law Section 2 # 29 Sub Division 5008 # 30 Violation Legal Code 27532215 # 31 Days Parking In Effect 8418387 # 32 From Hours In Effect 15613695 # 33 To Hours In Effect 15613692 # 34 Vehicle Color 415377 # 35 Unregistered Vehicle? 29595048 # 36 Vehicle Year 4 # 37 Meter Number 27290553 # 38 Feet From Curb 4 # 39 Violation Post Code 9349625 # 40 Violation Description 3647162 # 41 No Standing or Stopping Violation 33239159 # 42 Hydrant Violation 33239159 # 43 Double Parking Violation 33239159 # 44 Latitude 33239160 # 45 Longitude 33239160 # 46 Community Board 33239160 # 47 Community Council 33239160 # 48 Census Tract 33239160 # 49 BIN 33239160 # 50 BBL 33239160 # 51 NTA 33239160 # A lot of columns towards end of data frame are completely empty # Based on questions, we will not need these column for analysis hence we can remove these. # There are some columns with high number of nulls.These are not needed for analysis hence we decided to leave those as it is. # `violation location` and `Violation Time` are needed for analysis and have nulls to be checked as part of analysis # Hence we'll treat those nulls later during the analysis as needed for this case study NYC_Ticket_Base<-SparkR::dropna(NYC_Ticket_Base,how="any", cols=c("Vehicle Body Type","Vehicle Make","Violation Precinct","Issuer Precinct")) NYC_Ticket_Base$`No Standing or Stopping Violation`<-NULL NYC_Ticket_Base$`Hydrant Violation`<-NULL NYC_Ticket_Base$`Double Parking Violation`<-NULL NYC_Ticket_Base$`Latitude`<-NULL NYC_Ticket_Base$`Longitude`<-NULL NYC_Ticket_Base$`Community Board`<-NULL NYC_Ticket_Base$`Community Council `<-NULL NYC_Ticket_Base$`Census Tract`<-NULL NYC_Ticket_Base$`BIN`<-NULL NYC_Ticket_Base$`BBL`<-NULL NYC_Ticket_Base$`NTA`<-NULL #unique values unique_NYC_Ticket<-SparkR:::lapply(names(NYC_Ticket_Base),function(x) alias(countDistinct(NYC_Ticket_Base[[x]]), x)) head(do.call(agg, c(x = NYC_Ticket_Base, unique_NYC_Ticket))) # # Summons Number Plate ID Registration State Plate Type Issue Date Violation Code Vehicle Body Type # 1 31864311 6021216 69 89 3379 100 3824 # Vehicle Make Issuing Agency Street Code1 Street Code2 Street Code3 Vehicle Expiration Date # 1 12679 20 7024 7319 7072 9555 # Violation Location Violation Precinct Issuer Precinct Issuer Code Issuer Command Issuer Squad # 1 579 580 847 60900 5977 50 # Violation Time Time First Observed Violation County Violation In Front Of Or Opposite House Number # 1 2169 2547 19 6 78608 # Street Name Intersecting Street Date First Observed Law Section Sub Division Violation Legal Code # 1 189521 407674 1517 9 144 5 # Days Parking In Effect From Hours In Effect To Hours In Effect Vehicle Color Unregistered Vehicle? # 1 190 784 905 4418 4 # Vehicle Year Meter Number Feet From Curb Violation Post Code Violation Description # 1 100 54500 17 1234 110 #It is observed that there are nearly 1.1M(32932223 vs 31864311 unique) records with same Summons Numbers. #Summon Number is expected to be unique hence based on https://learn.upgrad.com/v/course/126/question/99083 #We will remove these records createOrReplaceTempView(NYC_Ticket_Base, "NYC_Ticket_Base_tab") NYC_Ticket_Base<-SparkR::sql("select * from NYC_Ticket_Base_tab where `Summons Number` in ( select `Summons Number` from NYC_Ticket_Base_tab group by `Summons Number` having count(*) = 1 )") ncol(NYC_Ticket_Base)#40 nrow(NYC_Ticket_Base)#30842504 NYC_Ticket_Base$`Issue Date`<-SparkR::to_date(NYC_Ticket_Base$`Issue Date`, "MM/dd/yyyy") NYC_Ticket_Base$Issue_Date_Year<-SparkR::year(NYC_Ticket_Base$`Issue Date`) #Creating a table for quering createOrReplaceTempView(NYC_Ticket_Base, "NYC_Ticket_Base_tab") ####################################################################################################################################################### #5.Assignment Tasks - Examining Data ####################################################################################################################################################### #------------------------------------------------------------------------------------------------------------------------------------------------------ #A1. Find the total number of tickets for each year. #We'll check if the data is clean first before getting answers Year_Wise_Record <- SparkR::sql("SELECT Issue_Date_Year, count(*) REC_COUNT FROM NYC_Ticket_Base_tab group by Issue_Date_Year") head(Year_Wise_Record) # Issue_Date_Year REC_COUNT # 1 1990 3 # 2 2025 40 # 3 1975 1 # 4 1977 1 # 5 2027 48 # 6 2003 5 # looks like the year spred is more than the expected 3 years #For the scope of this analysis, we wish to compare the phenomenon related to parking tickets over three different years - 2015, 2016, 2017 nrow(Year_Wise_Record) #71 seems lie a lot of years data is present, lets filter out unwanted years NYC_Ticket_Base <- SparkR::sql("SELECT * FROM NYC_Ticket_Base_tab where Issue_Date_Year in (2015, 2016, 2017)") createOrReplaceTempView(NYC_Ticket_Base, "NYC_Ticket_Base_tab") #let's check yearwise number of tickets in the data Year_Wise_Record <- SparkR::sql("SELECT Issue_Date_Year, count(*) REC_COUNT FROM NYC_Ticket_Base_tab group by Issue_Date_Year") head(Year_Wise_Record) # Issue_Date_Year REC_COUNT # 1 2015 10008087 # 2 2016 10146977 # 3 2017 5377542 #The ticket counts have decreased over the years, specially 2017 seems to have very low numbers #Alternate -Let's check this on a plot c <- SparkR::count(groupBy(NYC_Ticket_Base, "Issue_Date_Year")) c.r <- SparkR::collect(c) year_count <- c.r[c.r$Issue_Date_Year %in% c(2015,2016,2017), 1:2] # plot showing yearly number of tickets g <- ggplot(year_count, aes(x=year_count$Issue_Date_Year, y=year_count$count)) g + geom_bar(stat = "identity") + geom_text(label = year_count$count, position = position_stack(vjust = 0.5)) #------------------------------------------------------------------------------------------------------------------------------------------------------ #A2. Find out the number of unique states from where the cars that got parking tickets came from. #(Hint: Use the column 'Registration State') # There is a numeric entry in the column which should be corrected. #Replace it with the state having maximum entries. Give the number of unique states for each year again. #First check the data for cleanup needs- as mentioned, there is numeric entry in the data that needs to be fixed. State_Wise<-SparkR::sql("select `Registration State`,Count(*) Record_count from NYC_Ticket_Base_tab group by `Registration State` order by Count(*) desc") nrow(State_Wise) #Number of states in overall data - 69.This contains 99 which is invalid. head(State_Wise,69) #Maximum entries NY 20024087 #Numeric entries 99 73154 #Fix the numeric state records with NY NYC_Ticket_Base <- SparkR::sql("SELECT NYC_Ticket_Base_tab.*, case when `Registration State`=99 then 'NY' else `Registration State` end Registration_State FROM NYC_Ticket_Base_tab ") createOrReplaceTempView(NYC_Ticket_Base, "NYC_Ticket_Base_tab") Year_State<- SparkR::sql("select Issue_Date_Year, count(distinct Registration_State) REC_COUNT FROM NYC_Ticket_Base_tab group by Issue_Date_Year") head(Year_State,nrow(Year_State)) #Number of unique states in parking tickets data # Issue_Date_Year REC_COUNT # 1 2015 68 # 2 2016 67 # 3 2017 64 #Plot - Number of tickets by Registration State rs <- SparkR::count(groupBy(NYC_Ticket_Base, "Registration State")) rs <- SparkR::collect(rs) rs <- (arrange(rs, desc(rs$count))) # plot showing number of tickets based on Registration State g <- ggplot(rs, aes(x=reorder(rs$`Registration State`,-rs$count), y=rs$count)) g + geom_bar(stat = "identity") + theme(axis.text.x = element_text(angle = 90, hjust = 1)) head(rs) # Registration State count # 1 NY 20024087 # 2 NJ 2266677 # 3 PA 638387 # 4 CT 339772 # 5 FL 330976 # 6 MA 214649 # Number of tickets by Plate Type pt <- SparkR::count(groupBy(NYC_Ticket_Base, "Plate Type")) pt <- SparkR::collect(pt) pt <- (arrange(pt, desc(pt$count))) # plot indicating number of tickets based on Plate Type g <- ggplot(pt, aes(x=reorder(pt$`Plate Type`,-pt$count), y=pt$count)) g + geom_bar(stat = "identity") + theme(axis.text.x = element_text(angle = 90, hjust = 1)) head(pt) # Plate Type count # 1 PAS 18673869 # 2 COM 4702049 # 3 OMT 939756 # 4 OMS 230251 # 5 SRF 218486 # 6 IRP 138984 #Passenger vehicles in NY have highest violations #------------------------------------------------------------------------------------------------------------------------------------------------------ #A3. Some parking tickets don't have the address for violation location on them, #which is a cause for concern. Write a query to check the number of such tickets. #The values should not be deleted or imputed here. This is just a check. violation_location_missing<-SparkR::sql("select count(*) REC_COUNT from NYC_Ticket_Base_tab where `violation location` is null") head(violation_location_missing) # REC_COUNT # 4448923 (4448923*100)/nrow(NYC_Ticket_Base) #17.4% records have violation location missing from the records violation_location_missing_yearwise<-SparkR::sql("select Issue_Date_Year,count(*) REC_COUNT from NYC_Ticket_Base_tab where `violation location` is null group by Issue_Date_Year") head(violation_location_missing_yearwise) #Yearwise records - Violation location missing # Issue_Date_Year REC_COUNT # 1 2015 1555016 # 2 2016 1970527 # 3 2017 923380 ####################################################################################################################################################### #6.Assignment Tasks - Aggregation tasks ####################################################################################################################################################### #------------------------------------------------------------------------------------------------------------------------------------------------------ #A1.How often does each violation code occur? Display the frequency of the top five violation codes. violation_code_freq<-SparkR::sql("select `violation code`, count(*) REC_COUNT from NYC_Ticket_Base_tab group by `violation code` order by count(*) desc") head(violation_code_freq,5) # violation code REC_COUNT # 1 21 3595056 # 2 36 2964148 # 3 38 2749510 # 4 14 2138952 # 5 37 1594604 #Year wise top 5 violation_code_freq_yearwise<-SparkR::sql("select Issue_Date_Year,`violation code`, count(*) REC_COUNT from NYC_Ticket_Base_tab group by `violation code`,Issue_Date_Year order by count(*) desc") violation_code_freq_2015<-SparkR::collect(SparkR::filter(violation_code_freq_yearwise, violation_code_freq_yearwise$Issue_Date_Year == 2015)) head(violation_code_freq_2015,5) #Top 5 violation code in 2015 # Issue_Date_Year violation code REC_COUNT # 1 2015 21 1425779 # 2 2015 38 1143343 # 3 2015 36 951024 # 4 2015 14 851119 # 5 2015 37 668394 violation_code_freq_2016<-SparkR::collect(SparkR::filter(violation_code_freq_yearwise, violation_code_freq_yearwise$Issue_Date_Year == 2016)) head(violation_code_freq_2016,5) #Top 5 violation code in 2016 # Issue_Date_Year violation code REC_COUNT # 1 2016 21 1409997 # 2 2016 36 1351297 # 3 2016 38 1066339 # 4 2016 14 815800 # 5 2016 37 633584 violation_code_freq_2017<-SparkR::collect(SparkR::filter(violation_code_freq_yearwise, violation_code_freq_yearwise$Issue_Date_Year == 2017)) head(violation_code_freq_2017,5) #Top 5 violation code in 2017 # Issue_Date_Year violation code REC_COUNT # 1 2017 21 759280 # 2 2017 36 661827 # 3 2017 38 539828 # 4 2017 14 472033 # 5 2017 20 317551 #Plot the violation codes for visual analysis violation_code_freq_yearwise<-SparkR::collect(SparkR::filter(violation_code_freq_yearwise, violation_code_freq_yearwise$REC_COUNT>50000)) plot <- ggplot(violation_code_freq_yearwise,aes(x = factor(Issue_Date_Year), y = REC_COUNT,col=`violation code`,label=`violation code`)) + geom_point() + geom_label_repel(aes(label = `violation code`), box.padding = 0.35, point.padding = 0.5, segment.color = 'grey50') + theme_classic()+ xlab("violation code") + ylab("REC_COUNT") plot #21,38,36,14 and 37 all are top 5 violation codes #Let's check number of tickets by violation code through another visualization vc <- SparkR::count(groupBy(NYC_Ticket_Base, "Violation Code")) vc <- SparkR::collect(vc) vc <- (arrange(vc, desc(vc$count))) # plot indicating number of tickets based on violation code g <- ggplot(vc, aes(x=reorder(vc$`Violation Code`,-vc$count), y=vc$count)) g + geom_bar(stat = "identity") + theme(axis.text.x = element_text(angle = 90, hjust = 1)) head(vc) # Violation Code count # 1 21 3595056 # 2 36 2964148 # 3 38 2749510 # 4 14 2138952 # 5 37 1594604 # 6 20 1475446 #21,38,36,14 and 37 all are overall top 5 violation codes #------------------------------------------------------------------------------------------------------------------------------------------------------ #A2.How often does each 'vehicle body type' get a parking ticket? #How about the 'vehicle make'? (Hint: find the top 5 for both) #Overall #Body Type vehicle_body_type_freq<-SparkR::sql("select `vehicle body type`, count(*) REC_COUNT from NYC_Ticket_Base_tab group by `vehicle body type` order by count(*) desc") head(vehicle_body_type_freq,5) # vehicle body type REC_COUNT # 1 SUBN 8551193 # 2 4DSD 7296843 # 3 VAN 3571103 # 4 DELV 1754513 # 5 SDN 994615 #Make vehicle_make_freq<-SparkR::sql("select `vehicle make`, count(*) REC_COUNT from NYC_Ticket_Base_tab group by `vehicle make` order by count(*) desc") head(vehicle_make_freq,5) # vehicle make REC_COUNT # 1 FORD 3155810 # 2 TOYOT 2801819 # 3 HONDA 2486750 # 4 NISSA 2077591 # 5 CHEVR 1799535 ## Year wise analysis #Body Type vehicle_body_type_year_freq<-SparkR::sql("select `vehicle body type`,Issue_Date_Year, count(*) REC_COUNT from NYC_Ticket_Base_tab group by `vehicle body type`,Issue_Date_Year order by count(*) desc") #2015 vehicle_body_type_year_freq_2015<-SparkR::collect(SparkR::filter(vehicle_body_type_year_freq, vehicle_body_type_year_freq$Issue_Date_Year == 2015)) head(vehicle_body_type_year_freq_2015,5) #Top 5 vehicle body type in 2015 # vehicle body type Issue_Date_Year REC_COUNT # 1 SUBN 2015 3245663 # 2 4DSD 2015 2862109 # 3 VAN 2015 1448572 # 4 DELV 2015 734937 # 5 SDN 2015 390372 #2016 vehicle_body_type_year_freq_2016<-SparkR::collect(SparkR::filter(vehicle_body_type_year_freq, vehicle_body_type_year_freq$Issue_Date_Year == 2016)) head(vehicle_body_type_year_freq_2016,5) # vehicle body type Issue_Date_Year REC_COUNT # 1 SUBN 2016 3425471 # 2 4DSD 2016 2888670 # 3 VAN 2016 1403728 # 4 DELV 2016 667754 # 5 SDN 2016 413483 #2017 vehicle_body_type_year_freq_2017<-SparkR::collect(SparkR::filter(vehicle_body_type_year_freq, vehicle_body_type_year_freq$Issue_Date_Year == 2017)) head(vehicle_body_type_year_freq_2017,5) # vehicle body type Issue_Date_Year REC_COUNT # 1 SUBN 2017 1880059 # 2 4DSD 2017 1546064 # 3 VAN 2017 718803 # 4 DELV 2017 351822 # 5 SDN 2017 190760 #Let's visualize these inferences vehicle_body_type_year_freq<-SparkR::collect(SparkR::filter(vehicle_body_type_year_freq, vehicle_body_type_year_freq$REC_COUNT>50000)) plot <- ggplot(vehicle_body_type_year_freq,aes(x = factor(Issue_Date_Year), y = REC_COUNT,col=`vehicle body type`,label=`vehicle body type`)) + geom_point() + geom_label_repel(aes(label = `vehicle body type`), box.padding = 0.35, point.padding = 0.5, segment.color = 'grey50') + theme_classic()+ xlab("violation code") + ylab("REC_COUNT") plot #Year wise SUBN,4DSD,VAN,DELV,SDN are consistant top 5 violating body types #Make vehicle_make_year_freq<-SparkR::sql("select `vehicle make`,Issue_Date_Year, count(*) REC_COUNT from NYC_Ticket_Base_tab group by `vehicle make`,Issue_Date_Year order by count(*) desc") #Overall Analysis head(vehicle_make_year_freq,5) # vehicle make Issue_Date_Year REC_COUNT # 1 FORD 2015 1267693 # 2 FORD 2016 1252931 # 3 TOYOT 2016 1132494 # 4 TOYOT 2015 1065459 # 5 HONDA 2016 997047 #Yearwise Analysis #2015 vehicle_make_year_freq_2015<-SparkR::collect(SparkR::filter(vehicle_make_year_freq, vehicle_make_year_freq$Issue_Date_Year == 2015)) head(vehicle_make_year_freq_2015,5) # vehicle make Issue_Date_Year REC_COUNT # 1 FORD 2015 1267693 # 2 TOYOT 2015 1065459 # 3 HONDA 2015 952177 # 4 NISSA 2015 780454 # 5 CHEVR 2015 748049 #2016 vehicle_make_year_freq_2016<-SparkR::collect(SparkR::filter(vehicle_make_year_freq, vehicle_make_year_freq$Issue_Date_Year == 2016)) head(vehicle_make_year_freq_2016,5) # vehicle make Issue_Date_Year REC_COUNT # 1 FORD 2016 1252931 # 2 TOYOT 2016 1132494 # 3 HONDA 2016 997047 # 4 NISSA 2016 836343 # 5 CHEVR 2016 696322 #2017 vehicle_make_year_freq_2017<-SparkR::collect(SparkR::filter(vehicle_make_year_freq, vehicle_make_year_freq$Issue_Date_Year == 2017)) head(vehicle_make_year_freq_2017,5) # vehicle make Issue_Date_Year REC_COUNT # 1 FORD 2017 635186 # 2 TOYOT 2017 603866 # 3 HONDA 2017 537526 # 4 NISSA 2017 460794 # 5 CHEVR 2017 355164 #Let's visualize these inferences vehicle_make_year_freq<-SparkR::collect(SparkR::filter(vehicle_make_year_freq, vehicle_make_year_freq$REC_COUNT>50000)) plot <- ggplot(vehicle_make_year_freq,aes(x = factor(Issue_Date_Year), y = REC_COUNT,col=`vehicle make`,label=`vehicle make`)) + geom_point() + geom_label_repel(aes(label = `vehicle make`), box.padding = 0.35, point.padding = 0.5, segment.color = 'grey50') + theme_classic()+ xlab("violation code") + ylab("REC_COUNT") plot #Year wise FORD,TOYOT(A),HONDA,NISSA,CHEVR are top 5 brands with violation tickets #Let's perform overall analysis another way # Number of tickets by Vehicle Body Type vbt <- SparkR::count(groupBy(NYC_Ticket_Base, "Vehicle Body Type")) vbt <- SparkR::collect(vbt) vbt <- (arrange(vbt, desc(vbt$count))) head(vbt) # Vehicle Body Type count # 1 SUBN 8551193 # 2 4DSD 7296843 # 3 VAN 3571103 # 4 DELV 1754513 # 5 SDN 994615 # 6 2DSD 667143 # Number of tickets by Vehicle Make vm <- SparkR::count(groupBy(NYC_Ticket_Base, "Vehicle Make")) vm <- SparkR::collect(vm) vm <- (arrange(vm, desc(vm$count))) head(vm) # Vehicle Make count # 1 FORD 3155810 # 2 TOYOT 2801819 # 3 HONDA 2486750 # 4 NISSA 2077591 # 5 CHEVR 1799535 # 6 FRUEH 1022941 #This analysis confirms our earlier findings as well #------------------------------------------------------------------------------------------------------------------------------------------------------ #3.A precinct is a police station that has a certain zone of the city under its command. Find the (5 highest) frequency of tickets for each of the following: #3.1 'Violation Precinct' (this is the precinct of the zone where the violation occurred). #Using this, can you make any insights for parking violations in any specific areas of the city? #Overall Analysis Violation_Precinct_freq<-SparkR::sql("select `Violation Precinct`, count(*) REC_COUNT from NYC_Ticket_Base_tab group by `Violation Precinct` order by count(*) desc") head(Violation_Precinct_freq,6) # Violation Precinct REC_COUNT # 1 0 4448923 # 2 19 1320716 # 3 14 830512 # 4 18 790596 # 5 1 750644 # 6 114 712183 #Let's try alternate way # Number of tickets by Violation Precinct vp <- SparkR::count(groupBy(NYC_Ticket_Base, "Violation Precinct")) vp <- SparkR::collect(vp) vp <- (arrange(vp, desc(vp$count))) head(vp) # Violation Precinct count # 1 0 4448923 # 2 19 1320716 # 3 14 830512 # 4 18 790596 # 5 1 750644 # 6 114 712183 #Year-wise analysis Violation_Precinct_year_freq<-SparkR::sql("select `Violation Precinct`,Issue_Date_Year, count(*) REC_COUNT from NYC_Ticket_Base_tab where `Violation Precinct` !=0 group by `Violation Precinct`,Issue_Date_Year order by count(*) desc") #2015 Violation_Precinct_year_freq_2015<-SparkR::collect(SparkR::filter(Violation_Precinct_year_freq, Violation_Precinct_year_freq$Issue_Date_Year == 2015)) head(Violation_Precinct_year_freq_2015,5) # Violation Precinct Issue_Date_Year REC_COUNT # 1 19 2015 526252 # 2 18 2015 340438 # 3 14 2015 334275 # 4 114 2015 286258 # 5 1 2015 273800 #2016 Violation_Precinct_year_freq_2016<-SparkR::collect(SparkR::filter(Violation_Precinct_year_freq, Violation_Precinct_year_freq$Issue_Date_Year == 2016)) head(Violation_Precinct_year_freq_2016,5) # Violation Precinct Issue_Date_Year REC_COUNT # Violation Precinct Issue_Date_Year REC_COUNT # 1 19 2016 522311 # 2 1 2016 304737 # 3 14 2016 295558 # 4 18 2016 284146 # 5 114 2016 279142 #2017 Violation_Precinct_year_freq_2017<-SparkR::collect(SparkR::filter(Violation_Precinct_year_freq, Violation_Precinct_year_freq$Issue_Date_Year == 2017)) head(Violation_Precinct_year_freq_2017,5) # Violation Precinct Issue_Date_Year REC_COUNT # 1 19 2017 272153 # 2 14 2017 200679 # 3 1 2017 172107 # 4 18 2017 166012 # 5 114 2017 146783 #Let's visualize these inferernces ##Reducing counts of smaller numbers for plotting Violation_Precinct_year_freq<-SparkR::collect(SparkR::filter(Violation_Precinct_year_freq, Violation_Precinct_year_freq$REC_COUNT>100000)) plot <- ggplot(Violation_Precinct_year_freq,aes(x = factor(Issue_Date_Year), y = REC_COUNT,col=`Violation Precinct`,label=`Violation Precinct`)) + geom_point() + geom_label_repel(aes(label = `Violation Precinct`), box.padding = 0.35, point.padding = 0.5, segment.color = 'grey50') + theme_classic()+ xlab("Violation Precinct") + ylab("REC_COUNT") plot # 19,14,18,1,114 are top 5 Violation precincts where violations occur the most with 19 being consistently highest #3.2 'Issuer Precinct' (this is the precinct that issued the ticket) #Here you would have noticed that the dataframe has 'Violating Precinct' or 'Issuing Precinct' as '0'. These are the erroneous entries. #Hence, provide the record for five correct precincts. (Hint: print top six entries after sorting) #Overall Analysis Issuer_Precinct_freq<-SparkR::sql("select `Issuer Precinct`, count(*) REC_COUNT from NYC_Ticket_Base_tab group by `Issuer Precinct` order by count(*) desc") head(Issuer_Precinct_freq,6) # Issuer Precinct REC_COUNT # 1 0 5061166 # 2 19 1288533 # 3 14 811162 # 4 18 771257 # 5 1 731608 # 6 114 699980 #Let's try alternate way # Number of tickets by Issuer Precinct ip <- SparkR::count(groupBy(NYC_Ticket_Base, "Issuer Precinct")) ip <- SparkR::collect(ip) ip <- (arrange(ip, desc(ip$count))) head(ip) # Issuer Precinct count # 1 0 5061166 # 2 19 1288533 # 3 14 811162 # 4 18 771257 # 5 1 731608 # 6 114 699980 #Year-wise analysis Issuer_Precinct_year_freq<-SparkR::sql("select `Issuer Precinct`,Issue_Date_Year, count(*) REC_COUNT from NYC_Ticket_Base_tab where `Issuer Precinct` !=0 group by `Issuer Precinct`,Issue_Date_Year order by count(*) desc") #2015 Issuer_Precinct_year_freq_2015<-SparkR::collect(SparkR::filter(Issuer_Precinct_year_freq, Issuer_Precinct_year_freq$Issue_Date_Year == 2015)) head(Issuer_Precinct_year_freq_2015,5) # Issuer Precinct Issue_Date_Year REC_COUNT # 1 19 2015 513583 # 2 18 2015 334355 # 3 14 2015 325592 # 4 114 2015 282353 # 5 1 2015 268316 #2016 Issuer_Precinct_year_freq_2016<-SparkR::collect(SparkR::filter(Issuer_Precinct_year_freq, Issuer_Precinct_year_freq$Issue_Date_Year == 2016)) head(Issuer_Precinct_year_freq_2016,5) # Issuer Precinct Issue_Date_Year REC_COUNT # Issuer Precinct Issue_Date_Year REC_COUNT # 1 19 2016 510002 # 2 1 2016 296457 # 3 14 2016 287561 # 4 18 2016 276548 # 5 114 2016 274101 #2017 Issuer_Precinct_year_freq_2017<-SparkR::collect(SparkR::filter(Issuer_Precinct_year_freq, Issuer_Precinct_year_freq$Issue_Date_Year == 2017)) head(Issuer_Precinct_year_freq_2017,5) # Issuer Precinct Issue_Date_Year REC_COUNT # 1 19 2017 264948 # 2 14 2017 198009 # 3 1 2017 166835 # 4 18 2017 160354 # 5 114 2017 143526 #Let's visualize these inferernces ##Reducing counts of smaller numbers for plotting Issuer_Precinct_year_freq<-SparkR::collect(SparkR::filter(Issuer_Precinct_year_freq, Issuer_Precinct_year_freq$REC_COUNT>100000)) plot <- ggplot(Issuer_Precinct_year_freq,aes(x = factor(Issue_Date_Year), y = REC_COUNT,col=`Issuer Precinct`,label=`Issuer Precinct`)) + geom_point() + geom_label_repel(aes(label = `Issuer Precinct`), box.padding = 0.35, point.padding = 0.5, segment.color = 'grey50') + theme_classic()+ xlab("Issuer Precinct") + ylab("REC_COUNT") plot # 19,14,18,1,114 are top 5 issuer precincts where violations tickets are issued.The most with 19 being consistently highest # From the counts it appears that not all tickets are issued in the same precinct where violation occured #------------------------------------------------------------------------------------------------------------------------------------------------------ #A4. Find the violation code frequency across three precincts which have issued the most number of tickets - #do these precinct zones have an exceptionally high frequency of certain violation codes? #Are these codes common across precincts? #Hint: You can analyse the three precincts together using the 'union all' attribute in SQL view. #In the SQL view,use the 'where' attribute to filter among three precincts and combine them using 'union all'. #From data analyzed above, #Three prestine with most issued tickets # 19 1372464 # 14 870724 # 18 831708 #Overall analysis Issuer_Precinct_freq<-SparkR::sql("select `Issuer Precinct`,`violation code`, count(*) REC_COUNT from NYC_Ticket_Base_tab where `Issuer Precinct` in (19,14,18) group by `Issuer Precinct`,`violation code` order by `Issuer Precinct`,count(*) desc") Issuer_Precinct_freq_19<-SparkR::collect(SparkR::filter(Issuer_Precinct_freq, Issuer_Precinct_freq$`Issuer Precinct` == 19)) head(Issuer_Precinct_freq_19,5) # Issuer Precinct violation code REC_COUNT # 1 19 38 186785 # 2 19 37 182166 # 3 19 46 177442 # 4 19 14 145387 # 5 19 21 133729 Issuer_Precinct_freq_14<-SparkR::collect(SparkR::filter(Issuer_Precinct_freq, Issuer_Precinct_freq$`Issuer Precinct` == 14)) head(Issuer_Precinct_freq_14,5) # Issuer Precinct violation code REC_COUNT # 1 14 14 166912 # 2 14 69 160205 # 3 14 31 92628 # 4 14 47 66760 # 5 14 42 56359 Issuer_Precinct_freq_18<-SparkR::collect(SparkR::filter(Issuer_Precinct_freq, Issuer_Precinct_freq$`Issuer Precinct` == 18)) head(Issuer_Precinct_freq_18,5) # Issuer Precinct violation code REC_COUNT # 1 18 14 239395 # 2 18 69 107960 # 3 18 47 59707 # 4 18 31 56661 # 5 18 42 38060 #Let's visualize this inferences Issuer_Precinct_freq<-SparkR::collect(SparkR::filter(Issuer_Precinct_freq, Issuer_Precinct_freq$REC_COUNT>1000)) plot <- ggplot(Issuer_Precinct_freq,aes(x = `violation code`, y = REC_COUNT,col=`Issuer Precinct`)) + geom_point() + xlab("violation code") + ylab("REC_COUNT") plot #violation code 14 is consistently high for all three issue prectine location #38,37,46,21 are high for 19 precinct #69,31,47,42 are high for 14 as well as 19 precinct #Year wise analysis Issuer_Precinct_freq_yrly<-SparkR::sql("select Issue_Date_Year,`Issuer Precinct`,`violation code`, count(*) REC_COUNT from NYC_Ticket_Base_tab where `Issuer Precinct` in (19,14,18) group by Issue_Date_Year,`Issuer Precinct`,`violation code` order by Issue_Date_Year,count(*) desc") #2015 Issuer_Precinct_freq_yrly_2015<-SparkR::collect(SparkR::filter(Issuer_Precinct_freq_yrly, Issuer_Precinct_freq_yrly$Issue_Date_Year == 2015)) head(Issuer_Precinct_freq_yrly_2015,15) # Issue_Date_Year Issuer Precinct violation code REC_COUNT # 1 2015 18 14 104232 # 2 2015 19 38 76862 # 3 2015 19 37 71766 # 4 2015 14 69 71005 # 5 2015 14 14 67082 # 6 2015 19 14 59317 # 7 2015 19 46 57692 # 8 2015 19 21 53514 # 9 2015 19 16 52939 # 10 2015 18 69 50557 # 11 2015 14 31 35147 # 12 2015 19 20 28217 # 13 2015 18 47 26188 # 14 2015 18 31 24848 # 15 2015 14 47 24683 #2016 Issuer_Precinct_freq_yrly_2016<-SparkR::collect(SparkR::filter(Issuer_Precinct_freq_yrly, Issuer_Precinct_freq_yrly$Issue_Date_Year == 2016)) head(Issuer_Precinct_freq_yrly_2016,15) # Issue_Date_Year Issuer Precinct violation code REC_COUNT # 1 2016 18 14 85542 # 2 2016 19 37 74379 # 3 2016 19 38 73734 # 4 2016 19 46 72431 # 5 2016 14 69 58999 # 6 2016 19 14 56463 # 7 2016 14 14 55300 # 8 2016 19 21 51844 # 9 2016 19 16 46045 # 10 2016 18 69 37369 # 11 2016 14 31 34980 # 12 2016 19 20 26752 # 13 2016 14 47 23770 # 14 2016 14 42 22608 # 15 2016 18 31 19959 #2017 Issuer_Precinct_freq_yrly_2017<-SparkR::collect(SparkR::filter(Issuer_Precinct_freq_yrly, Issuer_Precinct_freq_yrly$Issue_Date_Year == 2017)) head(Issuer_Precinct_freq_yrly_2017,15) # Issue_Date_Year Issuer Precinct violation code REC_COUNT # 1 2017 18 14 49621 # 2 2017 19 46 47319 # 3 2017 14 14 44530 # 4 2017 19 38 36189 # 5 2017 19 37 36021 # 6 2017 14 69 30201 # 7 2017 19 14 29607 # 8 2017 19 21 28371 # 9 2017 14 31 22501 # 10 2017 18 69 20034 # 11 2017 14 47 18307 # 12 2017 19 20 14601 # 13 2017 18 47 14050 # 14 2017 18 31 11854 # 15 2017 19 40 11380 #14,37,38,46,69,21,31,47,42 are top violation codes occuring across 2015,2016 and 2017 across precincts #------------------------------------------------------------------------------------------------------------------------------------------------------ #5. You'd want to find out the properties of parking violations across different times of the day: #Find a way to deal with missing values, if any. #Hint: Check for the null values using 'isNull' under the SQL. #Also, to remove the null values, check the 'dropna' command in the API documentation. #The Violation Time field is specified in a strange format. #Find a way to make this into a time attribute that you can use to divide into groups. #Divide 24 hours into six equal discrete bins of time. #The intervals you choose are at your discretion. For each of these groups, #find the three most commonly occurring violations. #Hint: Use the CASE-WHEN in SQL view to segregate into bins. #For finding the most commonly occurring violations, #a similar approach can be used as mention in the hint for question 4. #Now, try another direction. For the 3 most commonly occurring violation codes, #find the most common time of the day (in terms of the bins from the previous part) Violation_Time_cnt<-SparkR::sql("select count(*) REC_COUNT from NYC_Ticket_Base_tab where `Violation Time` is null") head(Violation_Time_cnt) #1314 violation time fields are null , lets remove them with dropna NYC_Ticket_Base<-SparkR::dropna(NYC_Ticket_Base,how="all", cols="Violation Time") createOrReplaceTempView(NYC_Ticket_Base, "NYC_Ticket_Base_tab") Violation_Time_cnt<-SparkR::sql("select count(*) REC_COUNT from NYC_Ticket_Base_tab where `Violation Time` is null") head(Violation_Time_cnt) #0 Violation Time records are null head(NYC_Ticket_Base[,"Violation Time"]) # Violation Time # 1 1002A # 2 0820P # 3 0240P # 4 0749A # 5 0848A # 6 1010P # converting these values into army hours NYC_Ticket_Base<-SparkR::sql("select NYC_Ticket_Base_tab.*, substr(`Violation Time`,0,2) + case when substr(`Violation Time`,-1)=='A' then 0 else 12 end Violation_Time_Hour from NYC_Ticket_Base_tab "); createOrReplaceTempView(NYC_Ticket_Base, "NYC_Ticket_Base_tab") head(NYC_Ticket_Base[,"Violation_Time_Hour"]) # Violation_Time_Hour # 1 10 # 2 20 # 3 14 # 4 7 # 5 8 # 6 22 min_max_hour<-SparkR::sql("select max(Violation_Time_Hour) max_hour, min(Violation_Time_Hour) min_hour from NYC_Ticket_Base_tab") head(min_max_hour) #max_hour min_hour # 99 0 #There seems to be invalid values in the data garbage_hour<-SparkR::sql("select count(*) from NYC_Ticket_Base_tab where Violation_Time_Hour>23") head(garbage_hour) #2343248 records 2343248/nrow(NYC_Ticket_Base) ##9.2% records contain invalid data, lets remove them NYC_Ticket_Base<-SparkR::sql("select * from NYC_Ticket_Base_tab where Violation_Time_Hour<=23") createOrReplaceTempView(NYC_Ticket_Base, "NYC_Ticket_Base_tab") #validate the data min_max_hour<-SparkR::sql("select max(Violation_Time_Hour) max_hour, min(Violation_Time_Hour) min_hour from NYC_Ticket_Base_tab") head(min_max_hour) #max_hour min_hour # 23 0 #hours ranging from 00-23 - army time #Creating 6 buckets 4 hours each NYC_Ticket_Base<-SparkR::sql("select NYC_Ticket_Base_tab.*, case when Violation_Time_Hour between 0 and 3 then 1 when Violation_Time_Hour between 4 and 7 then 2 when Violation_Time_Hour between 8 and 11 then 3 when Violation_Time_Hour between 12 and 15 then 4 when Violation_Time_Hour between 16 and 19 then 5 when Violation_Time_Hour between 20 and 23 then 6 end Violation_Time_bucket from NYC_Ticket_Base_tab") createOrReplaceTempView(NYC_Ticket_Base, "NYC_Ticket_Base_tab") Violation_Time_bucket_Codes<-SparkR::sql("select Violation_Time_bucket,`violation code`, count(*) REC_COUNT from NYC_Ticket_Base_tab group by Violation_Time_bucket,`violation code` ") head(Violation_Time_bucket_Codes) # Violation_Time_bucket violation code REC_COUNT # 1 6 19 42520 # 2 4 54 5 # 3 2 46 74526 # 4 1 12 14 # 5 5 93 8 # 6 4 89 2222 #Time of the day for most 3 commonly occuring codes commoncode_violation_time_buckets <- SparkR::collect(Violation_Time_bucket_Codes) commoncode_violation_time_buckets<- arrange(commoncode_violation_time_buckets, desc(commoncode_violation_time_buckets$REC_COUNT)) head(commoncode_violation_time_buckets,3) #Time bucket for 3 top codes # Violation_Time_bucket violation code REC_COUNT # 1 3 21 2825191 # 2 3 36 1483548 # 3 3 38 928803 #It appears that the top violation codes occur in time bucket 3 #filtering for higher values for plotting Violation_Time_bucket_Codes<-SparkR::collect(SparkR::filter(Violation_Time_bucket_Codes, Violation_Time_bucket_Codes$REC_COUNT>50000)) plot <- ggplot(Violation_Time_bucket_Codes,aes(x = Violation_Time_bucket, y = REC_COUNT,col=`violation code`,label=`violation code`)) + geom_point() + geom_label_repel(aes(label = `violation code`), box.padding = 0.35, point.padding = 0.5, segment.color = 'grey50') + theme_classic()+ xlab("Time Bucket") + ylab("REC_COUNT") plot # for 0-3 21,40,78 are the top violation code # for 4-7 14,21,40 # for 8-11 21, 36,38 # for 12-15 38,36,37 # for 15-19 38, 37,14 # for 19-23 14, 38, 7 #36-38 , 40 , 14, 21 are mostly common across time buckets #Year wise analysis #Query based Violation_Time_bucket_Year_Codes<-SparkR::sql(" select * from (select Year_time_bucket, `violation code`,REC_COUNT , dense_rank() OVER (PARTITION BY Year_time_bucket ORDER BY REC_COUNT DESC) as rank from ( select concat(Violation_Time_bucket,'_',Issue_Date_Year) Year_time_bucket, `violation code`, count(*) REC_COUNT from NYC_Ticket_Base_tab group by concat(Violation_Time_bucket,'_',Issue_Date_Year), `violation code`)T)V where rank<=3 ") head(Violation_Time_bucket_Year_Codes,nrow(Violation_Time_bucket_Year_Codes)) # Year_time_bucket violation code REC_COUNT rank # 1 3_2016 21 1099915 1 # 2 3_2016 36 686587 2 # 3 3_2016 38 355058 3 # 4 5_2016 38 205800 1 # 5 5_2016 37 155279 2 # 6 5_2016 14 131121 3 # 7 4_2017 38 184105 1 # 8 4_2017 36 184050 2 # 9 4_2017 37 130466 3 # 10 4_2015 38 367090 1 # 11 4_2015 37 289704 2 # 12 4_2015 36 282136 3 # 13 6_2015 7 59612 1 # 14 6_2015 38 55596 2 # 15 6_2015 40 42432 3 # 16 1_2015 21 59280 1 # 17 1_2015 40 33684 2 # 18 1_2015 78 27319 3 # 19 2_2017 14 73567 1 # 20 2_2017 40 60397 2 # 21 2_2017 21 56737 3 # 22 3_2017 21 592259 1 # 23 3_2017 36 347650 2 # 24 3_2017 38 175693 3 # 25 1_2017 21 33956 1 # 26 1_2017 40 23216 2 # 27 1_2017 14 13866 3 # 28 2_2015 14 131702 1 # 29 2_2015 21 101552 2 # 30 2_2015 40 86878 3 # 31 2_2016 14 131765 1 # 32 2_2016 21 107374 2 # 33 2_2016 40 93228 3 # 34 5_2017 38 102533 1 # 35 5_2017 14 75000 2 # 36 5_2017 37 70223 3 # 37 4_2016 36 382783 1 # 38 4_2016 38 348430 2 # 39 4_2016 37 278588 3 # 40 6_2017 7 26238 1 # 41 6_2017 40 22011 2 # 42 6_2017 14 20778 3 # 43 3_2015 21 1133017 1 # 44 3_2015 36 449311 2 # 45 3_2015 38 398052 3 # 46 1_2016 21 66975 1 # 47 1_2016 40 38369 2 # 48 1_2016 78 27160 3 # 49 5_2015 38 196455 1 # 50 5_2015 37 151789 2 # 51 5_2015 14 130146 3 # 52 6_2016 7 59420 1 # 53 6_2016 38 47491 2 # 54 6_2016 14 42775 3 #Plot based plot <- ggplot(SparkR::collect(Violation_Time_bucket_Year_Codes), aes(x = Year_time_bucket , y = REC_COUNT,col=`violation code`,label=`violation code`)) + geom_point() + geom_label_repel(aes(label = `violation code`), box.padding = 0.35, point.padding = 0.5, segment.color = 'grey50') + theme_classic()+ xlab("Year_time_bucket") + ylab("REC_COUNT") plot #Please zoom the plot for better viewing # We can see that 3rd bucket that is 8-11 morning hrs the counts are really high for all three years # 21,36,38 being the key violation codes, atlthough 2017 has seen improment overall and these codes are repeated here as well # For the first window 21, 40 are common one and two, and 78 is replaced by 14 in 2017 post 2015 & 2016 # For 2nd window 14,21,40 are first three across years # For 4th window 36,37,38 are interchanging across three years # 38 is common top for 5th window # 37 is 2nd for first two year and 3rd for the last one # 7 for 1st and 14 other two yearas is filling up the remaing positions # 7 is common 1st in the 6th bucket # 38,14 40 are filling up the other postions across the years #So in a day violations start with codes 21,36,38 as time passed 14, 40 appears then afterwords again 36,37,38 dominate #follwed by introduction of 7 and finishing with 38,14,40 # All 3 top violations happened in 3rd time window in 2016 # Maximum violation in code 36,38 happened in 4th time window in 2015 and 2017 # Maximum violation in code 21 happened in 1st time window in 2015 # Maximum violation in code 21 happened in 2nd time window in 2017 #------------------------------------------------------------------------------------------------------------------------------------------------------ #6.Let's try and find some seasonality in this data #First, divide the year into some number of seasons, #and find frequencies of tickets for each season. #(Hint: Use Issue Date to segregate into seasons) #Then, find the three most common violations for each of these seasons. #(Hint: A similar approach can be used as mention in the hint for question 4.) #lets divide the year into 4 set of month , 1-3,4-6,7-9 & 1-12 NYC_Ticket_Base$Issue_Date_Month<-SparkR::month(NYC_Ticket_Base$"Issue Date") createOrReplaceTempView(NYC_Ticket_Base, "NYC_Ticket_Base_tab") #Overall analysis NYC_Ticket_Base<-SparkR::sql("select NYC_Ticket_Base_tab.*, case when Issue_Date_Month between 1 and 3 then 1 when Issue_Date_Month between 4 and 6 then 2 when Issue_Date_Month between 7 and 9 then 3 when Issue_Date_Month between 10 and 12 then 4 end Month_bucket from NYC_Ticket_Base_tab") createOrReplaceTempView(NYC_Ticket_Base, "NYC_Ticket_Base_tab") # Season wise spread across the three years Month_Wise<-SparkR::sql("select Month_bucket,count(*) as REC_COUNT from NYC_Ticket_Base_tab group by Month_bucket") Month_Wise <- SparkR::arrange(Month_Wise, Month_Wise$Month_bucket) head(Month_Wise,4) # Month_bucket REC_COUNT 1 6489169 2 7108573 3 4692555 4 4897743 # Season wise violation code analysis across all the three years Month_Wise_Violation_1 <- filter(Month_Wise_Violation, Month_Wise_Violation$Month_bucket == 1) head(arrange(Month_Wise_Violation_1, desc(Month_Wise_Violation_1$REC_COUNT))) #Month_bucket violation code REC_COUNT # 1 21 824866 # 1 38 759550 # 1 36 688665 # 1 14 571222 # 1 37 404433 # 1 20 397134 Month_Wise_Violation_2 <- filter(Month_Wise_Violation, Month_Wise_Violation$Month_bucket == 2) head(arrange(Month_Wise_Violation_2, desc(Month_Wise_Violation_2$REC_COUNT))) # Month_bucket violation code REC_COUNT # 2 21 1033621 # 2 36 740325 # 2 38 727504 # 2 14 632942 # 2 37 427841 # 2 20 418218 Month_Wise_Violation_3 <- filter(Month_Wise_Violation, Month_Wise_Violation$Month_bucket == 3) head(arrange(Month_Wise_Violation_3, desc(Month_Wise_Violation_3$REC_COUNT))) # Month_bucket violation code REC_COUNT # 3 21 713757 # 3 38 493050 # 3 14 405934 # 3 36 364038 # 3 37 293070 # 3 20 277801 Month_Wise_Violation_4 <- filter(Month_Wise_Violation, Month_Wise_Violation$Month_bucket == 4) head(arrange(Month_Wise_Violation_4, desc(Month_Wise_Violation_4$REC_COUNT))) # Month_bucket violation code REC_COUNT # 4 36 751179 # 4 21 699299 # 4 38 482932 # 4 14 392281 # 4 20 277341 # 4 37 270932 Month_Wise_Violation<-SparkR::sql("select Month_bucket,`violation code` ,count(*) REC_COUNT from NYC_Ticket_Base_tab group by Month_bucket,`violation code` ") Month_Wise_Violation<-SparkR::collect(SparkR::filter(Month_Wise_Violation, Month_Wise_Violation$REC_COUNT>50000)) plot <- ggplot(Month_Wise_Violation,aes(x = Month_bucket, y = REC_COUNT,col=`violation code`,label=`violation code`)) + geom_point() + geom_label_repel(aes(label = `violation code`), box.padding = 0.35, point.padding = 0.5, segment.color = 'grey50') + theme_classic()+ xlab("Season") + ylab("REC_COUNT") plot #all seasons/months have 21,38,36, 14 as most common in differnt order #Year wise analysis Month_year_Wise_Violation<-SparkR::sql("select * from ( select Month_year_Bucket,`violation code`,REC_COUNT, dense_rank() OVER (PARTITION BY Month_year_Bucket ORDER BY REC_COUNT DESC) as Rank from ( select concat(Month_bucket,'_',Issue_Date_Year) Month_year_Bucket, `violation code` ,count(*) REC_COUNT from NYC_Ticket_Base_tab group by concat(Month_bucket,'_',Issue_Date_Year),`violation code` ) T) V where Rank<=3 ") head(Month_year_Wise_Violation,nrow(Month_year_Wise_Violation)) #Yearly Month/Seasonwise violation codes and records # Month_year_Bucket violation code REC_COUNT Rank # 1 3_2016 21 345564 1 # 2 3_2016 38 220875 2 # 3 3_2016 36 204171 3 # 4 4_2017 46 209 1 # 5 4_2017 40 132 2 # 6 4_2017 21 116 3 # 7 4_2015 21 388242 1 # 8 4_2015 36 375921 2 # 9 4_2015 38 245213 3 # 10 1_2015 38 226022 1 # 11 1_2015 21 175707 2 # 12 1_2015 14 159197 3 # 13 2_2017 21 354076 1 # 14 2_2017 36 266183 2 # 15 2_2017 14 232710 3 # 16 3_2017 21 243 1 # 17 3_2017 46 202 2 # 18 3_2017 40 112 3 # 19 1_2017 21 333263 1 # 20 1_2017 36 293799 2 # 21 1_2017 38 256831 3 # 22 2_2015 21 369613 1 # 23 2_2015 38 276722 2 # 24 2_2015 14 214645 3 # 25 2_2016 21 309932 1 # 26 2_2016 36 285092 2 # 27 2_2016 38 223546 3 # 28 4_2016 36 375258 1 # 29 4_2016 21 310941 2 # 30 4_2016 38 237713 3 # 31 3_2015 21 367950 1 # 32 3_2015 38 272166 2 # 33 3_2015 14 217383 3 # 34 1_2016 21 315896 1 # 35 1_2016 36 294616 2 # 36 1_2016 38 276697 3 #Let's visualize this inference plot <- ggplot(SparkR::collect(Month_year_Wise_Violation),aes(x = Month_year_Bucket, y = REC_COUNT,col=`violation code`,label=`violation code`)) + geom_point() + geom_label_repel(aes(label = `violation code`), box.padding = 0.35, point.padding = 0.5, segment.color = 'grey50') + theme_classic()+ xlab("Month_year_Bucket") + ylab("REC_COUNT") plot # #Violation codes analysis #21 38 are common for first 3 months across years #14 for first year followed by 36 for remaining months are the norm # #21 is the category for 2nd set of month in all the three years #with differnce being remarkably high in 2015 & 2017 # 38,36,14 share 2 slot each across the years # #Year 2017 have very low counts for this month # 21,38 again feature as dominating the for years 2015 and 2016 # 14, 36 are in 3rd ranking for these two years respectively # # 46 and 40 are the odd one for 2017 #2017 observation for 4th set is same as for 3rd month #21,36,38 are the are interchangely presnt for 2015 and 2016 #------------------------------------------------------------------------------------------------------------------------------------------------------ #7.The fines collected from all the parking violation constitute a revenue source for the NYC police department. Let’s take an example of estimating that for the three most commonly occurring codes. #Find total occurrences of the three most common violation codes #Then, visit the website: # http://www1.nyc.gov/site/finance/vehicles/services-violation-codes.page #It lists the fines associated with different violation codes. #They're divided into two categories, one for the highest-density locations of the city, #the other for the rest of the city. For simplicity, take an average of the two. #Using this information, find the total amount collected for the three violation codes with maximum tickets. #State the code which has the highest total collection. #What can you intuitively infer from these findings? TOP_3_violation_code<-SparkR::sql("select `violation code`,count(*) REC_COUNT from NYC_Ticket_Base_tab group by `violation code` order by count(*) desc limit 3") head(TOP_3_violation_code,3) #violation code REC_COUNT Manhattan_fine all_others avg_fine Definition #21 3271543 65 45 55 Street Cleaning: No parking where parking is not allowed by sign, street marking or traffic control device. #36 2544207 50 50 50 Exceeding the posted speed limit in or near a designated school zone. #38 2463036 65 35 50 (38) Failing to show a receipt or tag in the windshield.Drivers get a 5-minute grace period past the expired time on parking meter receipts. #Total amount collected for top 3 violations - calculation #Violation code 21 3271543*55 #179934865 #Violation code 36 2544207*50 #127210350 #Violation code 38 2463036*50 #123151800 #Parking related violation are most common type and are a good source of revenue #Year wise analysis TOP_3_violation_code_year<-SparkR::sql("select Issue_Date_Year,`violation code`,count(*) REC_COUNT from NYC_Ticket_Base_tab group by Issue_Date_Year,`violation code` order by count(*) desc") #2015 TOP_3_violation_code_year_2015<-SparkR::collect(SparkR::filter(TOP_3_violation_code_year, TOP_3_violation_code_year$Issue_Date_Year == 2015)) head(TOP_3_violation_code_year_2015,3) #Top 3 violation codes in 2015 # Issue_Date_Year violation code REC_COUNT # 1 2015 21 1301512 # 2 2015 38 1020123 # 3 2015 36 825088 #2016 TOP_3_violation_code_year_2016<-SparkR::collect(SparkR::filter(TOP_3_violation_code_year, TOP_3_violation_code_year$Issue_Date_Year == 2016)) head(TOP_3_violation_code_year_2016,3) #Top 3 violation codes in 2016 # Issue_Date_Year violation code REC_COUNT # 1 2016 21 1282333 # 2 2016 36 1159137 # 3 2016 38 958831 #2017 TOP_3_violation_code_year_2017<-SparkR::collect(SparkR::filter(TOP_3_violation_code_year, TOP_3_violation_code_year$Issue_Date_Year == 2017)) head(TOP_3_violation_code_year_2017,3) #Top 3 violation codes in 2017 # Issue_Date_Year violation code REC_COUNT # 1 2017 21 687698 # 2 2017 36 559982 # 3 2017 38 484082 #Violation codes 21,36,38 are common across all 3 years #Let's visualize these inferernces ##Reducing counts of smaller numbers for plotting TOP_3_violation_code_year<-SparkR::collect(SparkR::filter(TOP_3_violation_code_year, TOP_3_violation_code_year$REC_COUNT>100000)) plot <- ggplot(TOP_3_violation_code_year,aes(x = Issue_Date_Year, y = REC_COUNT,col=`violation code`,label=`violation code`)) + geom_point() + geom_label_repel(aes(label = `violation code`), box.padding = 0.35, point.padding = 0.5, segment.color = 'grey50') + theme_classic()+ xlab("Issue_Date_Year") + ylab("REC_COUNT") plot #21,36,38 are top violation codes across the years #Yearly amounts #2015-21 1301512*55 #71583160 #2016-21 1282333*55 #70528315 #2017-21 687698*55 #37823390 #2015-36 825088*50 #41254400 #2016-36 1159137*50 #57956850 #2016-36 559982*50 #27999100 #2015-38 1020123*50 #51006150 #2016-38 958831*50 #47941550 #2017-38 484082*50 #24204100 #2015+2016+2017-total amount collected for 3 top violation codes #Violation code 21 (1301512+1282333+687698)*55 #179934865 #Violation code 36 (825088+1159137+559982)*50 #127210350 #Violation code 38 (1020123+958831+484082)*50 #123151800 #Parking related violation are most common violation type and are a good source of revenue ####################################################################################################################################################### #7.Closure ####################################################################################################################################################### sparkR.stop()

990551a74656fb8d792fca617d1deb36cd1acaff

478dff15dbb67b960d4386bf393ec289ddd82b6f

/plot2.R

84c9978cb54e54402e24916d137fb6777685ddd9

[]

no_license

panzerfauster/ExData_Plotting1

1ac21596542a21fc6e4a710121df573be9634fb4

a4dcef416d5c77ff28ba72ef84197461d9447c12

refs/heads/master

2021-01-17T23:40:06.477457

2015-09-13T18:55:23

42,335,710

0

null

2015-09-11T22:47:45

null

ISO-8859-2

R

false

899

r

plot2.R

# Data Science Specialization # Course 4: Exploratory Data Analysis # Course Project 1: Consumption Plots # Fausto Martín López ### This code generates the plots required for Course Project 1 in the current Working Directory. ## Read and process the source file colClasses=c("character", "character", rep("numeric", 7)) file <- read.table(file="household_power_consumption.txt", header=TRUE, sep=";", quote="", na.strings="?", colClasses=colClasses) # Filter the dates as instructed file <- file[as.Date(file$Date, "%d/%m/%Y")=="2007-02-01"|as.Date(file$Date, "%d/%m/%Y")=="2007-02-02",] # Merge the Date and Time columns into a timestamp file$DateTime <- strptime(paste(file$Date, file$Time), "%d/%m/%Y %T") file <- file[,c(10,3:9)] ## Create the plot # Plot 2 png("plot2.png") plot(file$DateTime, file$Global_active_power, type="l", xlab="", ylab="Global Active Power (kilowatts)") dev.off()

8da034814cb5f5debe9d58bc4ba3277a87a6f322

277dafa77508edd5d298730aacae4866d8587d32

/ensm.R

c1f1df4a4b164db7a27a70c9da9286f866b2e55d

[]

no_license

Allisterh/NaturalRate_ensm

ec1bac3bf9a7b41af5e2080ad362e1246aae6a2d

ce42313ceb6fe29f61c633c6f3ee5e40e2d1cb52

refs/heads/main

2023-08-26T10:02:44.443160

2021-11-05T10:05:55

null

0

null

UTF-8

R

false

1,461

r

ensm.R

## extract file df <- read.csv("treasurynotes12311993.csv") ## convert maturity to standard date format df <- transform(df, "Maturity" = as.Date(as.character(df[,1]), "%Y%m%d")) ## calculate ttm = years to maturity df$ttm <- (as.numeric(df[,1])-as.numeric(as.Date("1993-12-31")))/365 ##determine coupon.no = number of coupon payments df$coupon.no <- floor((df[,5]-0.01)*2)+1 ##set parameters and ensm function to calculate discount factor Z(0,T) p <- c(0.0687, -0.0422, -0.2399, 0.2116, 0.9652, 0.8825) z.ensm <- function(ttm) { ifelse(ttm==0, 0, exp(-(p[1] + (p[2]+p[3])*(1-exp(-ttm/p[5]))/(ttm/p[5]) - p[3]*exp(-ttm/p[5]) + p[4]*( (1-exp(-ttm/p[6]))/(ttm/p[6]) - exp(-ttm/p[6])))*ttm)) } #set a function to calculate pv of coupon payments coupon.price <- function(ttm1, c1, n) { result <- 0 for(k in n) { l <- ttm1 - 0.5*(k-1) result <- result + (c1/2)*z.ensm(l) if(l < 0.5) { break } } return(result) } #create output = price table price.list <- matrix(0, 224,5, dimnames = list(1:224, c("coupon.pv", "par.pv", "price.model", "price.gross", "price.diff"))) for (i in 1:224) { price.list[i,1] <- coupon.price(df[i,5], df[i,2], c(1:df[i,6])) price.list[i,2] <- 100*z.ensm(df[i,5]) price.list[i,3] <- price.list[i,1]+price.list[i,2] price.list[i,4] <- df[i,3] + (1 - (df[i,5]*2 - floor(df[i,5]*2)))*df[i,2]/2 price.list[i,5] <- price.list[i,4] - price.list[i,3] }

ac5646c0c9278891f328e9efdd931f50d9007449

67426b6f11131a696dfcd535183a722772a182d8

/Simulation/predict.R

6b71785b798cd2923f00e76311039f1ce8579e88

[ "MIT" ]

permissive

jingeyu/CSSN_data_code

700c93eeeac4bba23287d7d0630bc89deb7364d3

e54c5820d3df3a1e46a631f41e83c945ebbffd12

refs/heads/main

2023-03-19T21:41:42.355100

2021-03-06T12:47:24

331,542,553

0

null

UTF-8

R

false

2,858

r

predict.R

############################################################### ######################### Prediction ########################## ############################################################### rm(list = ls()) library(CholWishart) library(MASS) set.seed(20201231) # only 20201205 load(paste0("RData/", 20201205, "_Sigma.RData")) load(paste0("RData/", 20201205, "_Sim_network.RData")) load("RData/Result_20201205_0.1_70.RData") Corr.true[abs(Corr.true) != 0] <- 1 Sparse.Corr[Sparse.Corr != 0] <- 1 G <- nrow(X) n <- ncol(X) # set number of missing genes miss.num <- 50 # generate coordinates of missing genes miss.x <- runif(miss.num, 0, 750) miss.y <- runif(miss.num, 0, 1000) # whether there coordinates in cell.info already sum(miss.x %in% cell.info[,2]) sum(miss.y %in% cell.info[,3]) miss.indx <- cbind(miss.x, miss.y) # Square neighborhood radius r <- 80 NeiFind <- function(miss.indx){ nei.indx <- which(abs(cell.info[,2] - miss.indx[1]) < r & abs(cell.info[,3] - miss.indx[2]) < r) cell.type.nei <- cell.type[nei.indx] return(cbind(nei.indx, cell.type.nei)) } cell.type <- cell.info[,1] ExpSigma <- function(miss.indx, r, nu.i){ nei.mat <- data.frame(NeiFind(i)) colnames(nei.mat) <- c("cell.index", "cell.type") ni <- nrow(nei.mat) if(ni == 0){ Lambda.i <- (nu.i - G - 1) * Sigma.k[,,cell.type[i]] }else{ cell.label <- as.integer(names(table(nei.mat$cell.type))) nei.nk <- as.numeric(table(nei.mat$cell.type)) weight <- nei.nk / ni tmp <- 0 for(j in 1:length(cell.label)){ tmp <- tmp + Sigma.k[,,cell.label[j]] * weight[j] } Lambda.i <- (nu.i - G - 1) * tmp } } nu <- rep(G + 50, miss.num) Sigma.miss <- array(NA, dim = c(G, G, miss.num)) X.miss <- matrix(NA, G, miss.num) Lambda.miss <- array(NA, dim = c(G, G, miss.num)) c.thre <- 0.5 Corr.miss <- array(NA, dim = c(G, G, miss.num)) for(i in 1:miss.num){ Lambda.miss[,, i] <- ExpSigma(i, r, nu[i]) Sigma.miss[,, i] <- rInvWishart(1, nu[i], Lambda.miss[,, i])[,,1] Sigma.miss[,, i][abs(Sigma.miss[,, i]) < c.thre] <- 0 # ensure Sigma_i are positive definite diag(Sigma.miss[,,i]) <- diag(Sigma.miss[,,i]) + 5 Corr.miss[,,i] <- diag(diag(Sigma.miss[,,i])^(-0.5)) %*% Sigma.miss[,,i] %*% diag(diag(Sigma.miss[,,i])^(-0.5)) X.miss[,i] <- mvrnorm(1, mu = rep(0, G), Sigma = Sigma.miss[,,i]) } Corr.miss[Corr.miss != 0] <- 1 #-------- Predictions of missing cells-------- est.miss <- array(NA, dim = c(G, G, miss.num)) for(i in 1:miss.num){ miss.nei <- NeiFind(miss.indx[i,]) tmp <- Sparse.Corr[,, miss.nei[,1]] tmp1 <- apply(tmp, 1:2, mean) tmp1[tmp1 < 0.5] <- 0 tmp1[tmp1 >= 0.5] <- 1 est.miss[,, i] <- tmp1 } pre.error <- rep(0, miss.num) for(i in 1:miss.num){ pre.error[i] <- sum(abs(est.miss[,,i][upper.tri(est.miss[,,i])] - Corr.miss[,,i][upper.tri(Corr.miss[,,i])])) } sum(pre.error) / miss.num # 347.84

0c422c8d9b383fbeafd5877d9add45d3944ec2fa

02b178b7ebb101940d6ede02b10c52dec501dcd6

/microarray/UseRMA.R

cce12db544bae178318068cbf7f0e0e5cbae22e4

[ "MIT" ]

permissive

radio1988/bpipes

21ea7c124f1bd962afe32644c445da3bb7a7d177

0aceb97070210c2361adb45ee0040b6aa5be771b

refs/heads/master

2023-08-24T12:40:19.129216

2023-08-24T00:49:42

140,731,030

3

2

null

UTF-8

R

false

3,111

r

UseRMA.R

#Generate csv file with readaffy and RMA #By Lihua Julie Zhu #on December 7th 2007 #using .CEL file rm(list=ls()) library(affy) setwd("~/Documents/ConsultingActivities/MicroarrayExp/MarianWalhout"); require(affy) Data = ReadAffy(celfile.path="embryo") #Data<-ReadAffy(filenames=targets$FileName,celfile.path="CEL"); #par(mfrow=c(1,2)); temp = unlist(strsplit(sampleNames(Data), "\\.")) sampleNames(Data) = cbind(temp[1],temp[3], temp[5], temp[7], temp[9], temp[11]) boxplot(Data, col=c(2,3,4,5,6,7)) #boxplot(Data, col=c(1,2,3,4)); eset<-rma(Data); write.exprs(eset, file="dataRMA.txt"); pcNorm <-read.table("dataRMA.txt", header=TRUE, sep="\t", dec="."); colnames(pcNorm)[1] = "Probe" sampleNames(Data) slotNames(Data) library("simpleaffy") Data.qc <- qc(Data) avbg(Data.qc) #comparable #Data.qc <- qc.affy(Data,normalised=NULL,tau=0.015,logged=TRUE, cdfn=cleancdfname(cdfName(Data))) #Data.qc <- qc.affy(Data,normalised=NULL,tau=0.015,logged=TRUE, cdfn=cdfName(Data)) #scaling factor sfs(Data.qc) #comparable percent.present(Data.qc) # comparable ratios(Data.qc)[,1:2] # <3 spikeInProbes(Data.qc) ## Normalization of the data using MAS5.0 eset.mas5 <- mas5(Data) ##################### Get the P/A call info ############## APInfo <- mas5calls(Data) #exprs2excel(APInfo, file="Results/dataMas5_PresentCall.csv") #setwd('./Results') #exprs2excel(eset.mas5, file="dataMas5.csv") #write.exprs(eset.mas5, file="Results/dataMas5.csv") slotNames(APInfo) present.call <- exprs(APInfo) colnames(present.call) = paste("PresentCall",colnames(present.call), sep="."); colnames(present.call) #boxplot(pcNorm[,2:dim(pcNorm)[2]], col=c(rep(2,4),rep(3,3),rep(4,4),rep(5,4)),range=0); boxplot(pcNorm[,2:dim(pcNorm)[2]], col=c(2,3,4,5,6,7),range=0); #par(mfrow=c(2,2)); #image(Data); #geneIDS <- "need to put a list of ids in" library(annaffy) annotation(eset) #Symbol = aafSymbol(geneIDs, "zebrafish") ##########################################The following is for exploring purpose############################# deg <- AffyRNAdeg(Data) plotAffyRNAdeg(deg,col=c(2,3,4,5,6,7)) #summaryAffyRNAdeg(deg) deg$sample.names legend(5, 10, deg$sample.names, pch = rep(16, 6), col=c(2,3,4,5,6,7)) #legend(0,46,c("LL_Drosophila_2_1","LL_Drosophila_2_2","LL_Drosophila_2_3","LL_Drosophila_2_4"),pch=rep(16,4),col=c(2,3,4,5)) probeNames(Data)[1:10] dim(pm(Data)); pm(Data)[1:10,] dim(mm(Data)); dim(intensity(Data)); #intensity for a given probe of the same cdf type across all chips geneNames(Data)[1:10] prenorm<- cbind(as.character(probeNames(Data)),pm(Data)); write.table(prenorm, file="preNorm_PM.csv", sep=","); normPM <- normalize(Data, method="quantiles") dim(pm(normPM)) qqplot(pm(Data)[,1],pm(Data)[,2]); qqplot(pm(normPM)[,1], pm(normPM)[,2]); qqplot(pm(normPM)[,1],pm(normPM)[,3]) qqplot(pm(Data)[,1],pm(Data)[,3]) postnorm <- cbind(as.character(probeNames(Data)),pm(normPM)); write.table(postPM, file="postNorm_PM.csv", sep=","); #write.exprs(eset, file="StatRMA.csv");

4296390d3edf80ac0660324e31ffa054e64a544c

457c6af00135a67a0c7969dba0348214a26b4335

/plot2.R

0a8de962e0d4caec2e2b70025d9bf8e98ed90ced

[]

no_license

hiicharles/ExData_Plotting1

912f46b2ac0bd984c5fe78a6cdb134f6bb279482

815ca53b5742c54f9da1ba3554608339e3b5d03b

refs/heads/master

2020-12-14T08:57:14.561822

2014-12-07T17:59:13

null

0

null

UTF-8

R

false

1,227

r

plot2.R

## Exploratory Data Analysis (exdata-016) ## Course Project 1 ## By hiicharles@gmail.com ## plot2.R ## If you want to test, change the filePath. setwd("~/Development/data/exdata-016/") filePath <- "~/Development/data/exdata-016/household_power_consumption.txt" ## Read file. ## Replace ? with NA. ## 2075259 obs data <- read.table(file = filePath, header = TRUE, sep = ";", na.strings = "?") ## Only want data with date "1/2/2007" and "2/2/2007" ## 1/2/2007 - 1440 observations ## 2/2/2007 - 1440 observations sub_data <- data[ data$Date %in% c("1/2/2007", "2/2/2007"), ] ## Remove data to free up memory rm(data) ## Add a column Date1 and Time1 ## Date1 - "2007-02-01" of class Date ## Time1 - "2007-02-01 00:00:00" of class POSIXct sub_data$Date1 <- as.Date(x = sub_data$Date, format = "%d/%m/%Y") sub_data$Time1 <- as.POSIXct(x = paste(sub_data$Date, sub_data$Time, sep = " "), format = "%d/%m/%Y %H:%M:%S") ## Graphic File Device to PNG png(filename = "plot2.png", width = 480, height = 480, units = "px" ) ## Generate plot plot(x = sub_data$Time1, y = sub_data$Global_active_power, type="l", xlab = "", ylab = "Global Active Power (kilowatts)") ## Close device dev.off()

df9db2dbb308f6b3471d20ca1aa5e4471897d8ed

a44a3f17ac568ae03e4a23121cda5e4ab9bfa275

/Assignemnt1.R

1a3feb29f852e666d03f1780b311bc8abbcf5f53

[]

no_license

y437li/AI_Market_assignment

f1673ed9749cfe7705b49b6d74d65c37673042ed

7ecfb6790eff63f52870e8187cdea80617865e21

refs/heads/master

2022-11-17T07:14:44.228196

2020-07-19T14:21:18

279,747,057

0

null

UTF-8

R

false

7,118

r

Assignemnt1.R

##Load data from local csv file path = '/Users/yangli/OneDrive/MMAI/MMAI831/AIOS1_adv_sales.csv' #path = 'C:/Users/y437l/OneDrive/MMAI/MMAI831/AIOS1_adv_sales.csv' data <- read.csv(file =path) #drop index column data <- subset(data, select = -c(X)) ##No missing data #basic descriptive statistics summary(data[,c(1:6)]) data[,1:6]<- scale(data[,1:6]) #check data suitability #generalized pair grapjs to check for bivariate correlations library(gpairs) gpairs(data[,c(1:6)]) ###seems like there is no correlation relationship between independent variables library(corrplot) corrplot(cor(data[,c(1:6)]), method = "color", type="full", addgrid.col = "red", addshade = "positive", addCoef.col = "black") ##The goal of a sales driver analysis is to discover relationships between ##the sale volume with features of the the price and number of stores for the ## product and different type of advertisements. ###split the data train_end <- floor(0.75*nrow(data)) test_star <- train_end+1 train_data <- data[0:train_end,] test_data <- data[test_star:nrow(data),] ##Fitting the model ##Start with simple linear regression model1<-lm(sales~price,data=train_data) summary(model1) #predict test data model1_test_result <- predict(model1, newdata=test_data) TSS1 <- sum((test_data$sales-mean(test_data$sales))^2) RSS1 <- sum((test_data$sales-model1_test_result)^2) test_R_squared1 = 1 - (RSS1/TSS1) ########################### #train R squared:0.06081 #test R squared:0.04407 ########################### ##Multiple linear regression with two factors library(lmtest) model2<-lm(sales~price+store,data=train_data) ##Anova table,F-test anova(model2) #R squared summary(model2) # check heteroskedasticity par(mfrow=c(1,1)) plot(model2) plot(model2$residuals) ###Breusch-pagen test bptest(model2) ###Durbin-Watson test serial correlation dwtest(model2) #predict test data model2_test_result <- predict(model2, newdata=test_data) TSS2 <- sum((test_data$sales-mean(test_data$sales))^2) RSS2 <- sum((test_data$sales-model2_test_result)^2) test_R_squared2 = 1 - (RSS2/TSS2) ########################### #train R squared:0.3237 #test R squared:0.3121 ########################### ##Multiple linear regression with three factors model3<-lm(sales~price+store+billboard,data=train_data) ##Anova table,F-test anova(model3) #R squared summary(model3) # check heteroskedasticity par(mfrow=c(1,1)) plot(model3) plot(model3$residuals) ###Breusch-pagen test bptest(model3) ###Durbin-Watson test serial correlation dwtest(model3) #predict test data model3_test_result <- predict(model3, newdata=test_data) TSS3 <- sum((test_data$sales-mean(test_data$sales))^2) RSS3 <- sum((test_data$sales-model3_test_result)^2) test_R_squared3 = 1 - (RSS3/TSS3) ########################### #train R squared:0.84 #test R squared:0.854 ########################### ##Multiple linear regression with four factors model4<-lm(sales~price+store+billboard+printout,data=train_data) ##Anova table,F-test anova(model4) #R squared summary(model4) # check heteroskedasticity par(mfrow=c(1,1)) plot(model4) ###Breusch-pagen test bptest(model4) ###Durbin-Watson test serial correlation plot(model4$residuals) dwtest(model4) #predict test data model4_test_result <- predict(model4, newdata=test_data) TSS4 <- sum((test_data$sales-mean(test_data$sales))^2) RSS4 <- sum((test_data$sales-model4_test_result)^2) test_R_squared4 = 1 - (RSS4/TSS4) ########################### #train R squared:0.84 #test R squared:0.854 ########################### ##Multiple linear regression with five factors model5<-lm(sales~price+store+billboard+printout+sat,data=train_data) ##Anova table,F-test anova(model5) #R squared summary(model5) # check heteroskedasticity par(mfrow=c(1,1)) plot(model5) plot(model5$residuals) ###Breusch-pagen test heteroskedasticity bptest(model5) ###Durbin-Watson test serial correlation dwtest(model5) #predict test data model5_test_result <- predict(model5, newdata=test_data) TSS5 <- sum((test_data$sales-mean(test_data$sales))^2) RSS5 <- sum((test_data$sales-model5_test_result)^2) test_R_squared5 = 1 - (RSS5/TSS5) ########################### #train R squared:0.9135 #test R squared:0.9158 ########################### ##Multiple linear regression with six factors model6<-lm(sales~price+store+billboard+printout+sat+comp,data=train_data) ##Anova table,F-test anova(model6) #R squared summary(model6) # check heteroskedasticity par(mfrow=c(1,1)) plot(model6) plot(model6$residuals) ###Breusch-pagen test heteroskedasticity bptest(model6) ###Durbin-Watson test serial correlation dwtest(model6) #predict test data model6_test_result <- predict(model6, newdata=test_data) TSS6 <- sum((test_data$sales-mean(test_data$sales))^2) RSS6 <- sum((test_data$sales-model6_test_result)^2) test_R_squared6 = 1 - (RSS6/TSS6) ########################### #train R squared:0.9201 #test R squared:0.92018 ########################### ##Multiple linear regression with five factors model5_1<-lm(sales~price+store+billboard+sat+comp,data=train_data) ##Anova table,F-test anova(model5_1) #R squared summary(model5_1) # check heteroskedasticity par(mfrow=c(1,1)) plot(model5_1) plot(model5_1$residuals) ###Breusch-pagen test heteroskedasticity bptest(model5_1) ###Durbin-Watson test serial correlation dwtest(model5_1) #predict test data model5_1_test_result <- predict(model5_1, newdata=test_data) TSS5_1 <- sum((test_data$sales-mean(test_data$sales))^2) RSS5_1 <- sum((test_data$sales-model5_1_test_result)^2) test_R_squared5_1 = 1 - (RSS5_1/TSS5_1) ########################### #train R squared:0.9201 #test R squared:0.92026 ########################### ##Multiple linear regression with nine factors model9<-lm(sales~price+store+billboard+printout+sat+comp +store:billboard+store:printout+billboard:printout,data=train_data) ##Anova table,F-test anova(model9) #R squared summary(model9) # check heteroskedasticity par(mfrow=c(1,1)) plot(model9) plot(model9$residuals) ###Breusch-pagen test heteroskedasticity bptest(model9) ###Durbin-Watson test serial correlation dwtest(model9) #predict test data model9_test_result <- predict(model9, newdata=test_data) TSS9 <- sum((test_data$sales-mean(test_data$sales))^2) RSS9 <- sum((test_data$sales-model9_test_result)^2) test_R_squared9 = 1 - (RSS9/TSS9) ########################### #train R squared:0.9259 #test R squared:0.92839 ########################### ##Multiple linear regression with six factors model6_1<-lm(sales~price+store+billboard+sat+comp +store:billboard,data=train_data) ##Anova table,F-test anova(model6_1) #R squared summary(model6_1) # check heteroskedasticity par(mfrow=c(1,1)) plot(model6_1) plot(model6_1$residuals) ###Breusch-pagen test heteroskedasticity bptest(model6_1) ###Durbin-Watson test serial correlation dwtest(model6_1) #predict test data model6_1_test_result <- predict(model6_1, newdata=test_data) TSS6_1 <- sum((test_data$sales-mean(test_data$sales))^2) RSS6_1 <- sum((test_data$sales-model6_1_test_result)^2) test_R_squared6_1 = 1 - (RSS6_1/TSS6_1) ########################### #train R squared:0.9257 #test R squared:0.9284961 ###########################

d46c29cffbe95ce0e246b6ea90c5335aaf82632f

9a3b4965c85af90f870baba83c23a0103a986353

/assignments/a6/plot_null_alt.R

03fe14d274f729ca01258dc8a3f54e41b84ff271

[ "CC0-1.0", "GPL-1.0-or-later", "GPL-2.0-only", "GPL-3.0-only", "CC-BY-4.0" ]

permissive

ly129/EPIB607

2ed99374d99c4e8f003191782af7a6fd97707756

ac2f917bc064f8028a875766af847114cd306396

refs/heads/master

2020-07-05T19:28:44.636344

2018-12-11T13:13:52

202,746,856

0

1

CC0-1.0

2019-08-16T14:56:33

null

UTF-8

R

false

10,765

r

plot_null_alt.R

#' Plot null and alternative distributions #' @param n sample size #' @param s population standard deviation (or estimated standard deviation) #' @param mu0 mean under the null hypothesis #' @param mha mean under the alternative hypothesis #' @param alternative is alternative hypothesis greater than or less than or #' equal to mu0. Defaults to 'less'. If alternative='equal' then two cutoff #' points must be specified #' @param cutoff critical value(s). if alternative='equal', then you must #' provide two values, e.g., for alpha level 0.05, cutoffs = qnorm(c(0.025, #' 0.975), mu0, s/sqrt(n)) #' @param legend show legend? Defaults to TRUE #' @param ... other arguments passed to graphics::title #' @details requires the latex2exp package to be installed power_plot <- function(n, s, mu0, mha, cutoff, alternative = c("less", "greater", "equal"), legend = TRUE, ...) { if (!requireNamespace("latex2exp")) stop("you need to install the 'latex2exp' package for this function to work") cbPalette <- c("#999999", "#E69F00", "#56B4E9", "#009E73", "#F0E442", "#0072B2", "#D55E00", "#CC79A7") alternative <- match.arg(alternative) SEM <- s / sqrt(n) if (alternative == "greater") { if (mha < mu0) stop("mean under Ha is less than the null. select alternative='less'") x <- seq(mu0 - 4.25*SEM, mha + 3*SEM, length = 1000) dh0 <- dnorm(x, mu0, SEM) dh1 <- dnorm(x, mha, SEM) ht <- 1.1 * dnorm(mu0, mu0, SEM) plot.new() plot.window(xlim = range(x), ylim = c(.01*ht, 3.2*ht)) axis(1) title(...) # null green <- seq(mu0 - 3 * SEM, cutoff, length.out = 1000) d <- c(dnorm(green,mu0,SEM),0) I <- 1 polygon(c(green,cutoff),d + I * ht, col = cbPalette[4], border = NA) red <- seq(cutoff, mu0 + 3 * SEM, length.out = 1000) d <- c(dnorm(red, mu0, SEM), 0) polygon(c(red,cutoff), d + I * ht, col = "red", border = NA) points(mu0, I * ht, cex = 0.7, pch = 19) text(labels = latex2exp::TeX(sprintf("$\\mu_{H_0} = %#.4f$", mu0)), x = mu0, y = I*ht*.90) # alternative green <- seq(mha - 3 * SEM, cutoff, length.out = 1000) d <- c(dnorm(green, mha, SEM), 0) polygon(c(green,cutoff),d + (I + 1) * ht, col = cbPalette[6], border = NA) points(mha, (I + 1) * ht, cex = 0.7, pch = 19) text(labels = latex2exp::TeX(sprintf("$\\mu_{H_A} = %#.4f$", mha)), x = mha, y = (I + 1) * ht * .95) red <- seq(cutoff, mha + 3 * SEM, length.out = 1000) d <- c(dnorm(red, mha, SEM), 0) polygon(c(red,cutoff), d + (I + 1) * ht, col = cbPalette[2], border = NA) alpha <- pnorm(cutoff, mu0, SEM, lower.tail = FALSE) beta <- pnorm(cutoff, mha, SEM) labs.h0 <- latex2exp::TeX(sprintf("$\\alpha$ = %#.3f", alpha)) labs.h1a <- latex2exp::TeX(sprintf("$\\beta$ = %#.3f", beta)) labs.h1b <- latex2exp::TeX(sprintf("$1 - \\beta$ = %#.3f",1 - beta)) if (legend) legend("topleft", legend = c(labs.h0, labs.h1a, labs.h1b), pch = 15,cex = 1.2, col = c("red", cbPalette[c(6,2)])) segments(cutoff,(I)*ht*0.2, cutoff,(I+1)*ht,lwd=0.5,col="red") text(labels = latex2exp::TeX(sprintf("cutoff = %#.4f$", cutoff)), x = cutoff, y = (I)*ht*0.15) arrows(mu0,(I-0.25)*ht, cutoff,(I-0.25)*ht,length=0.05, code=3,angle=20,col=cbPalette[6],lwd=1.5) arrows(cutoff,(I+1-0.25)*ht, mha,(I+1-0.25)*ht,length=0.05, code=3,angle=15,col="red",lwd=1.5) segments(mha, I * ht, mha, (I + 2) * ht * 1.2,lwd=0.5,col="grey60") } if (alternative == "less") { if (mha > mu0) stop("mean under Ha is greater than the null. select alternative='greater'") # browser() x <- seq(mha - 4.25*SEM, mu0 + 3*SEM, length = 1000) dh0 <- dnorm(x, mu0, SEM) dh1 <- dnorm(x, mha, SEM) ht <- 1.1 * dnorm(mu0, mu0, SEM) plot.new() plot.window(xlim = range(x), ylim = c(.01*ht, 3.2*ht)) axis(1) title(...) # null green <- seq(mu0 + 3 * SEM, cutoff, length.out = 1000) d <- c(dnorm(green,mu0,SEM),0) I <- 1 polygon(c(green,cutoff),d + I * ht, col = cbPalette[4], border = NA) red <- seq(mu0 - 3 * SEM, cutoff, length.out = 1000) d <- c(dnorm(red, mu0, SEM), 0) polygon(c(red,cutoff), d + I * ht, col = "red", border = NA) points(mu0, I * ht, cex = 0.7, pch = 19) text(labels = latex2exp::TeX(sprintf("$\\mu_{H_0} = %#.4f$", mu0)), x = mu0, y = I*ht*.90) # alternative green <- seq(mha + 3 * SEM, cutoff, length.out = 1000) d <- c(dnorm(green, mha, SEM), 0) polygon(c(green,cutoff),d + (I + 1) * ht, col = cbPalette[6], border = NA) points(mha, (I + 1) * ht, cex = 0.7, pch = 19) text(labels = latex2exp::TeX(sprintf("$\\mu_{H_A} = %#.4f$", mha)), x = mha, y = (I + 1) * ht * .95) red <- seq(mha - 3 * SEM, cutoff, length.out = 1000) d <- c(dnorm(red, mha, SEM), 0) polygon(c(red,cutoff), d + (I + 1) * ht, col = cbPalette[2], border = NA) alpha <- pnorm(cutoff, mu0, SEM, lower.tail = TRUE) beta <- pnorm(cutoff, mha, SEM, lower.tail = FALSE) labs.h0 <- latex2exp::TeX(sprintf("$\\alpha$ = %#.3f", alpha)) labs.h1a <- latex2exp::TeX(sprintf("$\\beta$ = %#.3f", beta)) labs.h1b <- latex2exp::TeX(sprintf("$1 - \\beta$ = %#.3f",1 - beta)) if (legend) legend("topleft", legend = c(labs.h0, labs.h1a, labs.h1b), pch = 15,cex = 1.2, col = c("red", cbPalette[c(6,2)])) segments(cutoff,(I)*ht*0.2, cutoff,(I+1)*ht,lwd=0.5,col="red") text(labels = latex2exp::TeX(sprintf("cutoff = %#.4f$", cutoff)), x = cutoff, y = (I)*ht*0.15) arrows(mu0,(I-0.25)*ht, cutoff,(I-0.25)*ht,length=0.05, code=3,angle=20,col=cbPalette[6],lwd=1.5) arrows(cutoff,(I+1-0.25)*ht, mha,(I+1-0.25)*ht,length=0.05, code=3,angle=15,col="red",lwd=1.5) segments(mha, I * ht, mha, (I + 2) * ht * 1.2,lwd=0.5,col="grey60") } if (alternative == "equal") { # if (mha > mu0) stop("mean under Ha is greater than the null. select alternative='greater'") if (length(cutoff) != 2) stop("cutoff should be a vector of length 2 when alternative='equal'") # browser() x <- seq(min(mha - 4.25*SEM, mu0 - 4.25*SEM), max(mha + 4.25*SEM, mu0 + 4.25*SEM), length = 1000) dh0 <- dnorm(x, mu0, SEM) dh1 <- dnorm(x, mha, SEM) ht <- 1.1 * dnorm(mu0, mu0, SEM) plot.new() plot.window(xlim = range(x), ylim = c(.01*ht, 3.2*ht)) axis(1) # axis(2) title(...) # null ---- green <- seq(min(cutoff), max(cutoff), length.out = 1000) d <- c(0,dnorm(green,mu0,SEM),0) I <- 1 polygon(c(min(cutoff),green,max(cutoff)), d + I * ht, col = cbPalette[4], border = NA) # lower tail for null red <- seq(mu0 - 3 * SEM, min(cutoff), length.out = 1000) d <- c(dnorm(red, mu0, SEM), 0) polygon(c(red,min(cutoff)), d + I * ht, col = "red", border = NA) # upper tail for null red <- seq(max(cutoff), mu0 + 3 * SEM, length.out = 1000) d <- c(0, dnorm(red, mu0, SEM)) polygon(c(max(cutoff), red), d + I * ht, col = "red", border = NA) points(mu0, I * ht, cex = 0.7, pch = 19) text(labels = latex2exp::TeX(sprintf("$\\mu_{H_0} = %#.4f$", mu0)), x = mu0, y = I*ht*.90) # alternative - upper ---- green <- seq(min(cutoff), max(cutoff), length.out = 1000) d <- c(0,dnorm(green,mha,SEM),0) polygon(c(min(cutoff),green,max(cutoff)), d + (I+1) * ht, col = cbPalette[6], border = NA) points(mha, (I + 1) * ht, cex = 0.7, pch = 19) text(labels = latex2exp::TeX(sprintf("$\\mu_{H_A} = %#.4f$", mha)), x = mha, y = (I + 1) * ht * .95) # lower tail for alternative red <- seq(mha - 4.25 * SEM, min(cutoff), length.out = 1000) d <- c(dnorm(red, mha, SEM), 0) polygon(c(red,min(cutoff)), d + (I + 1) * ht, col = cbPalette[2], border = NA) # upper tail for alternative red <- seq(max(cutoff), mha + 4.25 * SEM, length.out = 1000) d <- c(0, dnorm(red, mha, SEM)) polygon(c(max(cutoff), red), d + (I + 1) * ht, col = cbPalette[2], border = NA) alpha <- pnorm(min(cutoff), mu0, SEM, lower.tail = TRUE) + pnorm(max(cutoff), mu0, SEM, lower.tail = FALSE) power <- pnorm(min(cutoff), mha, SEM, lower.tail = TRUE) + pnorm(max(cutoff), mha, SEM, lower.tail = FALSE) beta <- 1 - power labs.h0 <- latex2exp::TeX(sprintf("$\\alpha$ = %#.3f", alpha)) labs.h1a <- latex2exp::TeX(sprintf("$\\beta$ = %#.3f", beta)) labs.h1b <- latex2exp::TeX(sprintf("$1 - \\beta$ = %#.3f",1 - beta)) if (legend) legend("topleft", legend = c(labs.h0, labs.h1a, labs.h1b), pch = 15,cex = 1.2, col = c("red", cbPalette[c(6,2)])) segments(cutoff,(I)*ht*0.2, cutoff,(I+1)*ht,lwd=0.5,col="red") text(labels = latex2exp::TeX(sprintf("cutoff = %#.4f$", cutoff)), x = cutoff, y = (I)*ht*0.15) arrows(mu0,(I-0.25)*ht, cutoff,(I-0.25)*ht,length=0.05, code=3,angle=20,col=cbPalette[6],lwd=1.5) # arrows(cutoff,(I+1-0.25)*ht, # mha,(I+1-0.25)*ht,length=0.05, # code=3,angle=15,col="red",lwd=1.5) segments(mha, I * ht, mha, (I + 2) * ht * 1.2,lwd=0.5,col="grey60") } } # examples # # less than alternative ---- # n <- 5 # sample size # s <- 0.0080 # standard deviation # mu0 <- -0.540 # mean undder the null # mha <- 1.01 * mu0 # mean under the alternative # cutoff <- mu0 + qnorm(0.05) * s / sqrt(n) # power_plot(n = n, # s = s, # mu0 = mu0, # mha = mha, # cutoff = cutoff, # alternative = "less", # xlab = "") # # # # greater than alternative ---- # n <- 5 # sample size # s <- 0.0080 # standard deviation # mu0 <- -0.540 # mean undder the null # mha <- 0.99 * mu0 # mean under the alternative # cutoff <- mu0 + qnorm(0.95) * s / sqrt(n) # power_plot(n = n, # s = s, # mu0 = mu0, # mha = mha, # cutoff = cutoff, # alternative = "greater", # xlab = "") # # # # two-sided alternative ---- # n <- 3 # sample size # s <- 0.088 # standard deviation # mu0 <- .86 # mean undder the null # mha <- .88 # mean under the alternative # cutoff <- mu0 + qnorm(c(0.025, 0.975)) * s / sqrt(n) # power_plot(n = n, # s = s, # mu0 = mu0, # mha = mha, # cutoff = cutoff, # alternative = "equal", # xlab = "")

8248479993e78b05536f9472d0199a288d5e1c0b

ffaf081897ef7781ec44634b757eae461e848ccb

/day03/day03.R

90501debf75661a0c8a92f9f58f19d048ad99931

[]

no_license

marcmace/AdventofCode2020

f8e80842f16fd1c9f7fd181bbcc0054f1b7031c0

642ce0922052bf7853c1dccf4710b7682345f085

refs/heads/main

2023-01-29T20:53:42.362419

2020-12-09T13:49:27

317,989,115

0

null

UTF-8

R

false

508

r

day03.R

library(dplyr) input <- readLines(paste(getwd(),"/day03/day03_input.txt",sep="")) geo <- transpose(as.data.frame(strsplit(x = input, split = ""))) day3 <- function(colmove, rowmove) { trees <- 0 i <- 1 j <- 1 while (i <= nrow(geo)) { if (geo[i,j]== "#") { trees <- trees + 1 # geo[i,j] <- "X" } # else geo[i,j] <- "O" i <- i + rowmove j <- (j + colmove - 1) %% ncol(geo) + 1 } return (trees) }

207ef8d944b16119cd17e00d52cee2400274ad9f

9425481e2f3e6218b31870f121445400e2c47530

/Binomial/man/bin_distribution.Rd

12bf52f8c4a479c0669b690040df0003a625a981

[]

no_license

stat133-sp19/hw-stat133-JingtongZhao

909c7e2cefeae1dd01e66842af132452b8170285

b60cab3396f9871239ff687707c36686d29fd7bb

refs/heads/master

2020-04-28T07:25:17.528497

2019-05-04T00:39:47

175,091,840

0

null

UTF-8

R

false

true

598

rd

bin_distribution.Rd

% Generated by roxygen2: do not edit by hand % Please edit documentation in R/Binomial.R \name{bin_distribution} \alias{bin_distribution} \title{Binomial Distribution} \usage{ bin_distribution(trials, prob) } \arguments{ \item{trials}{number of fixed trials} \item{prob}{probability of success on each trial} } \value{ an object of class \code{"bindis"} } \description{ calculates probabilities based on different number of successes in a fixed number of random trials performed under identical conditions } \examples{ #binomial probability distribution bin_distribution(trials = 5, prob = 0.5) }

59faa190a5796e6a10d0a2afc5286f30c36b802e

a47e15d8a4b9bd62db8a531dacbdaa7d8a797a3d

/man/layer.Rd

178a692cc1d99ca75b445ba0a6d3c7b76dc5f1ff

[]

no_license

johannes-titz/leabRa

f5d4189793db830e5f5aed7af1df0305b3ad58f3

237bc5c67c81fcd24e7dac895af3106bbad9974a

refs/heads/master

2021-05-15T00:03:52.568144

2017-09-25T08:53:48

103,940,125

1

0

null

UTF-8

R

false

true

7,197

rd

layer.Rd

% Generated by roxygen2: do not edit by hand % Please edit documentation in R/layer.R \docType{class} \name{layer} \alias{layer} \title{Leabra layer class} \format{\code{\link{R6Class}} object} \usage{ layer } \value{ Object of \code{\link{R6Class}} with methods for calculating changes of activation in a layer of neurons. } \description{ This class simulates a biologically realistic layer of neurons in the Leabra framework. It consists of several \code{\link{unit}} objects in the variable (field) \code{units} and some layer-specific variables. } \section{Fields}{ \describe{ \item{\code{units}}{A list with all \code{\link{unit}} objects of the layer.} \item{\code{avg_act}}{The average activation of all units in the layer (this is an active binding).} \item{\code{n}}{Number of units in layer.} \item{\code{weights}}{A receiving x sending weight matrix, where the receiving units (rows) has the current weight values for the sending units (columns). The weights will be set by the \code{\link{network}} object, because they depend on the connection to other layers.} \item{\code{ce_weights}}{Sigmoidal contrast-enhanced version of the weight matrix \code{weights}. These weights will be set by the \code{\link{network}} object.} \item{\code{layer_number}}{Layer number in network (this is 1 if you create a layer on your own, without the network class).} }} \section{Methods}{ \describe{ \item{\code{new(dim, g_i_gain = 2)}}{Creates an object of this class with default parameters. \describe{ \item{\code{dim}}{A pair of numbers giving the dimensions (rows and columns) of the layer.} \item{\code{g_i_gain}}{Gain factor for inhibitory conductance, if you want less activation in a layer, set this higher.} } } \item{\code{get_unit_acts()}}{Returns a vector with the activations of all units of a layer. } \item{\code{get_unit_scaled_acts()}}{Returns a vector with the scaled activations of all units of a layer. Scaling is done with \code{recip_avg_act_n}, a reciprocal function of the number of active units. } \item{\code{cycle(intern_input, ext_input)}}{Iterates one time step with layer object. \describe{ \item{\code{intern_input}}{Vector with inputs from all other layers. Each input has already been scaled by a reciprocal function of the number of active units (\code{recip_avg_act_n}) of the sending layer and by the connection strength between the receiving and sending layer. The weight matrix \code{ce_weights} is multiplied with this input vector to get the excitatory conductance for each unit in the layer. } \item{\code{ext_input}}{Vector with inputs not coming from another layer, with length equal to the number of units in this layer. If empty (\code{NULL}), no external inputs are processed. If the external inputs are not clamped, this is actually an excitatory conductance value, which is added to the conductance produced by the internal input and weight matrix. } } } \item{\code{clamp_cycle(activations)}}{Iterates one time step with layer object with clamped activations, meaning that activations are instantaneously set without time integration. \describe{ \item{\code{activations}}{Activations you want to clamp to the units in the layer. } } } \item{\code{get_unit_act_avgs()}}{Returns a list with the short, medium and long term activation averages of all units in the layer as vectors. The super short term average is not returned, and the long term average is not updated before being returned (this is done in the function \code{chg_wt()} with the method\code{updt_unit_avg_l}). These averages are used by the network class to calculate weight changes. } \item{\code{updt_unit_avg_l()}}{Updates the long-term average (\code{avg_l}) of all units in the layer, usually done after a plus phase. } \item{\code{updt_recip_avg_act_n()}}{Updates the \code{avg_act_inert} and \code{recip_avg_act_n} variables, these variables update before the weights are changed instead of cycle by cycle. This version of the function assumes full connectivity between layers. } \item{\code{reset(random = FALSE)}}{Sets the activation and activation averages of all units to 0. Used to begin trials from a stationary point. \describe{ \item{\code{random}}{Logical variable, if TRUE the activations are set randomly between .05 and .95 for every unit instead of 0. } } } \item{\code{set_ce_weights()}}{Sets contrast enhanced weight values. } \item{\code{get_unit_vars(show_dynamics = TRUE, show_constants = FALSE)}}{Returns a data frame with the current state of all unit variables in the layer. Every row is a unit. You can choose whether you want dynamic values and / or constant values. This might be useful if you want to analyze what happens in units of a layer, which would otherwise not be possible, because most of the variables (fields) are private in the unit class. \describe{ \item{\code{show_dynamics}}{Should dynamic values be shown? Default is TRUE. } \item{\code{show_constants}}{Should constant values be shown? Default is FALSE. } } } \item{\code{get_layer_vars(show_dynamics = TRUE, show_constants = FALSE)}}{Returns a data frame with 1 row with the current state of the variables in the layer. You can choose whether you want dynamic values and / or constant values. This might be useful if you want to analyze what happens in a layer, which would otherwise not be possible, because some of the variables (fields) are private in the layer class. \describe{ \item{\code{show_dynamics}}{Should dynamic values be shown? Default is TRUE. } \item{\code{show_constants}}{Should constant values be shown? Default is FALSE. } } } } } \examples{ l <- layer$new(c(5, 5)) # create a 5 x 5 layer with default leabra values l$g_e_avg # private values cannot be accessed # if you want to see alle variables, you need to use the function l$get_layer_vars(show_dynamics = TRUE, show_constants = TRUE) # if you want to see a summary of all units without constant values l$get_unit_vars(show_dynamics = TRUE, show_constants = FALSE) # let us clamp the activation of the 25 units to some random values between # 0.05 and 0.95 l <- layer$new(c(5, 5)) activations <- runif(25, 0.05, .95) l$avg_act l$clamp_cycle(activations) l$avg_act # what happened to the unit activations? l$get_unit_acts() # compare with activations activations # scaled activations are scaled by the average activation of the layer and # should be smaller l$get_unit_scaled_acts() } \references{ O'Reilly, R. C., Munakata, Y., Frank, M. J., Hazy, T. E., and Contributors (2016). Computational Cognitive Neuroscience. Wiki Book, 3rd (partial) Edition. URL: \url{http://ccnbook.colorado.edu} Have also a look at \url{https://grey.colorado.edu/emergent/index.php/Leabra} (especially the link to the 'MATLAB' code) and \url{https://en.wikipedia.org/wiki/Leabra} } \keyword{data}

71a2e3db2ff8bece607d7bb02a72bb08e98a6a4a

ffdea92d4315e4363dd4ae673a1a6adf82a761b5

/data/genthat_extracted_code/biolink/examples/urls.Rd.R

f9e2ad21c8b6b383ef381db07378c91fbf0bb51e

[]

no_license

surayaaramli/typeRrh

d257ac8905c49123f4ccd4e377ee3dfc84d1636c

66e6996f31961bc8b9aafe1a6a6098327b66bf71

refs/heads/master

2023-05-05T04:05:31.617869

2019-04-25T22:10:06

null

0

null

UTF-8

R

false

302

r

urls.Rd.R

library(biolink) ### Name: urls ### Title: Construct urls to online resources ### Aliases: urls url_go url_kegg url_pubmed url_entrez url_cran url_bioc ### ** Examples # gene ontology url url_go("GO:0005539") # KEGG pathway url url_kegg("hsa04915") # PubMed article url url_pubmed("23193287")

4ca68847ee5bb01f22ee69c3ca3817dbb4270179

3b0be5721a5478b1bac4e6b08cdcd1b88e3a4046

/inst/snippets/Example9.18d.R

305ac60005a35771249b0972cb4a7f558fa74046

[]

no_license

stacyderuiter/Lock5withR

b7d227e5687bc59164b9e14de1c8461cb7861b14

417db714078dc8eaf91c3c74001b88f56f09b562

refs/heads/master

2020-04-06T06:33:39.228231

2015-05-27T11:41:42

null

0

null

UTF-8

R

false

43

r

Example9.18d.R

Ink.Price(PPM=3.0, interval='confidence')

c64726ce65cc8ad5665eca698f05e658dcea35f5

e3847b953b7bb3e6464a7686801fc607fde3a0a1

/Plot2.R

711c11b6eee52fb193cf2ca603d04b7520ac8b2f

[]

no_license

asrulnb/ExData_Plotting1

d2b00aaead4434d9ef656ba60a68a04776671c73

a77d628debf1305cbad893b0f20a7f7dda04e033

refs/heads/master

2021-01-16T21:08:59.127160

2015-08-09T15:53:03

40,299,974

0

null

2015-08-06T10:52:21

null

UTF-8

R

false

910

r

Plot2.R

###[ Initializing Library ] rm(list = ls()) library(dplyr) library(data.table) library(lubridate) library(datasets) library(graphics) ###[ Set working Directory to where the R source file is ] this.dir <- dirname(parent.frame(2)$ofile) setwd(this.dir) ###[ Part 1 : Read data from file ] mainDT <- read.table("household_power_consumption.txt", header = TRUE, sep = ";", na.strings="?") ###[ Part 2 : Filter Data ] mainDT <- filter(mainDT,(Date == "1/2/2007"|Date == "2/2/2007")) ###[ Convert Data ] datetime <- paste(dmy(mainDT$Date), " ", mainDT$Time) ## put together with proper formating mainDT <- mutate(mainDT, DateAndTime = ymd_hms(datetime)) ## to ensure the final Date and Time format ###[ Output to PNG File ] plot(mainDT$Global_active_power~mainDT$DateAndTime, type="l", ylab="Global Active Power (kilowatts)", xlab="") dev.copy(png, file="plot2.png", height=480, width=480) dev.off()

7a917e2807b4c5bc1d669837d05bb7f75761f34d

71808098f4c7eb7ed63ea4806509feaf84fac756

/Regression/Guns_Regression.r

a59b799d7bd1a155b3b408cabc56eaf1d3baf7e5

[]

no_license

MarcoBuratti/BusinessAnalytics-HACKATON

b1f691da9477d3b87cfb09099ee40e5d15567291

efb2e3ecbc0a0967ecd49dda494d6fe35183c52c

refs/heads/master

2023-01-23T22:07:30.483411

2020-11-12T15:57:23

307,710,379

1

0

null

UTF-8

R

false

3,654

r

Guns_Regression.r

library("readxl") library(rgl) library(kknn) # data <- read_excel('Guns.xls') data <- Guns names(data) sapply(data, class) #response violent <- log(data$violent) #sapply(violent, class) #regressor afam <- data$afam # Plot data with the log value on y axis plot(afam, violent, xlim=c(0,15), ylim=c(3,8)) grid() abline(0,1, lty=3) # Fit a linear model fm <- lm(violent ~ afam) summary(fm) # plot the regression line and the fitted values abline(coefficients(fm), col='red') points(afam, fitted(fm), col='red', pch=16) # Plot the avg value of afam on x-axis and violent on y-axis abline(h=mean(violent)) abline(v=mean(afam)) # Create a qq-plot of the residuals qqnorm(residuals(fm)) # Perform a Normality test of the residuals shapiro.test(residuals(fm)) # Print the estimates of the regression coefficients coefficients(fm) # Compute CI of the regression coefficients confint(fm, level= 0.95) # Print the estimate of the error variance s2 <- sum(residuals(fm)^2)/fm$df sqrt(s2) # Compute a set of CIs # Build a new data base Z0 <- data.frame(cbind(afam=seq(0, 35, by=0.1))) # Compute the CIs CI <- predict(fm, Z0, interval='confidence') # Plot the CIs plot(afam, violent, xlim=c(0,35), ylim=c(4,8)) lines(Z0[,1], CI[,'fit']) lines(Z0[,1], CI[,'lwr'], lty=4) lines(Z0[,1], CI[,'upr'], lty=4) # Compute a set of PIs PI <- predict(fm, Z0, interval='prediction', level=0.95) # Plot the PIs lines(Z0[,1], PI[,'fit']) lines(Z0[,1], PI[,'lwr'], lty=2) lines(Z0[,1], PI[,'upr'], lty=2) # Import the second regressor population <- data$population # Plot data with the log value on y axis plot(population, violent, xlim=c(0,15), ylim=c(3,8)) grid() abline(0,1, lty=3) # Fit a linear model fm2 <- lm(violent ~ population) summary(fm2) # plot the regression line and the fitted values abline(coefficients(fm2), col='red') points(population, fitted(fm2), col='red', pch=16) # Add a third regressor income <- data$income # Plot data with the log value on y axis plot(income, violent, xlim=c(9000,20000), ylim=c(3,8)) grid() abline(0,1, lty=3) # Fit a linear model fm3 <- lm(violent ~ income) summary(fm3) # plot the regression line and the fitted values abline(coefficients(fm3), col='red') points(income, fitted(fm3), col='red', pch=16) fm3 <- lm(violent ~ afam + population + income) summary(fm3) pairs(cbind(violent, afam, population, income)) #Plot in 3D pop, afam and violent open3d() plot3d(x=afam, y=population, z=violent, size=10, col='black') # Fit the linear model fm2 <- lm(violent ~ afam + population) summary(fm2) # plot the regression surface and the fitted values points3d(x=afam, y=population, z=fitted(fm2), size=10, col='red') planes3d(coefficients(fm2)[2],coefficients(fm2)[3],-1,coefficients(fm2)[1], alpha=0.5, color='red') # Plot data with the log value on y axis plot(afam, violent, xlim=c(0,15), ylim=c(3,8)) grid() abline(0,1, lty=3) # Fit a linear model fm <- lm(violent ~ afam) summary(fm) # plot the regression line and the fitted values abline(coefficients(fm), col='red') # Compute the fitted values using k=8 and the rectangular kernel Z0 <- data.frame(cbind(afam=seq(0, 35, by=0.1))) predicted.response <- kknn(violent ~ afam, train = data.frame(afam, violent), test = Z0, k = 18, kernel='rectangular') # Plot the knn prediction line lines(Z0[,'afam'], predicted.response$fitted.values, col='blue') # Select the value of k (leave-one-out crossvalidation method) train.cv <- train.kknn(violent ~ afam, data = data.frame(afam, violent), kmax = 40, scale = F, kernel = 'rectangular') plot(train.cv) grid()

3f3466431535a58007eb4f3e0058a0a972d1a2aa

fadd039259d32ffee1b8387659295cd3c94067b2

/02 R Programming/Week 3/Programming Assignment/ProgrammingAssignment3.R

14f8bbe06c2079a3c57841aa610a248a96414d3c

[]

no_license

castner-jon/datasciencecoursera

e4198c7668ab98ae746fdc21531fbb0dab2e09b3

2b602b37d278aaf11547c9d8fd162ca587fef0a5

refs/heads/master

2021-09-09T16:29:55.751823

2018-03-18T01:00:53

119,195,856

0

null

UTF-8

R

false

346

r

ProgrammingAssignment3.R

## read in the data outcome <- read.csv("outcome-of-care-measures.csv", colClasses = "character") head(outcome) class(outcome) str(outcome) dim(outcome) colnames(outcome) ## histogram of mortality rates from heart attacks outcome[, 11] <- as.numeric(outcome[, 11]) hist(outcome$Hospital.30.Day.Death..Mortality..Rates.from.Heart.Attack)

80c242129acdba02b573b11b6dc45a819da6c6a7

2bec5a52ce1fb3266e72f8fbeb5226b025584a16

/uwot/tests/testthat/helper_data.R

e49ad62eb3c1bedcd002a288e8679f429596d23b

[]

no_license

akhikolla/InformationHouse

4e45b11df18dee47519e917fcf0a869a77661fce

c0daab1e3f2827fd08aa5c31127fadae3f001948

refs/heads/master

2023-02-12T19:00:20.752555

2020-12-31T20:59:23

325,589,503

9

2

null

UTF-8

R

false

1,665

r

helper_data.R

# Small -ve distances are possible dist2 <- function(X) { D2 <- rowSums(X * X) D2 + sweep(X %*% t(X) * -2, 2, t(D2), `+`) } # Squared Euclidean distances, ensuring no small -ve distances can occur safe_dist2 <- function(X) { D2 <- dist2(X) D2[D2 < 0] <- 0 D2 } # convert dataframe to distance matrix x2d <- function(X) { sqrt(safe_dist2(x2m(X))) } # Covert a vector into a 2D matrix for generating Y output c2y <- function(...) { matrix(unlist(list(...)), ncol = 2) } iris10 <- x2m(iris[1:10, ]) iris10_Y <- pca_scores(iris10, ncol = 2) diris10 <- dist(iris10) # Sparse iris10 dist dmiris10 <- as.matrix(diris10) dmiris10z <- dmiris10 dmiris10z[dmiris10z > 0.71] <- 0 dmiris10z <- Matrix::drop0(dmiris10z) # some Y data ycat <- as.factor(c(levels(iris$Species)[rep(1:3, each = 3)], NA)) ycat2 <- as.factor(c(NA, levels(iris$Species)[rep(1:3, times = 3)])) ynum <- (1:10) / 10 ynum2 <- seq(from = 10, to = -10, length.out = 10) / 100 nn <- find_nn(iris10, k = 4, method = "fnn", metric = "euclidean", n_threads = 0, verbose = FALSE ) # Just test that res is a matrix with valid numbers expect_ok_matrix <- function(res, nr = nrow(iris10), nc = 2) { expect_is(res, "matrix") expect_equal(nrow(res), nr) expect_equal(ncol(res), nc) expect_false(any(is.infinite(res))) } expect_is_nn <- function(res, nr = 10, k = 4) { expect_is(res, "list") expect_is_nn_matrix(res$dist, nr, k) expect_is_nn_matrix(res$idx, nr, k) } expect_is_nn_matrix <- function(res, nr = 10, k = 4) { expect_is(res, "matrix") expect_equal(nrow(res), nr) expect_equal(ncol(res), k) }

5062eb4fa0786f275049d0f8eb4660398441a086

f8f3d53abf579dfbf6d49cfb59295b1c3ddc3fb2

/R/add_flextable.R

fac9a391aa6d2161eeaa3602f0d181e0f74f598f

[]

no_license

cardiomoon/rrtable

9010574549a6fc41015f89638a708c691c7975cf

8346fca2bb0dc86df949fb31738e1af90eeb5a70

refs/heads/master

2023-03-15T20:43:07.685721

2023-03-12T11:36:34

127,721,282

3

2

null

2021-11-17T01:08:31

2018-04-02T07:32:08

R

UTF-8

R

false

1,455

r

add_flextable.R

#' Add a flextable or mytable object into a document object #' @param mydoc A document object #' @param ftable A flextable or mytable object #' @param code R code string #' @param echo whether or not display R code #' @param landscape Logical. Whether or not make a landscape section. #' @importFrom officer add_slide ph_with body_add_par body_end_section_landscape body_end_section_portrait #' @importFrom flextable body_add_flextable #' @return a document object #' @export #' @examples #' \dontrun{ #' require(rrtable) #' require(moonBook) #' require(officer) #' require(magrittr) #' ftable=mytable(Dx~.,data=acs) #' title="mytable Example" #' ft=df2flextable(head(iris)) #' title2="df2flextable Example" #' doc=read_docx() #' doc %>% add_text(title=title) %>% #' add_flextable(ftable) %>% #' add_text(title=title2) %>% #' add_flextable(ft) #'} add_flextable=function(mydoc,ftable,echo=FALSE,code="",landscape=FALSE){ if("mytable" %in% class(ftable)){ ft<-mytable2flextable(ftable) } else { ft<-ftable } pos=1.5 if(echo & (code!="")) pos=2 if(inherits(mydoc,"rpptx")){ mydoc<-mydoc %>% ph_with(value=ft,location = ph_location(left=1,top=pos)) } else { if(landscape) mydoc <- body_end_section_portrait(mydoc) mydoc<-mydoc %>% body_add_flextable(ft) if(landscape) mydoc <- body_end_section_landscape(mydoc) } mydoc }

6cc02fa368753709b26ae50ebc78cfdb4a716be8

79f08f05d41ab55c37bbb5216a827d4f03507a3f

/module3/homework3-Q2/app.R

b6db45c80b05455a6b56884411d308435d91a716

[]

no_license

pmalo46/CUNY_DATA_608

d353bcbbe186675f064d367f520babd12d17a185

5d15862117bdc7e7df926afd29b9ec43523b7b1a

refs/heads/master

2023-02-02T06:23:34.846447

2020-12-14T04:39:33

292,177,538

0

null

2020-09-02T04:22:08

2020-09-02T04:22:07

null

UTF-8

R

false

2,896

r

app.R

# author: Pat Maloney # Data for this project: # https://github.com/charleyferrari/CUNY_DATA608/tree/master/module3/data # Question 2: # Often you are asked whether particular States are improving their mortality # rates (per cause) faster than, or slower than, the national average. Create a # visualization that lets your clients see this for themselves for one cause of # death at the time. Keep in mind that the national average should be weighted by # the national population. library(ggplot2) library(dplyr) library(plotly) library(shiny) library(sqldf) library(rsconnect) df <- read.csv("https://raw.githubusercontent.com/charleyferrari/CUNY_DATA608/master/lecture3/data/cleaned-cdc-mortality-1999-2010-2.csv", header= TRUE) names(df) <- gsub('\\.', '_', names(df)) %>% tolower() national <-sqldf("select ICD_Chapter , Year , round(sum(Deaths)*100000.00 /sum(Population),2) as Crude_Rate , 'National' as State from df Group by ICD_Chapter , Year") names(national) <-tolower(names(national) ) state <- sqldf("select icd_chapter , year , crude_rate , state from df ") national2 <- sqldf("select * from state union all select * from national") ui <- fluidPage( headerPanel('State Mortality Rates Explorer'), sidebarPanel( selectInput('state', 'State', unique(national2$state), selected='NY'), selectInput('icd_chapter', 'Cause of Death', unique(national2$icd_chapter), selected='Certain infectious and parasitic diseases') ), mainPanel( plotlyOutput('plot1'), verbatimTextOutput('stats'), h6("Number of deaths per 100,000 people") ) ) server <- function(input, output, session) { nationalData <- reactive({ national %>% filter(icd_chapter == input$icd_chapter) }) statedata <- reactive({ dfSlice <- national2 %>% filter(state == input$state, icd_chapter == input$icd_chapter) }) combined <- reactive({ merge(x = nationalData(), y = statedata(), all = TRUE) }) output$plot1 <- renderPlotly({ df2 <- national2 %>% filter(state == input$state, icd_chapter == input$icd_chapter) line_colors <- c("red", "blue") plot_ly(combined(), x = ~year, y = ~crude_rate, color = ~state, colors = line_colors, type='scatter', mode = 'lines') }) output$stats <- renderPrint({ df3 <- statedata() %>% filter(state == input$state) summary(df3$crude_rate) }) } shinyApp(ui = ui, server = server)

c842a98a8f3826da7f3466c2d3a5142ec0ee4458

efa4ba01ec27df73d1a40b09ee82a0c7af3cc850

/CODE/13_sleep_site_occupancy_randomizations.R

fe05b52940bfd5c3ce831b7540b1978adc10237b

[]

no_license

CarterLoftus/intergroup_sleep

444e1c1ee4abce299865c216b027b326bb45d7a0

446448b8436f0408f87d13f7ec9359ef1ce860b7

refs/heads/main

2023-04-14T05:20:35.681620

2022-07-27T10:57:51

518,377,903

0

null

UTF-8

R

false

15,034

r

13_sleep_site_occupancy_randomizations.R

#### sleep site randomizations ###### library( infotheo ) library( hms ) library( brms ) library( boot ) library( data.table ) ## function for normalizing a vector normalize_func <- function( x ) return( (x - mean( x, na.rm = T ) )/ sd( x, na.rm = T ) ) which_spec <- 'baboon' if( which_spec == "baboon" ){ spec_df <- read.csv( "DATA/bab_complete.csv" ) }else{ if( which_spec == "vervet" ){ spec_df <- read.csv( "DATA/verv_complete.csv" ) }else{ if( which_spec == "leopard" ){ spec_df <- read.csv( "DATA/leo_complete.csv" ) } } } ## just to confirm that each group has a maximum of one sleep site assigned each night same_sleep_check <- aggregate( spec_df$sleep_clus, by = list( spec_df$group, spec_df$day ), FUN = function( x ) sum( !is.na( unique( x ) ) ) ) names( same_sleep_check ) <- c( 'group', 'day', 'num_sleep_sites') same_sleep_check[ same_sleep_check$num_sleep_sites != 1, ] # make a dataframe with one row per group per night, stating when they left the sleep site that morning and then they arrived at their sleep site in the evening (averaged across the individuals in the group when there is more than one) cosleep_dat <- aggregate( spec_df[ , c( 'arrive_sleep_site', 'leave_sleep_site' ) ], by = list( spec_df$group, spec_df$day, spec_df$sleep_clus ), FUN = function( x ) as_hms( mean( as.numeric( as_hms( as.character( x ) ) ), na.rm = T ) ) ) # rename the columns of the dataframe names( cosleep_dat )[ 1:3 ] <- c( 'group', 'day', 'sleep_clus' ) # reorder the dataframe cosleep_dat <- cosleep_dat[ order( cosleep_dat$day ), ] cosleep_dat <- cosleep_dat[ order( cosleep_dat$group ), ] # make a column that will declare whether the group coslept with another group that night. Instantiate the column with all 0s cosleep_dat$cosleep <- 0 # fill in the cosleep columns with 1's on nights when the group slept at the same site as another group cosleep_dat[ duplicated( cosleep_dat[ , c( 'day', 'sleep_clus' ) ] ) | duplicated( cosleep_dat[ , c( 'day', 'sleep_clus' ) ], fromLast = T ), 'cosleep' ] <- 1 #### how many nights do they cosleep on? sum( duplicated( cosleep_dat[ , c( 'day', 'sleep_clus' ) ] ) ) #### rand parameters #### # make a dataframe with one row per group per night, stating when they left the sleep site that morning and then they arrived at their sleep site in the evening (averaged across the individuals in the group when there is more than one) cosleep_dat <- aggregate( spec_df[ , c( 'arrive_sleep_site', 'leave_sleep_site' ) ], by = list( spec_df$group, spec_df$day, spec_df$sleep_clus ), FUN = function( x ) as_hms( mean( as.numeric( as_hms( as.character( x ) ) ), na.rm = T ) ) ) # rename the columns of the dataframe names( cosleep_dat )[ 1:3 ] <- c( 'group', 'day', 'sleep_clus' ) cosleep_dat$id <- cosleep_dat$group cosleep_dat$func_ind <- cosleep_dat$group tag_names <- as.character( unique( cosleep_dat$id ) ) n <- 1000 rand_day_thresh <- 30 min_to_rand <- 1 # create empty vectors. These vectors will be filled with entries that eventually be put into the final dataframe that declares which functuals need to be compared vec_a <- c() vec_b <- c() for( a in 1:( length( tag_names ) - 1 ) ){ for( b in ( a + 1 ): length( tag_names ) ){ # create vectors that represent the unique combinations that can be made of functuals in this category vec_a <- c( vec_a, tag_names[ a ] ) vec_b <- c( vec_b, tag_names[ b ] ) } } vec_cat_a <- rep( 'baboon', length( vec_a ) ) vec_cat_b <- vec_cat_a # set up the final dataframe that will be filled out to give the empirical coefficient of associations between each dyad of functuals real_final_cosleep <- data.frame( func_a = vec_a, func_b = vec_b, num = rep( NA, times = length( vec_a ) ), denom = rep( NA, times = length( vec_a ) ), MI = NA, cat_a = vec_cat_a, cat_b = vec_cat_b , stringsAsFactors = F) # set up the final dataframe that will be filled out to give the coefficient of associations between each dyad of functuals produced by the randomizations rand_final_cosleep <- data.frame( func_a = rep( vec_a, times = n ), func_b = rep( vec_b, times = n ), rand_n = rep( 1:n , each = length( vec_a ) ), num = rep( NA, times = length( rep( vec_a, times = n ) ) ), denom = rep( NA, times = length( rep( vec_a, times = n ) ) ), MI = NA, cat_a = rep( vec_cat_a, times = n ), cat_b = rep( vec_cat_b, times = n ), stringsAsFactors = F) # create a dataframe that will contain the metadata for the randomizations meta_cosleep_dat <- data.frame( func_a = vec_a, func_b = vec_b, start_rand_at = rep( NA, length( vec_a ) ), end_rand_at = rep( NA, length( vec_a ) ), cat_a = vec_cat_a, cat_b = vec_cat_b, stringsAsFactors = F ) cosleep_dat$day <- as.numeric( cosleep_dat$day ) # saves the first day of the study start_date <- min( cosleep_dat$day ) cosleep_sites_real <- data.frame( func_a = character(), func_b = character(), shared_site = integer() ) cosleep_sites_rand <- data.frame( func_a = character(), func_b = character(), rand_num = integer(), shared_site = integer() ) set.seed( 111 ) for( row in 1:nrow( real_final_cosleep) ){ print( row / nrow(real_final_cosleep) ) # subset the full gps dataframe to just including the functual dyad's data during the correct period. This appropriate combination is determined by the set up of the dataframes above pair_cosleep_dat <- cosleep_dat[ cosleep_dat$id %in% c( real_final_cosleep[ row, c('func_a', 'func_b') ]), ] # just making sure 'func_ind' is a character and not a factor. It messes things up if it is a factor pair_cosleep_dat$func_ind <- as.character( pair_cosleep_dat$func_ind ) # trims the dataframe so it starts on the first day that both members of the dyad have data. We don't want to analyze anything before this pair_cosleep_dat <- pair_cosleep_dat[ pair_cosleep_dat$day >= max( aggregate( pair_cosleep_dat$day , by=list( pair_cosleep_dat$func_ind ), FUN = min )[ , 2 ] ) , ] # makes a column with both of their day columns starting at 1 on the first day when they both have the data. Important for chunking up the data in the next line pair_cosleep_dat$temp_day <- as.numeric( pair_cosleep_dat$day - min( pair_cosleep_dat$day ) + 1, units = 'days' ) # assign each row of data to a subset so that days of the data are only randomized within an range determined by rand_day_thresh pair_cosleep_dat$chunk_num <- ceiling( pair_cosleep_dat$temp_day / rand_day_thresh ) # split up the data into the subsets created in the line above. Now we have a list of dataframes, which each dataframe corresponding to one time chunk within randomizaation is allowable chunked_cosleep_dat <- split( pair_cosleep_dat, f = pair_cosleep_dat$chunk_num ) # create an empty dataframe that will represent a dyadic distance at every simultaneous fix during the current study period total_real <- data.frame( day = integer(), sleep_site_a = integer(), sleep_site_b = integer() ) # perform n permutations of the dyad's dataset for( i in 1:n ){ # create an empty dataframe that will represent a dyadic distance at every derived simultaneous fix of the study period that results after the randomization total_rand <- data.frame( day = integer(), sleep_site_a = integer(), sleep_site_b = integer() ) # loop through each chunk and permute the data within the chunks for( w in 1:length( chunked_cosleep_dat ) ){ # For each chunk of data # save the chunk of data to the dataframe "chunk" chunk <- chunked_cosleep_dat[[w]] # save the number of days of data that individuals a and b have in this chunk num_unique_days_a <- length( unique( chunk[ chunk$func_ind == real_final_cosleep$func_a[ row ], 'day' ] ) ) num_unique_days_b <- length( unique( chunk[ chunk$func_ind == real_final_cosleep$func_b[ row ], 'day' ] ) ) # if one of the functuals has no data for this chunk, skip the rest of the body of the loop and move to the next chunk if( num_unique_days_a == 0 || num_unique_days_b == 0 ){ next } # if either individual a or b don't have the necessary number of days worth of data as determined by the min_to_rand parameter, skip the rest of the body of the loop if( num_unique_days_a < (min_to_rand * rand_day_thresh) || num_unique_days_b < (min_to_rand * rand_day_thresh) ){ next } # The first time through, we will make the empirical dataframe of dyadic distances. We only need to do this once if( i == 1 ){ # use the dyad_dist function to calculate the dyadic distance for this pair of functuals real_sub <- as.data.frame( dcast( as.data.table( chunk ), day ~ id, value.var = 'sleep_clus', drop = F ) ) names( real_sub ) <- c( 'day', 'sleep_site_a', 'sleep_site_b' ) # add this to the running dataframe of dyadic distances over the whole study for this dyad total_real <- rbind( total_real, real_sub ) # save the latest time that will be successfully randomized. For the first run through for each functual dyad, this will get updated every time the conditions above are surpassed such that a chunk of data is successfully randomized. Eventually this will serve to mark the end of successful randomizations end_of_rand <- max(chunked_cosleep_dat[[w]]$day) } # determine how many days to shift ID b's data by for the randomization ID_b_shifts <- sample( 1:( rand_day_thresh - 1 ), 1, replace = TRUE ) # shift ID b's data by the amount determined above new_days_b <- chunk[ chunk$func_ind == real_final_cosleep$func_b[ row ], 'day'] + ID_b_shifts # complicated line of code, but all it does is wrap the end of b's data back around to match the beginning of a's data. So if we shifted b's data by 3, a's data will still be 1, 2, 3, 4, 5, 6, 7; and b's data will be 5, 6, 7, 1, 2, 3, 4. new_days_b[ new_days_b > max( chunk[ chunk$func_ind == real_final_cosleep$func_b[ row ], 'day' ] ) ] <- new_days_b[ new_days_b > max( chunk [ chunk$func_ind == real_final_cosleep$func_b[ row ], 'day' ] ) ] - max( chunk[ chunk$func_ind == real_final_cosleep$func_b[ row ], 'day' ] ) + min( chunk[ chunk$func_ind == real_final_cosleep$func_b[ row ], 'day'] ) - 1 # replace b's day data with these 'fake' shifted days chunk[ chunk$func_ind == real_final_cosleep$func_b[ row ], 'day'] <- new_days_b rand_sub <- as.data.frame( dcast( as.data.table( chunk ), day ~ id, value.var = 'sleep_clus', drop = F ) ) names( rand_sub ) <- c( 'day', 'sleep_site_a', 'sleep_site_b' ) # add this to the running dataframe of derived dyadic distances for the randomized data over the whole study total_rand <- rbind( total_rand, rand_sub ) } # if we are on the first run through the loop, save the empirical CA of the dyad if( i == 1 ){ # adds the empirical coefficient of association for this dyad real_final_cosleep[ row , 'num' ] <- sum( total_real$sleep_site_a == total_real$sleep_site_b, na.rm = T ) real_final_cosleep[ row , 'denom' ] <- sum( !is.na( total_real$sleep_site_a == total_real$sleep_site_b ) ) real_final_cosleep[ row , 'MI' ] <- mutinformation( as.character( total_real$sleep_site_a ) , as.character( total_real$sleep_site_b ) ) meta_cosleep_dat[ row, c( 'start_rand_at', 'end_rand_at' ) ] <- c( min( chunked_cosleep_dat [[ 1 ]]$day ), end_of_rand ) sites_real <- total_real$sleep_site_a[ total_real$sleep_site_a == total_real$sleep_site_b & !is.na( total_real$sleep_site_a == total_real$sleep_site_b ) ] if( length( sites_real ) != 0 ){ sites_real_sub <- data.frame( func_a = real_final_cosleep$func_a[ row ], func_b = real_final_cosleep$func_b[ row ], shared_site = sites_real ) cosleep_sites_real <- rbind( cosleep_sites_real, sites_real_sub ) } } # save the randomized CA of the functual dyad for this randomization rand_final_cosleep[ ( row + nrow( real_final_cosleep ) * ( i - 1 ) ), 'num' ] <- sum( total_rand$sleep_site_a == total_rand$sleep_site_b, na.rm = T ) rand_final_cosleep[ ( row + nrow( real_final_cosleep ) * ( i - 1 ) ), 'denom' ] <- sum( !is.na( total_rand$sleep_site_a == total_rand$sleep_site_b ) ) rand_final_cosleep[ ( row + nrow( real_final_cosleep ) * ( i - 1 ) ) , 'MI' ] <- mutinformation( as.character( total_rand$sleep_site_a ), as.character( total_rand$sleep_site_b ) ) sites_rand <- total_rand$sleep_site_a[ total_rand$sleep_site_a == total_rand$sleep_site_b & !is.na( total_rand$sleep_site_a == total_rand$sleep_site_b ) ] if( length( sites_rand ) != 0 ){ sites_rand_sub <- data.frame( func_a = real_final_cosleep$func_a[ row ], func_b = real_final_cosleep$func_b[ row ], rand_num = i, shared_site = sites_rand ) cosleep_sites_rand <- rbind( cosleep_sites_rand, sites_rand_sub ) } } } real_agg <- aggregate( real_final_cosleep[ , c('num', 'denom') ], by = list( real_final_cosleep$cat_a, real_final_cosleep$cat_b ), FUN = sum, na.rm = T) names( real_agg ) <- c( 'cat_a', 'cat_b', 'num', 'denom' ) real_agg$prop <- real_agg$num / real_agg$denom rand_agg <- aggregate( rand_final_cosleep[ , c('num', 'denom') ], by = list( rand_final_cosleep$cat_a, rand_final_cosleep$cat_b, rand_final_cosleep$rand_n ), FUN = sum, na.rm = T) names( rand_agg ) <- c( 'cat_a', 'cat_b', 'rand_n', 'num', 'denom' ) rand_agg$prop <- rand_agg$num / rand_agg$denom p <- sum( real_agg$prop <= rand_agg$prop ) / max( rand_agg$rand_n ) final_p <- ifelse( p > 0.5, sum( real_agg$prop >= rand_agg$prop ) / max( rand_agg$rand_n ), p ) par( bg = 'white' ) dens_rand <- density( rand_agg$prop ) plot( dens_rand$x, dens_rand$y, col = 'black', col.axis = 'black', col.lab = 'black', col.main = 'black', xlab = 'Probability of two groups sleeping at same site', type = 'l', main = '', ylab = 'Probability density', bty = 'l' ) axis(1, col = 'black', tick = T, labels = F ) axis(2, col = 'black', tick = T, labels = F ) abline( v = quantile( rand_agg$prop, c( 0.025, 0.975 ) ), col = 'black', lty = 3) abline( v = real_agg$prop, col = 'red', lty = 1 ) print( paste( 'p-value = ', round( final_p, 4) ), side = 4 ) par( bg = 'black' ) dens_rand <- density( rand_agg$prop ) plot( dens_rand$x, dens_rand$y, col = 'white', col.axis = 'white', col.lab = 'white', col.main = 'white', xlab = 'Probability of two groups sleeping at same site', type = 'l', main = '', ylab = 'Probability density', bty = 'l' ) axis(1, col = 'white', tick = T, labels = F ) axis(2, col = 'white', tick = T, labels = F ) abline( v = quantile( rand_agg$prop, c( 0.025, 0.975 ) ), col = 'white', lty = 3) abline( v = real_agg$prop, col = 'red', lty = 1 ) print( paste( 'p-value = ', round( final_p, 4) ), side = 4 )

1876b8f39dcd4d53ef6f1d0173e0cd651d33182f

cf2af9741bbf4ab0ccf83c108ac34a6300cdbeff

/plot4.R

0aaf098cd4856e46e4197f3017ec64760d781666

[]

no_license

ejsheehan/ExData_Plotting1

59e896b67b4f3488cb5cbd7af006c69bf018d717

19d04ed06721b139dbbab8930d4a0fd8d02b23c3

refs/heads/master

2021-01-24T15:24:08.958774

2015-11-05T00:46:45

45,576,730

0

null

2015-11-05T00:31:08

null

UTF-8

R

false

1,453

r

plot4.R

#read text file dat<-read.table("household_power_consumption.txt", header=TRUE, sep=";", na.strings="?", stringsAsFactors=FALSE) #create a combined Date and Time column dat$Date_Time<-paste(dat$Date, dat$Time) #identify Date and Time as date class dat$Date_Time<-strptime(dat$Date_Time, "%e/%m/%Y %H:%M:%S") #identify Date column as date class for use of subsetting dat$Date<-strptime(dat$Date, "%e/%m/%Y") #subset data to only include 2/1/2007 and 2/2/2007 dat<-dat[dat$Date=="2007-02-01" | dat$Date=="2007-02-02",] #use lubridate to name dates with abbreviated weekday name library(lubridate) dat$day<-wday(dat$Date) #open connection to png graphic device png(file="plot4.png") #prepare to add 4 plots to same graphic par(mfrow=c(2,2)) #Upper left plot plot(dat$Date_Time, dat$Global_active_power, type="l", ylab="Global Active Power (kilowatts)", xlab="") #Upper right plot plot(dat$Date_Time, dat$Voltage, type="l", xlab="datetime", ylab="Voltage") #Lower left plot plot(dat$Date_Time, dat$Sub_metering_1, type="l", ylab="Energy sub metering", xlab="") lines(dat$Date_Time, dat$Sub_metering_2, col="red") lines(dat$Date_Time, dat$Sub_metering_3, col="blue") legend(x="topright", legend=c("Sub_metering_1", "Sub_metering_2", "Sub_metering_3"), lty=c(1,1), col=c("black","red","blue")) #Lower right plot plot(dat$Date_Time, dat$Global_reactive_power, type="l", xlab="datetime", ylab="Global_reactive_power") #end connection to graphic device dev.off()

149583870d7cbae07937369ea90f6a8a6d853d71

3c3cf41378249e9f378b5a0c821e7d06dda22729

/man/LoadCleanRaw.Rd

6472ee2be4e9903018e6a6f074bed7396d84d50c

[]

no_license

kaylafrisoli/iRland

e4d89628abd57cff5cb5fe2b4d91c04d192043ac

fbb47d445679fa4eb66d4ba1832f058384946f16

refs/heads/master

2021-06-04T11:04:20.115525

2020-09-15T04:25:26

123,225,559

0

null

UTF-8

R

false

true

945

rd

LoadCleanRaw.Rd

% Generated by roxygen2: do not edit by hand % Please edit documentation in R/LoadClean.R \name{LoadCleanRaw} \alias{LoadCleanRaw} \title{Consistently load raw Irish census data} \usage{ LoadCleanRaw(path, file_delim = " ", assignID = NULL, preProcess = FALSE) } \arguments{ \item{path}{path to data} \item{file_delim}{single character used to separate fields within a record; default is " " for .txt files} \item{assignID}{string of variable names for creating a unique identifier} \item{preProcess}{has the data been preprocessed?} } \value{ data in tibble format } \description{ Consistently load and standardize raw Irish census data } \examples{ LoadCleanRaw("~/GoogleDrive/irelandData/ticknock_kayla.csv", ",", preProcess = TRUE) LoadCleanRaw("~/GoogleDrive/irelandData/census_ireland_1901/Carlow/Ticknock.txt") LoadCleanRaw("~/GoogleDrive/irelandData/census_ireland_1901/Carlow/Ticknock.txt", assignID = c("County", "DED", "Year")) }

6181a5885902e061eac4896b073810d1039dabfb

0ad326a4515ac93236aecb01c7b2b49b9c6906c5

/getTags.R

23f11b1644929be749ad33307096be41a58f8084

[]

no_license

stunglan/Rtraining

70a34b287b9d62f97dd928d15aa501cfa94b7c00

d3fdafe3a92c45ad49e8a766aab1a54cab7e057c

refs/heads/master

2021-01-22T13:58:01.618725

2014-09-18T19:42:04

null

0

null

UTF-8

R

false

285

r

getTags.R

#getTags #if (!exists('getTags_R')) { getTags_R <-TRUE setwd("~/GitHub/Rtraining") getTags <- function() { data <- read.table("data/artistTags.csv",col.names=c("artist","tag","count"),stringsAsFactors = FALSE,na.strings = "NA") data <- unique(data) # END Fix DATA } #}

9e4f42f833a89178e892a8c8bc4ad3b83818efba

ae0e9f9203adc6348a26d2195e5a09d80790ba24

/R/functions.r

5fa3e376844d3b6e29a019c1edd949ec3ef6852e

[ "MIT" ]

permissive

Moritz-Kohls/taxaEstimator

b546b9fb544a8b0116205035f2a19c134211bac3

1b93eb55db3af905268917b064f39fa9dbfb6040

refs/heads/master

2023-02-28T00:54:25.406346

2021-02-02T14:49:51

335,295,331

1

0

null

UTF-8

R

false

30,490

r

functions.r

#' Class specific positive predictive values (PPV) #' #' Step 1 of the taxa classification and estimation procedure. #' This function creates artificial reads based on NCBI viral reference genomes FASTA file, #' computes input features and artificial neural network (ANN) model #' and finally stores the trained model and classification results in R-intern variables. #' #' @seealso \code{\link{ANN_new_sample}}, \code{\link{estimation_taxa_distribution}} #' @import Biostrings #' @import keras #' @import stringi #' @param fasta.file_path FASTA file path (e.g. FASTA file viral.genomic.fna downloaded from NCBI: ftp://ftp.ncbi.nih.gov/refseq/release/viral). #' @param taxonomy.file_path Taxonomy file path (e.g. taxonomy file taxonomy_viruses_available.csv delivered within this package). #' @param temp.directory Results directory containing accuracy results, generalised confusion matrix results and accuracy as well as loss graphics of the simulation runs. #' @param count.reads_training Number of sampled viruses and artificially generated reads per virus taxonomy, e.g. order (training data). #' @param read_length Read length of all artificially generated reads. #' @param simulation_runs Simulation runs #' @examples #' # Please specify your file paths and directories! #' fasta.file_path = "~/ag_bioinf/genomes/viruses_na/refseq/viral.genomic.fna" # Download from NCBI! #' taxonomy.file_path = "inst/extdata/taxonomy_viruses_available.csv" # Relative file path #' temp.directory = "~/ag_bioinf/research/metagenomics/temp" # Results directory #' count.reads_training = 100 #' read_length = 150 #' simulation_runs = 10 #' \dontrun{ #' class_specific_PPVs ( fasta.file_path, taxonomy.file_path, temp.directory, count.reads_training, read_length ) #' } class_specific_PPVs = function ( fasta.file_path, taxonomy.file_path, temp.directory, count.reads_training, read_length, simulation_runs ) { # Taxonomy file of available viruses: taxonomy.file = read.csv(taxonomy.file_path, sep=";", stringsAsFactors=TRUE) str(taxonomy.file) taxonomy.file$NC_Id = as.character(taxonomy.file$NC_Id) taxonomy.file$Species = as.character(taxonomy.file$Species) str(taxonomy.file) table(taxonomy.file$Order, useNA = "always") # Taxonomy file of viruses with available order information: df.taxonomy_order = taxonomy.file[is.na(taxonomy.file$Order) == F,] # Fasta file: fasta.file = readDNAStringSet(fasta.file_path, format = "fasta") head(names(fasta.file)) genome_lengths = width(fasta.file) head(genome_lengths) indices = regexpr(" ",names(fasta.file)) fasta.NC_Ids = substring(names(fasta.file),1,indices-1) count.viruses_all = table(df.taxonomy_order$Order) # All viruses of the 9 different orders count.viruses_order = length(count.viruses_all) # 9 orders count.viruses_validation = count.viruses_test = round ( count.viruses_all * 0.15 ) # Number of viruses used for validation resp. test data count.viruses_training = count.viruses_all - count.viruses_validation - count.viruses_test # Number of viruses used for training data count.reads_validation = count.reads_test = round ( count.reads_training / 70 * 15 ) # Number of sampled viruses and artificially generated reads per virus taxonomy, e.g. order (validation resp. test data) count.reads_all = count.reads_training + count.reads_validation + count.reads_test # Number of sampled viruses and artificially generated reads per virus taxonomy, e.g. order (training, validation and test data) accuracy_results = rep(NA_real_,simulation_runs) confusion_matrix_results = vector("list",simulation_runs) acc_loss_graphics_results = vector("list",simulation_runs) for ( iteration in 1:(simulation_runs) ) { print(iteration) set.seed(iteration) # Split all viruses of all orders into training, validation and test data: Order_NC_Id.all = vector("list", length = count.viruses_order) for ( i in 1:count.viruses_order ) { Order_NC_Id.all [[i]] = unname(unlist(subset(df.taxonomy_order, Order == names(count.viruses_all)[i], select = NC_Id))) } names(Order_NC_Id.all) = names(count.viruses_all) Order_NC_Id.training = Order_NC_Id.validation = Order_NC_Id.test = vector("list", length = count.viruses_order) for ( i in 1:count.viruses_order ) { indices.training = sample(count.viruses_all[i], count.viruses_training[i]) indices.temp = setdiff(1:count.viruses_all[i], indices.training) indices.validation = sample(indices.temp, count.viruses_validation[i]) indices.test = setdiff(indices.temp, indices.validation) Order_NC_Id.training [[i]] = Order_NC_Id.all[[i]] [indices.training] Order_NC_Id.validation [[i]] = Order_NC_Id.all[[i]] [indices.validation] Order_NC_Id.test [[i]] = Order_NC_Id.all[[i]] [indices.test] } names(Order_NC_Id.training) = names(Order_NC_Id.validation) = names(Order_NC_Id.test) = names(count.viruses_all) # NC Ids of viruses per order. From each order, count.reads_training, count.reads_validation and count.reads_test NC Ids are sampled. order_NC_Id_sampled.training = matrix(nrow = count.viruses_order, ncol = count.reads_training) order_NC_Id_sampled.validation = matrix(nrow = count.viruses_order, ncol = count.reads_validation) order_NC_Id_sampled.test = matrix(nrow = count.viruses_order, ncol = count.reads_test) rownames(order_NC_Id_sampled.training) = rownames(order_NC_Id_sampled.validation) = rownames(order_NC_Id_sampled.test) = names(count.viruses_all) for ( i in 1:count.viruses_order ) { if ( count.viruses_training[i] >= count.reads_training ) { order_NC_Id_sampled.training [i,] = sample(Order_NC_Id.training [[i]], size = count.reads_training) } if ( count.viruses_validation[i] >= count.reads_validation ) { order_NC_Id_sampled.validation [i,] = sample(Order_NC_Id.validation [[i]], size = count.reads_validation) } if ( count.viruses_test[i] >= count.reads_test ) { order_NC_Id_sampled.test [i,] = sample(Order_NC_Id.test [[i]], size = count.reads_test) } if ( count.viruses_training[i] < count.reads_training ) { NC_Ids = rep(Order_NC_Id.training [[i]], floor(count.reads_training/count.viruses_training[i])) { if ( count.reads_training %% count.viruses_training[i] > 0 ) { order_NC_Id_sampled.training [i,] = c(NC_Ids, sample(Order_NC_Id.training [[i]], size = count.reads_training %% count.viruses_training[i])) } else if ( count.reads_training %% count.viruses_training[i] == 0 ) { order_NC_Id_sampled.training [i,] = NC_Ids } } order_NC_Id_sampled.training [i,] = sample(order_NC_Id_sampled.training [i,]) } if ( count.viruses_validation[i] < count.reads_validation ) { NC_Ids = rep(Order_NC_Id.validation [[i]], floor(count.reads_validation/count.viruses_validation[i])) { if ( count.reads_validation %% count.viruses_validation[i] > 0 ) { order_NC_Id_sampled.validation [i,] = c(NC_Ids, sample(Order_NC_Id.validation [[i]], size = count.reads_validation %% count.viruses_validation[i])) } else if ( count.reads_validation %% count.viruses_validation[i] == 0 ) { order_NC_Id_sampled.validation [i,] = NC_Ids } } order_NC_Id_sampled.validation [i,] = sample(order_NC_Id_sampled.validation [i,]) } if ( count.viruses_test[i] < count.reads_test ) { NC_Ids = rep(Order_NC_Id.test [[i]], floor(count.reads_test/count.viruses_test[i])) { if ( count.reads_test %% count.viruses_test[i] > 0 ) { order_NC_Id_sampled.test [i,] = c(NC_Ids, sample(Order_NC_Id.test [[i]], size = count.reads_test %% count.viruses_test[i])) } else if ( count.reads_test %% count.viruses_test[i] == 0 ) { order_NC_Id_sampled.test [i,] = NC_Ids } } order_NC_Id_sampled.test [i,] = sample(order_NC_Id_sampled.test [i,]) } } # Indices (positions) of sampled NC Ids in fasta file: indices_NC_Id.training = matrix(sapply(order_NC_Id_sampled.training, FUN = function(x) grep(x,fasta.NC_Ids)), nrow = count.viruses_order, ncol = count.reads_training) indices_NC_Id.validation = matrix(sapply(order_NC_Id_sampled.validation, FUN = function(x) grep(x,fasta.NC_Ids)), nrow = count.viruses_order, ncol = count.reads_validation) indices_NC_Id.test = matrix(sapply(order_NC_Id_sampled.test, FUN = function(x) grep(x,fasta.NC_Ids)), nrow = count.viruses_order, ncol = count.reads_test) # Random start positions of reads: art_reads.start_pos.training = matrix(nrow = count.viruses_order, ncol = count.reads_training) art_reads.start_pos.validation = matrix(nrow = count.viruses_order, ncol = count.reads_validation) art_reads.start_pos.test = matrix(nrow = count.viruses_order, ncol = count.reads_test) for ( i in 1:count.viruses_order ) { for ( j in 1:count.reads_training ) { art_reads.start_pos.training [i,j] = sample.int(genome_lengths [ indices_NC_Id.training[i,j] ] - read_length + 1, size = 1) } } for ( i in 1:count.viruses_order ) { for ( j in 1:count.reads_validation ) { art_reads.start_pos.validation [i,j] = sample.int(genome_lengths [ indices_NC_Id.validation[i,j] ] - read_length + 1, size = 1) } } for ( i in 1:count.viruses_order ) { for ( j in 1:count.reads_test ) { art_reads.start_pos.test [i,j] = sample.int(genome_lengths [ indices_NC_Id.test[i,j] ] - read_length + 1, size = 1) } } # Generate artificial reads by subsequencing original reads: art_reads.training = vector("list", length = count.viruses_order * count.reads_training) for ( i in 1:count.viruses_order ) { for ( j in 1:count.reads_training ) { art_reads.training[[count.reads_training*(i-1)+j]] = subseq(fasta.file[[indices_NC_Id.training[i,j]]], start = art_reads.start_pos.training[i,j], width = read_length) } } art_reads.validation = vector("list", length = count.viruses_order * count.reads_validation) for ( i in 1:count.viruses_order ) { for ( j in 1:count.reads_validation ) { art_reads.validation[[count.reads_validation*(i-1)+j]] = subseq(fasta.file[[indices_NC_Id.validation[i,j]]], start = art_reads.start_pos.validation[i,j], width = read_length) } } art_reads.test = vector("list", length = count.viruses_order * count.reads_test) for ( i in 1:count.viruses_order ) { for ( j in 1:count.reads_test ) { art_reads.test[[count.reads_test*(i-1)+j]] = subseq(fasta.file[[indices_NC_Id.test[i,j]]], start = art_reads.start_pos.test[i,j], width = read_length) } } # One-mer, two-mer and three-mer distributions and inter-nucleotide distances (4 + 16 + 64 + 36 = 120 variables). two_mer_permutations = expand.grid(c("A","C","G","T"),c("A","C","G","T")) two_mer_permutations = as.matrix(two_mer_permutations) two_mer_permutations = apply(two_mer_permutations, MARGIN = 1, FUN = function (x) paste(x,collapse="")) two_mer_permutations = sort(two_mer_permutations) three_mer_permutations = expand.grid(c("A","C","G","T"),c("A","C","G","T"),c("A","C","G","T")) three_mer_permutations = as.matrix(three_mer_permutations) three_mer_permutations = apply(three_mer_permutations, MARGIN = 1, FUN = function (x) paste(x,collapse="")) three_mer_permutations = sort(three_mer_permutations) count_input_features = 4 + 16 + 64 + 36 x_training = matrix(nrow = count.viruses_order * count.reads_training, ncol = count_input_features) for ( i in 1:(count.viruses_order*count.reads_training)) { x_training [i,1:4] = letterFrequency(art_reads.training[[i]], letters = c("A","C","G","T"), as.prob = F) for ( j in 5:20 ) { x_training [i,j] = stri_count_fixed(art_reads.training[[i]],pattern = two_mer_permutations[j-4]) } for ( j in 21:84 ) { x_training [i,j] = stri_count_fixed(art_reads.training[[i]],pattern = three_mer_permutations[j-20]) } dist_A = diff(stri_locate_all(art_reads.training[[i]], regex = "A") [[1]] [,1]) dist_A = table(factor(dist_A, levels = 2:10)) dist_C = diff(stri_locate_all(art_reads.training[[i]], regex = "C") [[1]] [,1]) dist_C = table(factor(dist_C, levels = 2:10)) dist_G = diff(stri_locate_all(art_reads.training[[i]], regex = "G") [[1]] [,1]) dist_G = table(factor(dist_G, levels = 2:10)) dist_T = diff(stri_locate_all(art_reads.training[[i]], regex = "T") [[1]] [,1]) dist_T = table(factor(dist_T, levels = 2:10)) x_training [i,85:93] = dist_A x_training [i,94:102] = dist_C x_training [i,103:111] = dist_G x_training [i,112:120] = dist_T } for ( my.index in 85:count_input_features ) { x_training[is.na(x_training[,my.index]),my.index] = 0 } x_training = apply(x_training, MARGIN = 2, FUN = function(x) (x-min(x)) / diff(range(x))) colnames(x_training) = c("A","C","G","T",two_mer_permutations,three_mer_permutations, paste0("d_A_",2:10),paste0("d_C_",2:10),paste0("d_G_",2:10),paste0("d_T_",2:10)) x_validation = matrix(nrow = count.viruses_order * count.reads_validation, ncol = count_input_features) for ( i in 1:(count.viruses_order*count.reads_validation)) { x_validation [i,1:4] = letterFrequency(art_reads.validation[[i]], letters = c("A","C","G","T"), as.prob = F) for ( j in 5:20 ) { x_validation [i,j] = stri_count_fixed(art_reads.validation[[i]],pattern = two_mer_permutations[j-4]) } for ( j in 21:84 ) { x_validation [i,j] = stri_count_fixed(art_reads.validation[[i]],pattern = three_mer_permutations[j-20]) } dist_A = diff(stri_locate_all(art_reads.validation[[i]], regex = "A") [[1]] [,1]) dist_A = table(factor(dist_A, levels = 2:10)) dist_C = diff(stri_locate_all(art_reads.validation[[i]], regex = "C") [[1]] [,1]) dist_C = table(factor(dist_C, levels = 2:10)) dist_G = diff(stri_locate_all(art_reads.validation[[i]], regex = "G") [[1]] [,1]) dist_G = table(factor(dist_G, levels = 2:10)) dist_T = diff(stri_locate_all(art_reads.validation[[i]], regex = "T") [[1]] [,1]) dist_T = table(factor(dist_T, levels = 2:10)) x_validation [i,85:93] = dist_A x_validation [i,94:102] = dist_C x_validation [i,103:111] = dist_G x_validation [i,112:120] = dist_T } for ( my.index in 85:count_input_features ) { x_validation[is.na(x_validation[,my.index]),my.index] = 0 } x_validation = apply(x_validation, MARGIN = 2, FUN = function(x) (x-min(x)) / diff(range(x))) colnames(x_validation) = c("A","C","G","T",two_mer_permutations,three_mer_permutations, paste0("d_A_",2:10),paste0("d_C_",2:10),paste0("d_G_",2:10),paste0("d_T_",2:10)) x_test = matrix(nrow = count.viruses_order * count.reads_test, ncol = count_input_features) for ( i in 1:(count.viruses_order*count.reads_test)) { x_test [i,1:4] = letterFrequency(art_reads.test[[i]], letters = c("A","C","G","T"), as.prob = F) for ( j in 5:20 ) { x_test [i,j] = stri_count_fixed(art_reads.test[[i]],pattern = two_mer_permutations[j-4]) } for ( j in 21:84 ) { x_test [i,j] = stri_count_fixed(art_reads.test[[i]],pattern = three_mer_permutations[j-20]) } dist_A = diff(stri_locate_all(art_reads.test[[i]], regex = "A") [[1]] [,1]) dist_A = table(factor(dist_A, levels = 2:10)) dist_C = diff(stri_locate_all(art_reads.test[[i]], regex = "C") [[1]] [,1]) dist_C = table(factor(dist_C, levels = 2:10)) dist_G = diff(stri_locate_all(art_reads.test[[i]], regex = "G") [[1]] [,1]) dist_G = table(factor(dist_G, levels = 2:10)) dist_T = diff(stri_locate_all(art_reads.test[[i]], regex = "T") [[1]] [,1]) dist_T = table(factor(dist_T, levels = 2:10)) x_test [i,85:93] = dist_A x_test [i,94:102] = dist_C x_test [i,103:111] = dist_G x_test [i,112:120] = dist_T } for ( my.index in 85:count_input_features ) { x_test[is.na(x_test[,my.index]),my.index] = 0 } x_test = apply(x_test, MARGIN = 2, FUN = function(x) (x-min(x)) / diff(range(x))) colnames(x_test) = c("A","C","G","T",two_mer_permutations,three_mer_permutations, paste0("d_A_",2:10),paste0("d_C_",2:10),paste0("d_G_",2:10),paste0("d_T_",2:10)) # Categories to predict (Virus orders 1 to 9, here indices 0 to 8): y_training = rep(0:(count.viruses_order-1),each = count.reads_training) y_validation = rep(0:(count.viruses_order-1),each = count.reads_validation) y_test = rep(0:(count.viruses_order-1),each = count.reads_test) df.order_id = data.frame(Order = names(count.viruses_all), Id = 0:(count.viruses_order-1)) # Build the model. At first, setup the layers: model <- keras_model_sequential() model %>% layer_dense(units = 64, activation = 'relu', input_shape = c(count_input_features)) %>% layer_dense(units = count.viruses_order, activation = 'softmax') # Compile the model: model %>% compile( optimizer = 'adam', loss = 'sparse_categorical_crossentropy', metrics = c('accuracy') ) # Train the model: history <- model %>% fit( x_training, y_training, validation_data = list(x_validation, y_validation), verbose = 0, shuffle = T, epochs = 100, batch_size = floor(sqrt(nrow(x_training)))^2, callbacks = callback_early_stopping(monitor = "val_loss", patience = 10, verbose = 0, restore_best_weights = T) ) # Evaluate accuracy: score <- model %>% evaluate(x_test, y_test) cat('Test loss:', score$loss, "\n") cat('Test accuracy:', score$acc, "\n") # Predict classes: predictions = model %>% predict_classes(x_test) (predictions = factor(predictions, levels = 0:(count.viruses_order-1))) Confusion.Matrix = data.frame(matrix(nrow = count.viruses_order, ncol = count.viruses_order+6)) colnames(Confusion.Matrix) = c("TPR","TNR","PPV","NPV","Order","Order_ID",paste("Order",0:(count.viruses_order-1),sep = "_")) Confusion.Matrix$Order = names(count.viruses_all) Confusion.Matrix$Order_ID = 0:(count.viruses_order-1) for ( i in 0:(count.viruses_order-1) ) { my.indices = which(y_test == i ) my.row = table(predictions[my.indices]) Confusion.Matrix [i+1,-(1:6)] = my.row } total_sum = sum(Confusion.Matrix[,-(1:6)]) for ( i in 0:(count.viruses_order-1) ) { my.matrix = Confusion.Matrix[,-(1:6)] TP = my.matrix [i+1,i+1] TN = sum(my.matrix [-(i+1),-(i+1)]) C_P = sum(my.matrix[i+1,]) C_N = sum(my.matrix[-(i+1),]) P_C_P = sum(my.matrix[,i+1]) P_C_N = sum(my.matrix[,-(i+1)]) TPR = TP / C_P TNR = TN / C_N PPV = TP / P_C_P NPV = TN / P_C_N Confusion.Matrix [i+1,1:4] = round(100*c(TPR,TNR,PPV,NPV),1) } Confusion.Matrix accuracy_results[iteration] = score$acc confusion_matrix_results[[iteration]] = Confusion.Matrix acc_loss_graphics_results[[iteration]] = history } saveRDS(accuracy_results, paste0(temp.directory,"/accuracy_results.rds")) saveRDS(confusion_matrix_results, paste0(temp.directory,"/confusion_matrix_results.rds")) saveRDS(acc_loss_graphics_results, paste0(temp.directory,"/acc_loss_graphics_results.rds")) } #' Artificial neural network (ANN) classification of a new sample #' #' Step 2 of the taxa classification and estimation procedure. #' This function loads an ANN model which was trained on artificial data, #' computes input features of the new, adjusted sample file (FASTQ or SAM) #' and stores the predicted classes of its read sequences. #' #' @seealso \code{\link{class_specific_PPVs}}, \code{\link{estimation_taxa_distribution}} #' @import Biostrings #' @import keras #' @import stringi #' @param fasta.file_path FASTA file path (e.g. FASTA file viral.genomic.fna downloaded from NCBI: ftp://ftp.ncbi.nih.gov/refseq/release/viral). #' @param taxonomy.file_path Taxonomy file path (e.g. taxonomy file taxonomy_viruses_available.csv delivered within this package). #' @param temp.directory Results directory containing accuracy results, generalised confusion matrix results and accuracy as well as loss graphics of the simulation runs. #' @param read_sequences.file_path Only read sequences without identifier or species names, extracted from FASTQ or SAM file! #' @param model.file_path File path of ANN model trained on artificially generated data. #' @param predictions.file_path File path of the result file of predicted taxonomic orders. #' @examples #' # Please specify your file paths and directories! #' fasta.file_path = "~/ag_bioinf/genomes/viruses_na/refseq/viral.genomic.fna" # Download from NCBI! #' taxonomy.file_path = "inst/extdata/taxonomy_viruses_available.csv" # Relative file path #' temp.directory = "~/ag_bioinf/research/metagenomics/temp" # Results directory #' read_sequences.file_path = "~/ag_bioinf/research/metagenomics/Data/Seehund_Mapping/read_sequences.txt" #' model.file_path = "inst/extdata/model_training_1_dataset.h5" #' predictions.file_path = "~/ag_bioinf/research/metagenomics/temp/900000_training_samples_1_iteration/test/predictions_seal_sample.rds" #' \dontrun{ #' ANN_new_sample ( fasta.file_path, taxonomy.file_path, temp.directory, read_sequences.file_path ) #' } ANN_new_sample = function ( ) { read_sequences = readLines(read_sequences.file_path) # read_sequences = read_sequences [1:100] read_sequences.count = length(read_sequences) seq_len = unname(sapply(read_sequences,nchar)) sequences = DNAStringSet(read_sequences) # One-mer, two-mer and three-mer distributions and inter-nucleotide distances (4 + 16 + 64 + 36 = 120 variables). two_mer_permutations = expand.grid(c("A","C","G","T"),c("A","C","G","T")) two_mer_permutations = as.matrix(two_mer_permutations) two_mer_permutations = apply(two_mer_permutations, MARGIN = 1, FUN = function (x) paste(x,collapse="")) two_mer_permutations = sort(two_mer_permutations) three_mer_permutations = expand.grid(c("A","C","G","T"),c("A","C","G","T"),c("A","C","G","T")) three_mer_permutations = as.matrix(three_mer_permutations) three_mer_permutations = apply(three_mer_permutations, MARGIN = 1, FUN = function (x) paste(x,collapse="")) three_mer_permutations = sort(three_mer_permutations) count_input_features = 4 + 16 + 64 + 36 x_test = matrix(nrow = read_sequences.count, ncol = count_input_features) for ( i in 1:read_sequences.count) { if ( i %% 100 == 0 ) print(round(i/read_sequences.count*100,2)) x_test [i,1:4] = letterFrequency(sequences[[i]], letters = c("A","C","G","T"), as.prob = F) for ( j in 5:20 ) { x_test [i,j] = stri_count_fixed(sequences[[i]],pattern = two_mer_permutations[j-4]) } for ( j in 21:84 ) { x_test [i,j] = stri_count_fixed(sequences[[i]],pattern = three_mer_permutations[j-20]) } dist_A = diff(stri_locate_all(sequences[[i]], regex = "A") [[1]] [,1]) dist_A = table(factor(dist_A, levels = 2:10)) dist_C = diff(stri_locate_all(sequences[[i]], regex = "C") [[1]] [,1]) dist_C = table(factor(dist_C, levels = 2:10)) dist_G = diff(stri_locate_all(sequences[[i]], regex = "G") [[1]] [,1]) dist_G = table(factor(dist_G, levels = 2:10)) dist_T = diff(stri_locate_all(sequences[[i]], regex = "T") [[1]] [,1]) dist_T = table(factor(dist_T, levels = 2:10)) x_test [i,85:93] = dist_A x_test [i,94:102] = dist_C x_test [i,103:111] = dist_G x_test [i,112:120] = dist_T } for ( my.index in 85:count_input_features ) { x_test[is.na(x_test[,my.index]),my.index] = 0 } x_test = apply(x_test, MARGIN = 2, FUN = function(x) (x-min(x)) / diff(range(x))) colnames(x_test) = c("A","C","G","T",two_mer_permutations,three_mer_permutations, paste0("d_A_",2:10),paste0("d_C_",2:10),paste0("d_G_",2:10),paste0("d_T_",2:10)) # Load your previously on artificially generated reads trained ANN model ("e.g. model_training_1_dataset.h5"): model = load_model_hdf5(model.file_path) predictions = model %>% predict_classes(x_test) # Save the classification results of your model: saveRDS(predictions,predictions.file_path) } #' Prior and posterior estimation of taxa distribution #' #' Step 3 of the taxa classification and estimation procedure. #' This function loads the predicted classes of the ANN model which was trained on artificial data, #' computes prior as well as posterior taxa distribution estimations #' and saves a graphics result file containing the estimation of the predicted classes. #' #' @seealso \code{\link{class_specific_PPVs}}, \code{\link{ANN_new_sample}} #' @import Biostrings #' @import keras #' @import stringi #' @import ggplot2 #' @import gridExtra #' @import ggpubr #' @import ggplotify #' @param temp.directory Results directory containing accuracy results, generalised confusion matrix results and accuracy as well as loss graphics of the simulation runs. #' @param graphics.directory Directory of graphics result file. #' @param a_priori_table.file_path Predictions of previously classified reads of new sample file. #' @examples #' # Please specify your file paths and directories! #' temp.directory = "inst/extdata/class_specific_PPVs_results/" # Results directory #' graphics.directory = "~/ag_bioinf/research/metagenomics/ManuscriptNeuralNet/Graphics/" #' a_priori_table.file_path = "inst/extdata/predictions_3154562.rds" #' \dontrun{ #' estimation_taxa_distribution ( temp.directory, graphics.directory, a_priori_table.file_path ) #' } estimation_taxa_distribution = function ( ) { estimation.a_priori.table = readRDS(a_priori_table.file_path) estimation.a_priori.table = estimation.a_priori.table[-10] confusion_matrix_results = readRDS(paste0(temp.directory,"confusion_matrix_results.rds")) loss_acc_results = readRDS(paste0(temp.directory,"loss_acc_results.rds")) loss_acc_graphics_results = readRDS(paste0(temp.directory,"loss_acc_graphics_results.rds")) order_names = c("Bunyavirales","Caudovirales","Herpesvirales","Ligamenvirales","Mononegavirales", "Nidovirales","Ortervirales","Picornavirales","Tymovirales") TPR = matrix(NA_real_, nrow = 9, ncol = 10) PPV = matrix(NA_real_, nrow = 9, ncol = 10) for ( i in 1:10 ) { TPR[,i] = confusion_matrix_results [[i]] [,1] PPV[,i] = confusion_matrix_results [[i]] [,3] } rownames(TPR) = rownames(PPV) = order_names df.TPR_and_PPV = data.frame(Order = rep(order_names, each = 10), TPR = as.vector(t(TPR)), PPV = as.vector(t(PPV))) estimation.a_priori = as.data.frame(matrix(nrow = 9, ncol = 12)) colnames(estimation.a_priori) = c("Order","Order_ID,",paste0("Iteration_",1:10)) estimation.a_priori[,1] = order_names estimation.a_priori[,2] = 0:8 for ( j in 3:12 ) { estimation.a_priori[,j] = unlist(estimation.a_priori.table) } estimation.a_posteriori = estimation.a_priori PPV_probabilities = vector("list",10) for ( j in 1:10 ) { PPV_probabilities[[j]] = matrix(NA_real_,9,9) rownames(PPV_probabilities[[j]]) = paste0("Order_",0:8) colnames(PPV_probabilities[[j]]) = paste0("Order_",0:8) } for ( i in 1:9 ) { for ( j in 1:10 ) { PPV_probabilities[[j]] [,i] = confusion_matrix_results[[j]] [,i+6] / sum(confusion_matrix_results[[j]] [,i+6]) } } PPV_probabilities.mean = vector("list",10) for ( j in 1:10 ) { PPV_probabilities.mean[[j]] = matrix(NA_real_,9,9) rownames(PPV_probabilities.mean[[j]]) = paste0("Order_",0:8) colnames(PPV_probabilities.mean[[j]]) = paste0("Order_",0:8) for ( i in 1:9 ) { temp = sapply(PPV_probabilities[-j], FUN = function(x) x[,i]) PPV_probabilities.mean[[j]] [,i] = rowMeans(temp) } } for ( j in 1:10 ) { estimation.a_posteriori[,j+2] = PPV_probabilities.mean[[j]] %*% matrix(estimation.a_priori[,j+2],ncol = 1) estimation.a_posteriori[,j+2] = PPV_probabilities[[j]] %*% matrix(estimation.a_priori[,j+2],ncol = 1) } my.data_frame = data.frame(Order = rep(rep(order_names, each = 10),2), Estimation = c(rep("Prior",9*10),rep("Posterior",9*10)), Iteration = rep(1:10, 2*9), Count = c(as.vector(t(as.matrix(estimation.a_priori[,-(1:2)]))), as.vector(t(as.matrix(estimation.a_posteriori[,-(1:2)]))))) my.data_frame$Order = as.factor(my.data_frame$Order) my.data_frame$Estimation = factor(my.data_frame$Estimation, levels = c("Prior","Posterior")) my.data_frame$Iteration = as.factor(my.data_frame$Iteration) my.data_frame.relative = my.data_frame for ( i in 1:10 ) { my.data_frame.relative[seq(0+i,90,by=10),4] = my.data_frame.relative[seq(0+i,90,by=10),4] / sum(my.data_frame.relative[seq(0+i,90,by=10),4]) * 100 my.data_frame.relative[seq(90+i,180,by=10),4] = my.data_frame.relative[seq(90+i,180,by=10),4] / sum(my.data_frame.relative[seq(90+i,180,by=10),4]) * 100 } my.data_frame.relative$Order = factor(my.data_frame.relative$Order, levels = rev(c("Bunyavirales","Caudovirales","Herpesvirales","Ligamenvirales","Mononegavirales", "Nidovirales","Ortervirales","Picornavirales","Tymovirales") )) require("ggplot2") .df <- data.frame(x = my.data_frame.relative$Order, y = my.data_frame.relative$Count, z = my.data_frame.relative$Estimation) .plot <- ggplot(data = .df, aes(x = factor(x), y = y, colour = z)) + stat_summary(fun.y = "mean", geom = "point", position = position_dodge(width = 0.6) ) + stat_summary(fun.data = "mean_cl_boot", geom = "errorbar", position = position_dodge(width = 0.6), pch = 10, size = 1, width = 0.1, fun.args = list(conf.int = 1.0)) + coord_flip() + xlab("Order") + ylab("Relative frequency (%)") + labs(colour = "Estimation") + theme_bw(base_size = 25, base_family = "sans") print(.plot) ggsave(filename = paste0(graphics.directory,"prior_posterior_estimation.pdf"), plot = .plot, width = 12, height = 8) rm(.df, .plot) dev.off() }

097d2ac5c0fb984f516a96c4bdfd498f155619f4

1f5bc0ade472404258b43525269d15f5e1b543f7

/R/CreateBasis.R

d8e05574f66138f60cd736e3c71ff9607db72c35

[]

no_license

mathchin/fdapace

d84d393d731a0fe64f7b3e15419d6637a524f490

67728c2d6be6f27bdfdb2eb97257b2c881847266

refs/heads/master

2021-01-13T13:32:24.601040

2016-07-15T10:19:26

72,411,264

0

1

null

2016-10-31T07:02:22

null

UTF-8

R

false

1,030

r

CreateBasis.R

# Create an orthogonal basis of K functions in [0, 1], with nGrid points. # Output: a K by nGrid matrix, each column containing an basis function. CreateBasis <- function(K, pts=seq(0, 1, length.out=50), type='sin') { nGrid <- length(pts) possibleTypes <- c('cos', 'sin', 'fourier', 'unknown') type <- possibleTypes[pmatch(type, possibleTypes, nomatch=length(possibleTypes))] stopifnot(is.numeric(K) && length(K) == 1 && K > 0) if (type == 'cos') { sapply(seq_len(K), function(k) if (k == 1) { rep(1, nGrid) } else { sqrt(2) * cos((k - 1) * pi * pts) } ) } else if (type == 'sin') { sapply(seq_len(K), function(k) sqrt(2) * sin(k * pi * pts)) } else if (type == 'fourier') { sapply(seq_len(K), function(k) if (k == 1) { rep(1, nGrid) } else if (k %% 2 == 0) { sqrt(2) * sin(k * pi * pts) } else { sqrt(2) * cos((k - 1) * pi * pts) } ) } else if (type == 'unknown') { stop('unknown basis type') } }

9c63a940a75211ff633909a7fe8c6e122bf40577

b914a79b9cc2f66614fb4f6ab784c8649901229e

/Liger/spatial.RNA.plot.a.gene.R

4156cb22ad131e68802636792d92854a8a13092a

[]

no_license

explorerwjy/ML_genomics

731111729e076038673851dc14170c2a03898295

91acbef43a49ee48205e088fac7c78a878a03752

refs/heads/main

2023-04-23T07:46:50.273596

2021-04-27T13:15:49

361,255,171

0

null

UTF-8

R

false

1,750

r

spatial.RNA.plot.a.gene.R

spatial.RNA.plot.a.gene <- function(spatial.RNA.filename, genename, log.normalize=TRUE, savefilename="tmp.pdf") { library(ggplot2) library(ggthemes) library(ggeasy) spatialRNA <- read.table(spatial.RNA.filename, header=T, sep="\t", row.names = 1) LogNormalize <- function(M, scale = 1e5, normalize = TRUE){ if(normalize){ MM <- apply(M, 2, function(x){ if(sum(x) > 0){ x/sum(x) }else{ x } }) } else { MM <- M } MM <- scale * MM MM <- log10(MM + 1) } if (!genename %in% row.names(spatialRNA)) { message(paste0(genename, " not in ", spatial.RNA.filename)) return(NULL) } else { spatialRNA.coordinate <- matrix(NA, nrow = dim(spatialRNA)[2], ncol = 2) for (i in 1:dim(spatialRNA)[2]) { spatialRNA.coordinate[i, 1] = as.integer(strsplit(strsplit(colnames(spatialRNA)[i], split = "x")[[1]][1], split = "X")[[1]][2]) spatialRNA.coordinate[i, 2] = as.integer(strsplit(colnames(spatialRNA)[i], split = "x")[[1]][2]) } # first log normalize, if necessary if (log.normalize) { spatialRNA <- LogNormalize(spatialRNA) } to.plot <- data.frame(expression = spatialRNA[match(genename, rownames(spatialRNA)),], X=spatialRNA.coordinate[,1], Y=spatialRNA.coordinate[,2]) mid<-1/2*(min(to.plot$expression)+max(to.plot$expression)) p <- ggplot(to.plot, aes(x=X, y=Y, col=expression)) + geom_point(alpha=0.5, ) + xlab("X") + ylab("Y") + ggtitle(genename) + theme_classic() + ggeasy::easy_center_title() scale_color_gradient2(midpoint=mid, low="blue", mid="white", high="red") ggsave(savefilename) } }

77d944dd3649ee565a68c5b49309fdbde1fcee39

ce0db4deeb74d83cb4704a0ab95134ddd40a77f6

/gr.R

b17962dd5f8e090005fbc2ee95d892f41d02e980

[]

no_license

lufanl/GrowthRespiration

ff313cd05ae1434889ae14fb57c4f773fa285a13

cdf51e50a6cc282c0fbc0e813f4445a7269e46c8

refs/heads/master

2020-04-06T04:57:21.108303

2014-05-31T01:53:03

null

0

null

UTF-8

R

false

19,584

r

gr.R

require(PEcAn.all) logger.setQuitOnSevere(FALSE) settings <- read.settings("gr.settings.xml") td <- get.trait.data(settings$pfts,settings$run$dbfiles,settings$database,TRUE) ## rescale trait data trait.file = file.path(settings$pfts$pft$outdir, 'trait.data.Rdata') load(trait.file) for(i in 1:length(trait.data)){ trait.data[[i]]$mean = trait.data[[i]]$mean/100 trait.data[[i]]$stat = trait.data[[i]]$stat/100 } save(trait.data,file=trait.file) ##PEcAn - get posterior priors run.meta.analysis(td, settings$meta.analysis$iter, settings$run$dbfiles, settings$database) load(file.path(settings$pfts$pft$outdir,"trait.mcmc.Rdata")) load(file.path(settings$pfts$pft$outdir,"post.distns.Rdata")) ######################################### c <- read.csv("cost.csv") cost <- c$CO2Produced NC <- length(cost) # #Components ## Convert gCO2 to gC ## gCO2*(12gC/44gCO2) cost = cost*(12/44) leafvariables = c('l_carbohydrates','l_lignin','l_lipids','l_minerals','l_organicacids','l_protein') stemvariables = c('s_carbohydrates','s_lignin','s_lipids','s_minerals','s_organicacids','s_protein') rootvariables = c('r_carbohydrates','r_lignin','r_lipids','r_minerals','r_organicacids','r_protein') variables=matrix(c(leafvariables,stemvariables,rootvariables),NC,3) NV=length(variables) #Number of variables ########################################################## ##FUNCTION FOR ANALYSIS ## function input: trait data and construction costs ## function outopt: list of 4 matrices of growth respiration values for uninformative prior, leaf, root, and stem ########################################################### getdistribution <- function(trait.mcmc,post.distns,cost,variables) { NC=length(cost) NV=length(variables) #R = Rl + Rs + Rr #leaf #Rl = kl*Gl #kl = cost (g C produced) * pcompl (percent composition of leaf components) ## calc mean and sd from meta-analysis mean = matrix(NA,NC,3) var = matrix(NA,NC,3) for(i in 1:NV){ if(variables[i] %in% names(trait.mcmc)){ y = as.matrix((trait.mcmc[[variables[i]]]))[,"beta.o"] mean[i]= mean(y) var[i]= var(y) } else { ## use the prior row = which(rownames(post.distns) == variables[i]) if(length(row)==1 & post.distns$distn[row] == 'beta'){ x = post.distns[row,] mean[i] = x$parama/(x$parama+x$paramb) var[i] = (x$parama*x$paramb)/((x$parama+x$paramb)^2*(x$parama+x$paramb+1)^2) } } } ## moment matching to est. alpha # USING DIRICHLET: # mean[i]=a[i]/a0 # var[i]=a[i]*(a0-a[i])/((a0)^2*(a0+1)) # a = matrix(NA,NC,3) # for(i in 1:length(variables)){ # a[i]=mean[i]*(((mean[i]-mean[i]^2)/var[i])-1) # } # USING BETA # E[x]=M=a/(a+B) # B=a(1-M)/M # Var[X]=aB/[(a+B)^2(a+B+1)] a=B=matrix(NA,NC,3) for(i in 1:NV) { a[i]=(1-mean[i])*mean[i]^2/var[i]-mean[i] B[i]= a[i]*(1-mean[i])/mean[i] } ########## functions to rescale percent composition to sum to 1 ############# NewP.oldDoesntWork <- function(k,p,a,b){ # calculate current quantile q0 = pbeta(p,a,b) qm = pbeta(a/(a+b),a,b) # adjust by k qnew = qm + k*(q0-qm) qnew[qnew<0] = 0 qnew[qnew>1] = 1 # convert back to p pnew = qbeta(qnew,a,b) return(pnew) } NewP <- function(k,p,a,b){ # calculate current quantile q0 = pbeta(p,a,b) # calc SD equivalent of current quantile sd0 = qnorm(q0) # adjust by k sd.new = sd0 + k # calc new quantile q.new = pnorm(sd.new) # convert back to p pnew = qbeta(q.new,a,b) return(pnew) } SumToOneFactor <- function(k,p,a,b){ pnew = NewP(k,p,a,b) # assess sum to 1 return((sum(pnew)-1)^2) } N = 5000 # Iterations ## l=leaf; s=stem; r=root; nd=assuming no parameter data G=Gl=Gs=Gr=matrix(1,N,1) Rl=Rs=Rr=Rnd=matrix(NA,N,1) pcompl=pcomps=pcompr=pcompnd=matrix(NA,N,NC) #storage for % composition kl=ks=kr=knd=matrix(NA,N,1) #cost*%composition # get percent composition using alpha and beta for(i in 1:N){ # rdirichlet(1,c(,1,1,1,1,1)) # pcompl[i,]=rdirichlet(1,c(a[,1])) # pcomps[i,]=rdirichlet(1,c(a[,2])) # pcompr[i,]=rdirichlet(1,c(a[,3])) for (j in 1:NC) { pcompnd[i,j]=rbeta(1,1,5) pcompl[i,j]=rbeta(1,a[j,1],B[j,1]) pcomps[i,j]=rbeta(1,a[j,2],B[j,2]) pcompr[i,j]=rbeta(1,a[j,3],B[j,3]) } ## Rescale pcomp output so sums to 1 kopt = optimize(SumToOneFactor,c(-10,10),p=pcompnd[i,],a=1,b=6) popt = NewP(kopt$minimum,p=pcompnd[i,],a=1,b=6) koptl = optimize(SumToOneFactor,c(-10,10),p=pcompl[i,],a=a[,1],b=B[,1]) poptl = NewP(koptl$minimum,p=pcompl[i,],a=a[,1],b=B[,1]) kopts = optimize(SumToOneFactor,c(-10,10),p=pcomps[i,],a=a[,2],b=B[,2]) popts = NewP(kopts$minimum,p=pcomps[i,],a=a[,2],b=B[,2]) koptr = optimize(SumToOneFactor,c(-10,10),p=pcompr[i,],a=a[,3],b=B[,3]) poptr = NewP(koptr$minimum,p=pcompr[i,],a=a[,3],b=B[,3]) knd[i,]=sum(cost*popt) kl[i,]=sum(cost*poptl) ks[i,]=sum(cost*popts) kr[i,]=sum(cost*poptr) if(i %% 1000 == 0) print(i) } # Calculate growth respiration for leaf, stem, and root Rnd=knd*G ## UNINFORMATIVE PRIOR; no percent composition data Rl=kl*Gl Rs=ks*Gs Rr=kr*Gr R<- list("Rnd"=Rnd,"Rl"=Rl,"Rs"=Rs,"Rr"=Rr,"var"=var) return(R) } #end of function ########################################################################## R.allplants <- getdistribution(trait.mcmc,post.distns,cost,variables) ########## Create Plot of Distributions ################################## cols = 1:4 dRnd = density(R.allplants$Rnd) plot(density(R.allplants$Rl),xlim=range(dRnd$x),col=cols[2]) lines(dRnd,col=cols[1]) lines(density(R.allplants$Rs),col=cols[3]) lines(density(R.allplants$Rr),col=cols[4]) legend("topright",legend=c("Null","Leaf","Stem","Root"),col=cols,lwd=2) ########### Variance Decomposition #################################################### ## sum(Pcomp^2*Var(cost) + sum(cost^2*Var(Pcomp)) ## no variance in construction costs vd = matrix(NA,NC,3) for (i in 1:NC){ vd[i,1]=cost[i]^2*var(pcompl[,i]) vd[i,2]=cost[i]^2*var(pcomps[,i]) vd[i,3]=cost[i]^2*var(pcompr[,i]) } ## alternative that doesn't have sum to 1 constraints for (i in 1:NC){ vd[i,1]=cost[i]^2*R.allplants$var[i,1] vd[i,2]=cost[i]^2*R.allplants$var[i,2] vd[i,3]=cost[i]^2*R.allplants$var[i,3] } colnames(vd) <- c("leaf","stem","root") rownames(vd) <- c("carb","lignin","lipid","mineral","OA","protein") totvar <- apply(vd,2,sum) t(vd)/totvar ## % variance totsd <- apply(sqrt(vd),2,sum) t(sqrt(vd))/totsd *100 ## % sd ########################################################## ## Build covariates table ctable=matrix(NA,0,NV) colnames(ctable) <-variables for(i in 1:length(variables)){ ##Find variable in trait data if(variables[i]%in%names(trait.data)){ tr=which(names(trait.data)==variables[i]) ##Create unique ID for trait v=paste(trait.data[[tr]]$specie_id,trait.data[[tr]]$site_id,sep="#") for(j in 1:length(v)){ ##if ID is already has a row in the table if(v[j]%in%rownames(ctable)){ rownumber=which(rownames(ctable)==v[j]) if(is.na(ctable[rownumber,i])) { ctable[rownumber,i]=trait.data[[i]]$mean[j] } else { ####But if space in table already full ##average current and new value ctable[rownumber,i]==mean(c(ctable[rownumber,i],trait.data[[i]]$mean[j])) } } else{ ##if ID is new newrow=matrix(NA,1,length(variables)) rownames(newrow)=v[j] newrow[,i]=trait.data[[i]]$mean[j] ctable=rbind(ctable,newrow) } } } } ## fit missing data model to estimate NAs MissingData = " model{ for(i in 1:n){ x[i,] ~ dmnorm(mu,tau) } mu ~ dmnorm(m0,t0) tau ~ dwish(R,k) x[2,4] <- xmis xmis ~ dnorm(0.2,1) } " w = ncol(ctable) data <- list(x = ctable,n=nrow(ctable),m0=rep(1/6,w),t0 = diag(1,w),R = diag(1e-6,w),k=w) #test #w = 4 #data <- list(x = ctable[1:2,1:w],n=2,m0=rep(1/6,w),t0 = diag(1,w),R = diag(1e-6,w),k=w) j.model = jags.model(file=textConnection(MissingData), data = data, n.chains=1, n.adapt=10, inits = list(xmis = 0.1)) logit <- function(p){ log(p/(1-p)) } ilogit <- function(x){ exp(x)/(1+exp(x)) } Z = logit(ctable) m = nrow(ctable) ###set up covariates Zorig <- as.matrix(Z) ncov <- ncol(as.matrix(Z)) #find Zobs Zobs <- apply(Z,2,mean,na.rm=TRUE) ## HACK## if(is.nan(Zobs[10])) Zobs[10] = Zobs[4] if(is.nan(Zobs[11])) Zobs[11] = Zobs[5] if(is.nan(Zobs[13])) Zobs[13] = Zobs[1] if(is.nan(Zobs[16])) Zobs[16] = Zobs[4] if(is.nan(Zobs[17])) Zobs[17] = Zobs[5] if(is.nan(Zobs[18])) Zobs[18] = Zobs[12] n.Z <- nrow(Z) for(i in 1:ncov){ Z[is.na(Zorig[,i]),i] <- Zobs[i] } ## initial guess Z.init = ilogit(Z) #priors for Zmis #mean mu muZ.ic <- Zobs mu.Z0 <- rep(logit(1/6),ncol(Z)) #post-normalization M.Z0 <- diag(rep(10,ncol(Z))) IM.Z0 <- solve(M.Z0) #cov V V.Z.ic <- diag(cov(Z,use="pairwise.complete.obs")) x.Z <- ncov + 2 V.Z0.all <- M.Z0*x.Z V.Z0 <- diag(V.Z0.all) IV.Z0 <- solve(V.Z0.all) mu.Z <- mu.Z0 V.Z <- V.Z0.all IV.Z <- solve(V.Z) library(MCMCpack) library(mvtnorm) ## set storage start = 1 ngibbs = 500 muZgibbs <- matrix(0,nrow=ngibbs,ncol=ncov) VZgibbs <- matrix(0,nrow=ngibbs,ncol=ncov*(ncov+1)/2) Zgibbs <- Z*0 #gibbs loop btimes <- 0 for(g in start:ngibbs){ print(g) ##missing Z's - mean bigv <- try(solve(n.Z*IV.Z + IM.Z0)) if(is.numeric(bigv)){ smallv <- apply(Z %*% IV.Z,2,sum) + IM.Z0 %*% mu.Z0 mu.Z <- rmvnorm(1,bigv %*% smallv,bigv) } muZgibbs[g,] <- mu.Z ##missing Z's - Variance u <- 0 for(i in 1:m){ u <- u + crossprod(Z[i,]-mu.Z) } V.Z.orig <- V.Z IV.Z.orig <- IV.Z V.Z <- riwish(x.Z + n.Z, V.Z0.all + u) IV.Z <- try(solve(V.Z)) if(!is.numeric(IV.Z)){ IV.Z <- IV.Z.orig V.Z <- V.Z.orig } VZgibbs[g,] <- vech(V.Z) ##missing Z's - draw missing values for(i in 1:m){ for(j in 1:ncov){ if(is.na(Zorig[i,j])){ bigv <- 1/IV.Z[j,j] smallv <- mu.Z[j]*IV.Z[j,j] zcols <- 1:ncov; zcols <- zcols[zcols != j] for(k in zcols){ smallv <- smallv + (Z[i,k] - mu.Z[k])*IV.Z[k,j] } Z[i,j] <- rnorm(1,bigv * smallv, sqrt(bigv)) } } } Zgibbs = Zgibbs + Z if(g %% 500 == 0){ save.image("GR.RData")} } #end Z.fillmissing Zbar = ilogit(Zgibbs/g) sum(is.na(Zbar)) cbind(apply(Zbar,2,mean),apply(Z.init,2,mean), apply(ctable,2,mean,na.rm=TRUE)) pdf("muZgibb.pdf") plot(as.mcmc(ilogit(muZgibbs))) dev.off() ################################################################## ## PCA & Cluster Analysis data.leaf <- Z.init[,1:6] data.stem <- Z.init[,7:12] data.root <- Z.init[,13:18] ## cluster analysis on raw leaf data cluster.leaf <- kmeans(data.leaf,2) plot(Z.init[,2],Z.init[,3]) plot(Z.init[,2],Z.init[,3],col=cluster.leaf$cluster) cluster.stem <- kmeans(data.stem,2) cluster.root <- kmeans(data.root,2) ## cluster analysis on leaf data weighted by construction costs cluster.leaf.cost <- kmeans(t(t(data.leaf)*cost),2) cluster.stem.cost <- kmeans(t(t(data.stem)*cost),2) cluster.root.cost <- kmeans(t(t(data.root)*cost),2) plot(Z.init[,2],Z.init[,3],col=cluster.leaf.cost$cluster) ## principal component analysis on raw leaf data pca.leaf <- prcomp(data.leaf,retx=TRUE) pca.leaf$sdev/sum(pca.leaf$sdev)*100 plot(pca.leaf) plot(pca.leaf$x[,1],pca.leaf$x[,2]) ## principal component analysis on leaf data weighed by construction costs pca.leaf.cost <- prcomp(data.leaf,scale=cost,retx=TRUE) ## cluster.pca.leaf <- kmeans(t(t(pca.leaf$x)*pca.leaf$sdev^2),2) plot(pca.leaf$x[,1],pca.leaf$x[,2],col=cluster.pca.leaf$cluster) cluster.pca.leaf <- kmeans(t(t(pca.leaf$x)*pca.leaf$sdev^2),2) phenol = rbinom(nrow(pca.leaf$x),1,0.5) ## replace this with real data ## phenol.char = c("E","D") plot(pca.leaf$x[,1],pca.leaf$x[,2],col=cluster.pca.leaf$cluster) library(MASS) library(vegan) library("RPostgreSQL") dbparms <- list(driver="PostgreSQL" , user = "bety", dbname = "bety", password = "bety") con <- db.open(dbparms) ## species category gymnosperm? input = db.query(paste('SELECT "id","scientificname","commonname","Category","GrowthForm" FROM species'),con) #input = db.query(paste("SELECT * FROM species"),con) ## vectors of categoires corresponding to data categories = growthform = speciesnames = commonname = id = vector() for (i in 1:length(input$id)) { k = grep(input$id[i], rownames(data.leaf), fixed=TRUE) ######## rownames(data.leaf) can't be redefined (should be id#site) categories[k]=input$Category[i] growthform[k]=input$GrowthForm[i] speciesnames[k]=input$scientificname[i] commonname[k]=input$commonname[i] id[k]=input$id[i] } #################### Fill missing categories ################################## growthform[1]="Single Crown"####???? growthform[2]="Bunch" ####?????? #American Beech categories[3]="Dicot" growthform[3]="Single Stem" commonname[3]="American beech" growthform[9]="Single Stem" growthform[17]="Single Stem" categories[18]="Dicot" growthform[18]="Single Crown" ###CHECK commonname[18]="Yellow Alpine Pasqueflower" growthform[21]="Single Stem" growthform[22:26]="Bunch" categories[29]="Dicot" growthform[29]="Single Stem" commonname[29]="Oak" growthform[30]="Single Stem" growthform[35]="Single Crown" ###??? woody <- vector() for (i in 1:length(growthform)) { if (growthform[i]=="Single Stem") { woody[i]=TRUE } else if (growthform[i]=="Multiple Stem") { woody[i]=TRUE } else if (growthform[i]=="Single Crown") { woody[i]=TRUE } else if (growthform[i]=="Rhizomatous") { woody[i]=TRUE } else if (growthform[i]=="Bunch") { woody[i]=FALSE } else { woody[i]=NA } } sel.vect = which(!is.na(categories)) rownames(data.leaf)=rownames(data.stem)=rownames(data.root)=speciesnames #################### renaming rownames(data.leaf) for pca labels characteristics = cbind(categories[sel.vect]=="Monocot",woody[sel.vect]==TRUE) colnames(characteristics)=c("Monocot","Woody") ## fit species charactaristics to compositional pca cluster.leaf.cost <- kmeans(t(t(data.leaf)*cost),2) pca.leaf.cost <- prcomp(data.leaf[sel.vect,],scale=cost,retx=TRUE) #ef.leaf <- envfit(pca.leaf.cost,as.factor(categories[sel.vect]),na.rm=TRUE) ef.leaf <- envfit(pca.leaf.cost,characteristics,na.rm = TRUE) biplot(pca.leaf.cost,cex=0.6,col=cluster.pca.leaf$cluster) plot (ef.leaf,cex=0.5) cluster.stem.cost <- kmeans(t(t(data.stem)*cost),2) pca.stem.cost <- prcomp(data.stem[sel.vect,],scale=cost,retx=TRUE) ef.stem <- envfit(pca.stem.cost,characteristics,na.rm=TRUE) biplot(pca.stem.cost,cex=0.8) plot(ef.stem,cex=0.5) cluster.root.cost <- kmeans(t(t(data.root)*cost),2) pca.root.cost <- prcomp(data.root[sel.vect,],scale=cost,retx=TRUE) ef.root <- envfit(pca.root.cost,characteristics,na.rm=TRUE) biplot(pca.root.cost,cex=0.8) plot(ef.root,cex=0.5) ################################## ## Split into 2 distributions ################################## ####### Monocot vs. Dicot ######## ## query species for char. j=which(categories=="Monocot") trait.data.mono = list() trait.data.dicot = list() for (i in 1:length(trait.data)) { sel.mono = which(trait.data[[i]]$specie_id %in% id[j]) trait.data.mono[[i]] = trait.data[[i]][sel.mono,] trait.data.dicot[[i]] = trait.data[[i]][-sel.mono,] # for(l in 1:length(trait.data[[i]]$specie_id)) { # if(trait.data[[i]]$specie_id[l]%in%id[j]) { # #m=as.matrix(trait.data[[i]]) # } # } #j = which(trait.data[[i]]$specie_id%in%id[j]) } names(trait.data.mono) = names(trait.data) names(trait.data.dicot) = names(trait.data) ## split trait.data #trait.data.foo1 = trait.data[k] #trat.data.foo2 = trait.data[!k] td.mono = td td.mono$pft$outdir = "/home/carya/pecan/pft/gr.mono" td.dicot = td td.dicot$pft$outdir = "/home/carya/pecan/pft/gr.dicot/" ## save #save(trait.data.foo1,file=foo1) #save(trait.data.foo2,file=foo2) save(trait.data.mono,file=file.path(td.mono$pft$outdir, 'trait.data.Rdata')) save(trait.data.dicot,file=file.path(td.dicot$pft$outdir, 'trait.data.Rdata')) ##PEcAn - get posterior priors #run.meta.analysis() run.meta.analysis(td.mono, settings$meta.analysis$iter, settings$run$dbfiles, settings$database) load(file.path(settings$pfts$pft$outdir,"trait.mcmc.Rdata")) load(file.path(settings$pfts$pft$outdir,"post.distns.Rdata")) R.mono = getdistribution(trait.mcmc,post.distns,cost,variables) run.meta.analysis(td.dicot, settings$meta.analysis$iter, settings$run$dbfiles, settings$database) load(file.path(settings$pfts$pft$outdir,"trait.mcmc.Rdata")) load(file.path(settings$pfts$pft$outdir,"post.distns.Rdata")) R.dicot = getdistribution(trait.mcmc,post.distns,cost,variables) #### Probability distributions different? ks.test(R.mono$Rl,R.dicot$Rl) ks.test(R.mono$Rs,R.dicot$Rs) ks.test(R.mono$Rr,R.dicot$Rr) cols = 1:4 dR.monond = density(R.mono$Rnd) plot(density(R.mono$Rl),xlim=range(dR.monond$x),col=cols[2]) lines(dR.monond,col=cols[1]) lines(density(R.mono$Rs),col=cols[3]) lines(density(R.mono$Rr),col=cols[4]) lines(density(R.dicot$Rnd),col=cols[1],lty=2) lines(density(R.dicot$Rl),col=cols[2],lty=2) lines(density(R.dicot$Rs),col=cols[3],lty=2) lines(density(R.dicot$Rr),col=cols[4],lty=2) legend("topright",legend=c("Null","Leaf","Stem","Root","Monocot","Dicot"),col=c(cols,1,1),lwd=2,lty=c(1,1,1,1,1,2)) ###### Woody vs Nonwoody ####### ## query species for char. j=which(woody==TRUE) trait.data.woody = list() trait.data.nonwoody = list() for (i in 1:length(trait.data)) { sel.woody = which(trait.data[[i]]$specie_id %in% id[j]) trait.data.woody[[i]] = trait.data[[i]][sel.woody,] trait.data.nonwoody[[i]] = trait.data[[i]][-sel.woody,] } names(trait.data.woody) = names(trait.data) names(trait.data.nonwoody) = names(trait.data) td.woody = td td.woody$pft$outdir = "/home/carya/pecan/pft/gr.woody/" td.nonwoody = td td.nonwoody$pft$outdir = "/home/carya/pecan/pft/gr.nonwoody/" ## save save(trait.data.woody,file=file.path(td.woody$pft$outdir, 'trait.data.Rdata')) save(trait.data.nonwoody,file=file.path(td.nonwoody$pft$outdir, 'trait.data.Rdata')) #run.meta.analysis() run.meta.analysis(td.woody, settings$meta.analysis$iter, settings$run$dbfiles, settings$database) load(file.path(settings$pfts$pft$outdir,"trait.mcmc.Rdata")) load(file.path(settings$pfts$pft$outdir,"post.distns.Rdata")) R.woody = getdistribution(trait.mcmc,post.distns,cost,variables) run.meta.analysis(td.nonwoody, settings$meta.analysis$iter, settings$run$dbfiles, settings$database) load(file.path(settings$pfts$pft$outdir,"trait.mcmc.Rdata")) load(file.path(settings$pfts$pft$outdir,"post.distns.Rdata")) R.nonwoody = getdistribution(trait.mcmc,post.distns,cost,variables) #### Test probability distributions ks.test(R.woody$Rl,R.nonwoody$Rl) ks.test(R.woody$Rs,R.nonwoody$Rs) ks.test(R.woody$Rr,R.nonwoody$Rr) cols = 1:4 dR.woodynd = density(R.woody$Rnd) plot(density(R.woody$Rl),xlim=range(dR.woodynd$x),col=cols[2]) lines(dR.woodynd,col=cols[1]) lines(density(R.woody$Rs),col=cols[3]) lines(density(R.woody$Rr),col=cols[4]) lines(density(R.nonwoody$Rnd),col=cols[1],lty=2) lines(density(R.nonwoody$Rl),col=cols[2],lty=2) lines(density(R.nonwoody$Rs),col=cols[3],lty=2) lines(density(R.nonwoody$Rr),col=cols[4],lty=2) legend("topright",legend=c("Null","Leaf","Stem","Root","Woody","Nonwoody"),col=c(cols,1,1),lwd=2,lty=c(1,1,1,1,1,2))

b5a397df0f0c92aac14390b2242a2e53e6507a74

384dd8ffffaf0b791f3934589ab008a0da22920b

/R/random.pseudoinverse.R

515d58bd001fe792f29f1a63e4bedd29fd83db88

[]

no_license

cran/ndl

51ca423e2cdad9411883eb723a79a74034cd72f0

52291ac2f05d4591d139240501c87b888093892b

refs/heads/master

2021-01-06T20:41:47.421368

2018-09-10T12:40:02

17,697,831

1

0

null

UTF-8

R

false

2,403

r

random.pseudoinverse.R

random.pseudoinverse = function (m, verbose=F, k = 0) { # computes an approximation of the SVD using the first k singular values of A stoch_svd = function(A, k, p = 200, verbose=F) { # default p=200, may need a larger value here # to use the the fast.svd function from the corpcor library # require(corpcor) # get the dimensions of the matrix n = dim(A)[1] m = dim(A)[2] # Make a random projection of A if (verbose) message("Making Random Projection") flush.console() Y <- (A %*% matrix(rnorm((k+p) * m,-1,1), ncol=k+p)) # the left part of the decomposition for A (approximately) if (verbose) message("Calculating QR decompostion of Random Projection") Q = qr.Q(qr(Y)) # taking that off gives us something small to decompose if (verbose) message("Multiply transposed QR with Orig.") B = t(Q) %*% A # decomposing B gives us singular values and right vectors for A if (verbose) message("Doing SVD of subset.") # s = fast.svd(B) s = svd(B) # Calculate U, Q time U of subset if (verbose) message("Get U from Q times U of subset.") U = Q %*% s$u # and then we can put it all together for a complete result if (verbose) message("Stoch SVD is complete.") flush.console() return (list(u=U, v=s$v, d=s$d)) } if (k<1) { #Default: K is the top 3/4 of the singular values. k = floor((dim(m)[1]) * 0.75) if (verbose) message(c("k = ",k)) } if (k>dim(m)[1]) { stop("k must be less that the size of the matrix") } if (verbose) message ("Starting reduced rank SVD approximation calc.") msvd = stoch_svd(m,k,verbose) if (length(msvd$d) == 0) #No singular values, so return zero matrix { return(array(0, dim(m)[2:1])) } else { if (verbose) message ("Done calculating pseudoinverse.") return( msvd$v %*% (1/msvd$d * t(msvd$u)) ) } }

c37fdf6b956f65f3e5f4fc811832f5de19f8b858

be0ac9cd7b4950edc946c6b5b93d92aa508cc4d6

/man/blk.diff.Rd

4e513a42baebe26ebecc10d442711737799350fe

[]

no_license

benjaminrich/PCSmisc

bbc325c0981af635039a16b704c2414fec2a41c6

b56c3d684f900b6498654405c3cff0b01d8b280b

refs/heads/master

2023-06-09T09:28:29.609251

2023-05-27T21:07:27

114,463,985

1

0

null

UTF-8

R

false

2,110

rd

blk.diff.Rd

\name{blk.diff} \alias{blk.diff} \alias{blk.intereventTime} \title{Blockwise Lagged Differences} \description{ Computes lagged difference within blocks and on selected elements. } \usage{ blk.diff(x, id, ind = NULL, lag = 1, fill = NA, diff.op = "-", ...) blk.intereventTime(time, id, ind = NULL, lag = 1, fill = NA, diff.op = difftime.default, ...) } \arguments{ \item{x,time}{A vector in \code{\link{block-format}} with respect to \code{id}.} \item{id}{A valid \code{\link{block-format}} ID.} \item{ind}{A logical vector that designates a subset of \code{x}. By default all are included.} \item{lag}{An integer specifying the lag.} \item{fill}{A value to use when no other value is appropriate.} \item{diff.op}{A function that subtracts one value from another.} \item{...}{Further arguments passed to \code{diff.op}.} } \details{ These functions operate on data sets in \code{\link{block-format}}. Essentially, the standard \code{\link{diff}} function is applied within each block, except that a function \code{diff.op} can be specified for doing the subtraction. Additionally, a subset on which to perform the operation can be selected with \code{ind}. For elements that are not selected the corresponding result is given by \code{fill}. The first \code{lag} elements of each block are the result are also assigned the value \code{fill} so that the result is in \code{\link{block-format}} with respect to \code{id}. For time values, \code{blk.intereventTime} is an alias with a more descriptive name and a different default \code{diff.op}. } \value{ A vector in \code{\link{block-format}} with respect to \code{id} containing the differenced values. } \author{Benjamin Rich <mail@benjaminrich.net>} \seealso{ \itemize{ \item \code{\link{block-format}} \item \code{\link{diff}} \item \code{\link{deltat}} } } \examples{ require(nlme) data(Phenobarb) dat <- Phenobarb[1:56,] # First 4 subjects attach(dat) cbind(dat, INTERDOSE.TIME=blk.intereventTime(time, asID(Subject), ind=!is.na(dose))) detach(dat) } \keyword{ utilities } % vim: tw=70 sw=2

390e38d52fbcaf76dcc782c124b92967d9849d87

8d11121c41ec3ea5e92c2789108c1fd1c43e7ca4

/run_analysis.R

ae05a2fa9e1579de32303931c9198542f80fc850

[]

no_license

joannaconti/cleandata

f54be5e526e0b61175f6aed1a6df6df367de6c5b

35c12cdf2a406042ed4ab9a10b25abce2a1c3680

refs/heads/master

2020-05-26T01:44:19.623105

2014-05-24T19:53:50

null

0

null

UTF-8

R

false

3,281

r

run_analysis.R

run_analysis <- function(x) { # Read all 8 files into R subjecttest = read.table("subject_test.txt") subjecttrain = read.table("subject_train.txt") xtest = read.table("X_test.txt") ytest = read.table("y_test.txt") xtrain = read.table("X_train.txt") ytrain = read.table("y_train.txt") features = read.table("features.txt") activity = read.table("activity_labels.txt") #Create vector of feature names to use for column names featuresvector = features[ , "V2"] colnames(xtest) <- featuresvector colnames(xtrain) <- featuresvector #Add description of activity to ytest & ytrain newtest = cbind(ytest, activity[ytest[ , 1], "V2"]) newtrain = cbind(ytrain, activity[ytrain[ , 1], "V2"]) #Create column names for subject and activity colnames(subjecttest) = "Subject" colnames(subjecttrain) = "Subject" colnames(newtest) = c("ActivityNum", "Activity") colnames(newtrain) = c("ActivityNum", "Activity") # Create new test and train databases with subject number included test = cbind(subjecttest, newtest, xtest) train = cbind(subjecttrain, newtrain, xtrain) #Create final database with test and train combined total = rbind(test, train) #Extract the columns that have mean or std but not meanFreq in their name to a new file extract = total[ , 1:3] x = 4 while (x <= 564) { addcolumn = FALSE columnname = colnames(total) [x] if (grepl("mean", columnname) == TRUE) {addcolumn = TRUE} if (grepl("std", columnname) == TRUE) {addcolumn = TRUE} if (grepl("meanFreq", columnname) == TRUE) {addcolumn = FALSE} if (addcolumn == TRUE) { extract = cbind(extract, total[, x]) columnnumb = ncol(extract) colnames(extract)[columnnumb] = columnname } x = x+1 } #Pretty up the column names numcolumns = ncol(extract) x = 4 while (x <= numcolumns) { oldname = colnames(extract) [x] numchar = nchar(oldname) z = 6 newname = substr(oldname, 1, 5) while (z <= numchar) { curchar = substr(oldname, z, z) if (curchar == "-") {curchar = "."} if (curchar == "(") {curchar = ""} if (curchar == ")") {curchar = ""} newname = paste(newname, curchar, sep="") z = z+1 } colnames(extract)[x] = newname x = x+1 } #Extract just the columns that have mean in their name to extract2 file extract2 = extract[ , 1:3] x = 4 while (x <= 69) { addcolumn = FALSE columnname = colnames(extract) [x] if (grepl("mean", columnname) == TRUE) {addcolumn = TRUE} if (addcolumn == TRUE) { extract2 = cbind(extract2, extract[, x]) columnnumb = ncol(extract2) colnames(extract2)[columnnumb] = columnname } x = x+1 } #Melt extract2 into a long file & dcast back the means meltdf = melt(extract2, id.vars = c("Subject", "ActivityNum", "Activity")) dcastdb=dcast(meltdf, Subject + Activity ~ variable, mean) return(dcastdb) }

738ae02fa5c530d180ed699e77c985cc6ab8a1a5

05ebb4d386cb2604bb7642bd79d09fa3ca76dc72

/man/tbk_data.Rd

307ee79c3e9492e62be0f223f5a7ca7d50e75ba9

[]

no_license

trichelab/tbmater

a322d5b3c558c4b45474e3ed1e394754543cc5d5

dafbf46ca7a021849a0e5b86c1669fe7d2ad3447

refs/heads/master

2023-01-08T04:53:43.023578

2020-11-12T02:59:17

312,152,018

0

null

UTF-8

R

false

true

1,228

rd

tbk_data.Rd

% Generated by roxygen2: do not edit by hand % Please edit documentation in R/tbk_data.R \name{tbk_data} \alias{tbk_data} \title{Get data from tbk} \usage{ tbk_data( tbk_fnames, idx_fname = NULL, probes = NULL, show.unaddressed = FALSE, chrm = NULL, beg = NULL, end = NULL, as.matrix = FALSE, simplify = FALSE, name.use.base = TRUE, max_addr = 3000, max_source = 10^6, max_pval = 0.05, min_coverage = 5, all_units = FALSE ) } \arguments{ \item{idx_fname}{index file. If not given, use the idx.gz in first path} \item{probes}{probe names} \item{simplify}{reduce matrix to vector if only one sample is queried} \item{name.use.base}{use basename for sample name} \item{max_addr}{random addressing if under max_addr} \item{max_pval}{maximum sig2 for float.float} \item{min_coverage}{minimum sig2 for float.int} \item{all_units}{retrieve all units for float.float and float.int} \item{fnames}{tbk_fnames} \item{min_source}{random addressing if source size is under min_source} } \value{ \preformatted{ a numeric matrix } } \description{ Assumptions: \enumerate{ \item All the tbks have the same index. \item If idx_fname not given, idx.gz is in the same folder as the first sample. } }

604d09c5d2bca6fa0b3b409611d8f6409130c6aa

5c255d5bba86ddd16c76a43e4f51c3099fb22ac6

/tests/testthat/test_ST_Ranking.R

3b7c7960e51023e269091c9b5c703d549b1d35c5

[]

no_license

tristanbains/rFootballAnalysis

92d7b4ef52c117d62a58f1ad901c70e9af1c3529

b898ba686baffd5da63c5312898e018241b9c127

refs/heads/master

2021-01-20T12:33:49.422731

2017-05-19T11:08:34

90,376,720

0

null

UTF-8

R

false

309

r

test_ST_Ranking.R

context("ST_Ranking") data = suppressMessages(suppressWarnings( DL_Matches("ESP1",season1 = "2014-2015",season2 = "2014-2015"))) result = suppressMessages(suppressWarnings(ST_Ranking(data = data))) test_that("Right ordering based on aggregate matches",{ expect_equal(result$GD[16:18],c(-25,-35,-21)) })

18619e7d3ab33e137b93b281616bdc2c1799c8ce

fc18e2ba06ff2e409f8a30c76d8841243c4b393a

/R/mergeROC.R

6b58eab6268572994b1f1ce891d2d9a55d1a0694

[]

no_license

zhixingfeng/seqPatch

59c0f36eb02f8ca63f467f1b4122a4736ca36b1c

5ef9d6084e9af2d01cc292334eed4f71ec4e6e7b

refs/heads/master

2020-12-24T15:05:33.081698

2014-07-31T06:54:56

6,462,374

8

0

null

UTF-8

R

false

4,101

r

mergeROC.R

mergeROC <- function(ROC.B, TPR.knots=seq(0.05,1,0.05)) { # spline interpolate model <- list() FDR.B <- list() for (i in 1:length(ROC.B)){ model[[i]] <- list() FDR.B[[i]] <- list() # fit spline for forward strand for (k in 1:length(ROC.B[[i]])){ TPR <- ROC.B[[i]][[k]]$TPR FDR <- ROC.B[[i]][[k]]$FDR chr <- names(ROC.B[[i]])[k] model[[i]][[chr]] <- approx(TPR,FDR, xout=TPR.knots, rule=2) FDR.B[[i]][[chr]] <- model[[i]][[chr]]$y } } FDR.mean <-list() FDR.sd <-list() for (k in 1:length(FDR.B[[i]]) ){ chr <- names(FDR.B[[i]])[k] FDR.mean[[chr]] <- 0 FDR.sd[[chr]] <- 0 for (i in 1:length(FDR.B)){ FDR.mean[[chr]] <- FDR.mean[[chr]] + FDR.B[[i]][[k]] FDR.sd[[chr]] <- FDR.sd[[chr]] + FDR.B[[i]][[k]]^2 } FDR.mean[[chr]] <- FDR.mean[[chr]] / length(FDR.B) FDR.sd[[chr]] <- sqrt(FDR.sd[[chr]]/length(FDR.B) - FDR.mean[[chr]]^2 + 1e-10)/sqrt(length(FDR.B)) FDR.mean[[chr]][FDR.mean[[chr]]<0] <- 0 FDR.mean[[chr]][FDR.mean[[chr]]>1] <- 1 } list(FDR.B=FDR.B, FDR.mean=FDR.mean, FDR.sd= FDR.sd) } mergeROC.bak <- function(ROC.B, TPR.knots=seq(0.05,1,0.05)) { # spline interpolate model <- list() FDR.B <- list() for (i in 1:length(ROC.B)){ model[[i]] <- list() FDR.B[[i]] <- list() # fit spline for forward strand for (k in 1:length(ROC.B[[i]]$pos)){ TPR <- ROC.B[[i]]$pos[[k]]$TPR FDR <- ROC.B[[i]]$pos[[k]]$FDR model[[i]]$pos[[k]] <- approx(TPR,FDR, xout=TPR.knots, rule=2) FDR.B[[i]]$pos[[k]] <- model[[i]]$pos[[k]]$y } names(model[[i]]$pos) <- names(ROC.B[[i]]$pos) names(FDR.B[[i]]$pos) <- names(ROC.B[[i]]$pos) # fit spline for backward strand for (k in 1:length(ROC.B[[i]]$neg)){ TPR <- ROC.B[[i]]$neg[[k]]$TPR FDR <- ROC.B[[i]]$neg[[k]]$FDR model[[i]]$neg[[k]] <- approx(TPR, FDR, xout=TPR.knots, rule=2) FDR.B[[i]]$neg[[k]] <- model[[i]]$neg[[k]]$y } names(model[[i]]$neg) <- names(ROC.B[[i]]$neg) names(FDR.B[[i]]$neg) <- names(ROC.B[[i]]$neg) } FDR.mean.pos <-list() FDR.sd.pos <-list() FDR.mean.neg <-list() FDR.sd.neg <-list() for (k in 1:length(ROC.B[[i]]$pos) ){ FDR.mean.pos[[k]] <- 0 FDR.sd.pos[[k]] <- 0 for (i in 1:length(ROC.B)){ FDR.mean.pos[[k]] <- FDR.mean.pos[[k]] + FDR.B[[i]]$pos[[k]] FDR.sd.pos[[k]] <- FDR.sd.pos[[k]] + FDR.B[[i]]$pos[[k]]^2 } FDR.mean.pos[[k]] <- FDR.mean.pos[[k]] / length(ROC.B) FDR.sd.pos[[k]] <- sqrt(FDR.sd.pos[[k]]/length(ROC.B) - FDR.mean.pos[[k]]^2 + 1e-10)/sqrt(length(ROC.B)) FDR.mean.pos[[k]][FDR.mean.pos[[k]]<0] <- 0 FDR.mean.pos[[k]][FDR.mean.pos[[k]]>1] <- 1 } for (k in 1:length(ROC.B[[i]]$neg) ){ FDR.mean.neg[[k]] <- 0 FDR.sd.neg[[k]] <- 0 for (i in 1:length(ROC.B)){ FDR.mean.neg[[k]] <- FDR.mean.neg[[k]] + FDR.B[[i]]$neg[[k]] FDR.sd.neg[[k]] <- FDR.sd.neg[[k]] + FDR.B[[i]]$neg[[k]]^2 } FDR.mean.neg[[k]] <- FDR.mean.neg[[k]] / length(ROC.B) FDR.sd.neg[[k]] <- sqrt(FDR.sd.neg[[k]]/length(ROC.B) - FDR.mean.neg[[k]]^2 + 1e-10)/sqrt(length(ROC.B)) FDR.mean.neg[[k]][FDR.mean.neg[[k]]<0] <- 0 FDR.mean.neg[[k]][FDR.mean.neg[[k]]>1] <- 1 } names(FDR.mean.pos) <- names(ROC.B[[1]]$pos) names(FDR.sd.pos) <- names(ROC.B[[1]]$pos) names(FDR.mean.neg) <- names(ROC.B[[1]]$neg) names(FDR.mean.neg) <- names(ROC.B[[1]]$neg) result <- list() result$FDR.mean.pos <- FDR.mean.pos result$FDR.sd.pos <- FDR.sd.pos result$FDR.mean.neg <- FDR.mean.neg result$FDR.sd.neg <- FDR.sd.neg result$FDR.B <- FDR.B result }

b734bd34f0454dc6e6a4c380217b230b62197887

6310ea884f52bfddeebc31ab9fb66bfce105ede0

/tools/r-packages/nibrs/R/IncidentFunctions.R

46ef07f12dbcf5269a5f58a8599550198e4b5ced

[ "Apache-2.0" ]

permissive

mark43/nibrs

24c1d59c005f4134a4f62c42c00b7b7306003d89

65dcd57f874f8211a9fb507fbfd082ec9177b669

refs/heads/master

2023-04-28T02:10:24.762013

2021-12-17T16:15:45

98,336,429

0

Apache-2.0

2023-04-15T04:45:34

2017-07-25T18:13:14

Java

UTF-8

R

false

5,902

r

IncidentFunctions.R

# Copyright 2016 SEARCH-The National Consortium for Justice Information and Statistics # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. # #' @importFrom readr read_fwf # loadIncidentFile <- function(file, maxRecords = -1) { # columnSpecsFile <- system.file("raw", "IncidentFileFormat.txt", package=getPackageName()) # columnSpecs <- getColumnSpecs(columnSpecsFile) # read_fwf(file=file, col_positions = fwf_positions(start = columnSpecs$start, end = columnSpecs$end, col_names = columnSpecs$name), # col_types=paste(columnSpecs$type, collapse=""), n_max = maxRecords) %>% ungroup() %>% mutate(AdministrativeSegmentID=row_number()) # } #' @importFrom readr read_fwf loadIncidentFile <- function(file, versionYear, maxRecords = -1) { columnSpecsFile <- system.file("raw", paste0('IncidentFileFormat-', versionYear, '.txt'), package=getPackageName()) columnSpecs <- getColumnSpecs(columnSpecsFile) read_fwf(file=file, col_positions = fwf_positions(start = columnSpecs$start, end = columnSpecs$end, col_names = columnSpecs$name), col_types=paste(columnSpecs$type, collapse=""), n_max = maxRecords, progress=FALSE) %>% ungroup() %>% mutate(AdministrativeSegmentID=row_number()) } #' @import dplyr addAdministrativeSegmentID <- function(rawIncidentsDataFrame) { rawIncidentsDataFrame %>% ungroup() %>% mutate(AdministrativeSegmentID=row_number()) } #' @importFrom DBI dbClearResult dbSendQuery truncateIncidents <- function(conn) { dbClearResult(dbSendQuery(conn, "truncate AdministrativeSegment")) } #' @import dplyr #' @importFrom DBI dbWriteTable #' @importFrom lubridate month year ymd writeIncidents <- function(conn, rawIncidentDataFrame, segmentActionTypeTypeID, agencyDataFrame) { currentMonth <- formatC(month(Sys.Date()), width=2, flag="0") currentYear <- year(Sys.Date()) %>% as.integer() AdministrativeSegment <- rawIncidentDataFrame %>% processingMessage('Incident') %>% select(AdministrativeSegmentID, ORI, IncidentNumber=INCNUM, INCDATE, IncidentHour=V1007, ClearedExceptionallyTypeID=V1013, ReportDateIndicator=V1006) %>% mutate(INCDATE=ifelse(INCDATE==-5, NA, INCDATE)) %>% mutate(IncidentDate=ymd(INCDATE), IncidentDateID=createKeyFromDate(IncidentDate), MonthOfTape=currentMonth, YearOfTape=currentYear, CityIndicator=NA, SegmentActionTypeTypeID=segmentActionTypeTypeID, ClearedExceptionallyTypeID=ifelse(ClearedExceptionallyTypeID==-6, 6L, ClearedExceptionallyTypeID), CargoTheftIndicatorTypeID=99998L, IncidentHour=ifelse(IncidentHour < 0, NA_integer_, IncidentHour)) %>% select(-INCDATE) %>% left_join(agencyDataFrame %>% select(AgencyID, ORI=AgencyORI), by='ORI') writeLines(paste0("Writing ", nrow(AdministrativeSegment), " administrative segments to database")) dbWriteTable(conn=conn, name="AdministrativeSegment", value=AdministrativeSegment, append=TRUE, row.names = FALSE) attr(AdministrativeSegment, 'type') <- 'FT' AdministrativeSegment } #' @import dplyr #' @import tidyr #' @import tibble #' @importFrom DBI dbWriteTable writeRawAdministrativeSegmentTables <- function(conn, inputDfList, tableList) { dfName <- load(inputDfList[2]) adminSegmentDf <- get(dfName) %>% mutate_if(is.factor, as.character) %>% inner_join(tableList$Agency %>% select(AgencyORI), by=c('V1003'='AgencyORI')) rm(list=dfName) currentMonth <- formatC(month(Sys.Date()), width=2, flag="0") currentYear <- year(Sys.Date()) %>% as.integer() AdministrativeSegment <- adminSegmentDf %>% select(ORI=V1003, IncidentNumber=V1004, INCDATE=V1005, IncidentHour=V1007, V1013, V1016, ReportDateIndicator=V1006) %>% mutate(IncidentHour=gsub(x=IncidentHour, pattern='\$([0-9]+)\$.+', replacement='\\1')) %>% mutate(INCDATE=ifelse(trimws(INCDATE)=='' | is.na(INCDATE), NA, INCDATE)) %>% mutate(IncidentDate=ymd(INCDATE), IncidentDateID=createKeyFromDate(IncidentDate), MonthOfTape=currentMonth, YearOfTape=currentYear, CityIndicator=NA_character_, SegmentActionTypeTypeID=99998L, V1016=ifelse(trimws(V1016)=='' | is.na(V1016), 99998L, V1016), IncidentHour=ifelse(is.na(IncidentHour), NA_integer_, as.integer(IncidentHour))) %>% select(-INCDATE) %>% mutate(AdministrativeSegmentID=row_number()) %>% left_join(tableList$Agency %>% select(AgencyID, ORI=AgencyORI), by='ORI') %>% left_join(tableList$ClearedExceptionallyType %>% select(ClearedExceptionallyTypeID, ClearedExceptionallyCode), by=c('V1013'='ClearedExceptionallyCode')) %>% left_join(tableList$CargoTheftIndicatorType %>% select(CargoTheftIndicatorTypeID, CargoTheftIndicatorCode), by=c('V1016'='CargoTheftIndicatorCode')) %>% mutate(ClearedExceptionallyTypeID=ifelse(is.na(ClearedExceptionallyTypeID), 99998L, ClearedExceptionallyTypeID)) %>% mutate(CargoTheftIndicatorTypeID=ifelse(is.na(CargoTheftIndicatorTypeID), 99998L, CargoTheftIndicatorTypeID)) %>% select(-V1013, -V1016) %>% as_tibble() rm(adminSegmentDf) writeLines(paste0("Writing ", nrow(AdministrativeSegment), " administrative segments to database")) dbWriteTable(conn=conn, name="AdministrativeSegment", value=AdministrativeSegment, append=TRUE, row.names = FALSE) attr(AdministrativeSegment, 'type') <- 'FT' tableList$AdministrativeSegment <- AdministrativeSegment tableList }

56aaf8cdedf06cd987cf302b89af7aec7c34fd9d

5b73b50251eef41c0e41dce6cba8b50f0a3765a3

/preprocessing/weather data/weather_merge2020.R

b73d556311c28b47f79b765b1c4e8cb335292738

[]

no_license

noranm/landslide-prediction

cc250cabb127bc5d93a198535e98992215e7507d

9c38b4b8305e74fe4fb116c1fcd901f1993a9aee

refs/heads/main

2023-06-17T20:55:56.925294

2021-07-20T12:06:25

null

0

null

UTF-8

R

false

5,751

r

weather_merge2020.R

########## KN library(dplyr) path = "./data2/경상남도/" KN_SGGs <- list.files(path) final1 <- final2 <- final3 <- NULL for (sgg in KN_SGGs) { print(sgg) sgg_dir = paste0(path, sgg, "/") DATEs = list.files(sgg_dir) for (dt in DATEs){ dt_dir = paste0(sgg_dir, dt, "/") FILEs = list.files(dt_dir) for (f in FILEs) { set <- strsplit(f, "_") umd = set[[1]][1]; value = set[[1]][2]; first_date = set[[1]][3]; end_date = substr(set[[1]][4],1,6) CSVfile = read.csv(paste0(dt_dir, f)) CSVfile <- na.omit(CSVfile) colnames(CSVfile) <- c("day", "hour", "val") CSVfile$day <- as.numeric(CSVfile$day) CSVfile$hour <- as.numeric(CSVfile$hour) CSVfile$val <- as.numeric(CSVfile$val) CSVfile$month <- c(rep(6,30*24), rep(7,31*24), rep(8,31*24), rep(9,30*24)) CSVfile$val[CSVfile$val == -1] <- 0 if (value == "강수") { tmp1 <- CSVfile %>% group_by(month, day) %>% summarise(var1 = max(val, na.rm=TRUE), var2 = sum(val, na.rm=TRUE)) %>% data.frame() colnames(tmp1) <- c("month", "day", "최대시우량", "하루총강수량") tmp1$SIDO <- "경상남도"; tmp1$SGG_NM <- sgg; tmp1$UMD <- umd; tmp1$날짜 <- as.character(as.numeric(202000 + tmp1$month)*100 + tmp1$day); final1 <- rbind(final1, tmp1) } else if (value == "풍속") { tmp2 <- CSVfile %>% group_by(month, day) %>% summarise(var = max(val, na.rm=TRUE)) %>% data.frame() colnames(tmp2) <- c("month", "day", "최대풍속") tmp2$SIDO <- "경상남도"; tmp2$SGG_NM <- sgg; tmp2$UMD <- umd; tmp2$날짜 <- as.character(as.numeric(202000 + tmp2$month)*100 + tmp2$day); final2 <- rbind(final2, tmp2) } else { # 없음(0), 비(1), 비/눈(2), 눈(3) tmp3 <- CSVfile %>% group_by(month, day) %>% summarise("var1" = sum(val != 0, na.rm=TRUE)/24) %>% data.frame() colnames(tmp3) <- c("month", "day", "강수비율") tmp3$SIDO <- "경상남도"; tmp3$SGG_NM <- sgg; tmp3$UMD <- umd; tmp3$날짜 <- as.character(as.numeric(202000 + tmp3$month)*100 + tmp3$day); final3 <- rbind(final3, tmp3) } } } } KN_final <- merge( final1[,c("SIDO", "SGG_NM", "UMD", "날짜", "최대시우량", "하루총강수량")], final2[,c("SIDO", "SGG_NM", "UMD", "날짜", "최대풍속")], by= c("SIDO", "SGG_NM", "UMD", "날짜"), all=TRUE) KN_final <- merge( KN_final, final3[,c("SIDO", "SGG_NM", "UMD", "날짜", "강수비율")], by= c("SIDO", "SGG_NM", "UMD", "날짜"), all=TRUE) colSums(is.na(KN_final)) KN_final[KN_final$UMD == "장목면",] ########## KB path = "./data2/경상북도/" KB_SGGs <- list.files(path) final1 <- final2 <- final3 <- NULL for (sgg in KB_SGGs) { print(sgg) sgg_dir = paste0(path, sgg, "/") DATEs = list.files(sgg_dir) for (dt in DATEs){ dt_dir = paste0(sgg_dir, dt, "/") FILEs = list.files(dt_dir) for (f in FILEs) { set <- strsplit(f, "_") umd = set[[1]][1]; value = set[[1]][2]; first_date = set[[1]][3]; end_date = substr(set[[1]][4],1,6) CSVfile = read.csv(paste0(dt_dir, f)) CSVfile <- na.omit(CSVfile) colnames(CSVfile) <- c("day", "hour", "val") CSVfile$day <- as.numeric(CSVfile$day) CSVfile$hour <- as.numeric(CSVfile$hour) CSVfile$month <- c(rep(6,30*24), rep(7,31*24), rep(8,31*24), rep(9,30*24), rep(10,31*24)) CSVfile$val[CSVfile$val == -1] <- 0 if (value == "강수") { tmp1 <- CSVfile %>% group_by(month, day) %>% summarise(var1 = max(val, na.rm=TRUE), var2 = sum(val, na.rm=TRUE)) %>% data.frame() colnames(tmp1) <- c("month", "day", "최대시우량", "하루총강수량") tmp1$SIDO <- "경상북도"; tmp1$SGG_NM <- sgg; tmp1$UMD <- umd; tmp1$날짜 <- as.character(as.numeric(202000 + tmp1$month)*100 + tmp1$day); final1 <- rbind(final1, tmp1) } else if (value == "풍속") { tmp2 <- CSVfile %>% group_by(month, day) %>% summarise(var = max(val, na.rm=TRUE)) %>% data.frame() colnames(tmp2) <- c("month", "day", "최대풍속") tmp2$SIDO <- "경상북도"; tmp2$SGG_NM <- sgg; tmp2$UMD <- umd; tmp2$날짜 <- as.character(as.numeric(202000 + tmp2$month)*100 + tmp2$day); final2 <- rbind(final2, tmp2) } else { # 없음(0), 비(1), 비/눈(2), 눈(3) tmp3 <- CSVfile %>% group_by(month, day) %>% summarise("var1" = sum(val != 0, na.rm=TRUE)/24) %>% data.frame() colnames(tmp3) <- c("month", "day", "강수비율") tmp3$SIDO <- "경상북도"; tmp3$SGG_NM <- sgg; tmp3$UMD <- umd; tmp3$날짜 <- as.character(as.numeric(202000 + tmp3$month)*100 + tmp3$day); final3 <- rbind(final3, tmp3) } } } } KB_final <- merge( final1[,c("SIDO", "SGG_NM", "UMD", "날짜", "최대시우량", "하루총강수량")], final2[,c("SIDO", "SGG_NM", "UMD", "날짜", "최대풍속")], by= c("SIDO", "SGG_NM", "UMD", "날짜"), all=TRUE) KB_final <- merge( KB_final, final3[,c("SIDO", "SGG_NM", "UMD", "날짜", "강수비율")], by= c("SIDO", "SGG_NM", "UMD", "날짜"), all=TRUE) colSums(is.na(KB_final)) final <- rbind(KN_final, KB_final) colSums(is.na(final)) final <- final[!is.na(final$최대시우량),] file_format <- file("./WeatherInfo2020.csv", encoding='euc-kr') write.csv(final, file_format, row.names=FALSE)

6cb769384acb6bd31533833b2951c059b941e5b0

a933950bb09c97480b7031abe7858d65e0bf2d87

/map_ze_breakfast.R

bfc1a9db45854ec94a4e590add5fa9211062f0b1

[]

no_license

rachwhatsit/breakfast-report-card-map

9816559a2c0823ec4ecc7abbffc6b6b66cab664a

9364d17a97fb6325f456d5593549276a221d0f2a

refs/heads/master

2021-01-01T03:55:22.160928

2017-12-24T14:59:25

59,143,117

0

null

UTF-8

R

false

1,491

r

map_ze_breakfast.R

library(choroplethr) library(choroplethrMaps) library(ggplot2);library(classInt) setwd("/Users/rachel_wilkerson/Box Sync/RLW THI Projects/breakfast_heat_map/") load("map_ze_breakfast.RData") df=read.csv("heatmap.csv",strip.white=T,stringsAsFactors = F) df$perc=as.numeric(substr(df$X..FRP.Breakfast.Participation,1,nchar(df$X..FRP.Breakfast.Participation)-1)) df=df[-c(253,252),] df[is.na(df)]=0 classIntervals(df$perc, 6,"jenks") data(county.regions) county.regions.tx=county.regions[which(county.regions$state.name=="texas"),] df$County=tolower(df$County) df=merge(df,county.regions.tx,by.x="County",by.y="county.name",all=T) df$perc[which(is.na(df$perc)==T)]<-0 df$value=cut(df$perc,breaks=c(0,10,20,30,40,50,60,70,80,90),include.lowest = T) #df$region=county.regions.tx$region[match(tolower(df$County),county.regions.tx$county.name)] #df$check=county.regions.tx$county.name[match(tolower(df$County),county.regions.tx$county.name)] #length(unique(df$region)) #df=df[-c(252,253),] df$region.y<-NULL colnames(df)[6]<-"region" county_choropleth(df, title = "Participation", legend = "Participation", num_colors = 1, state_zoom = "texas") df$value[which(is.na(df$value)==T)]<-0 choro = CountyChoropleth$new(df) choro$ggplot_scale = scale_fill_brewer(name="Participation",palette=7, drop=FALSE) choro$set_zoom("texas") choro$render() ggsave("map_me_some_breakfast_orangeCUTS.pdf",height=7,width=7,units="in")

bda0fb4f3ea73b70098b0ac2fcb689d3137bf57d

b07bf8cef773d2611d329c93bcb645ccc2b87cba

/Code/Chapter 6.2 - Ridge and Lasso Regression.R

516f81741eeb90483b4ddba1244d4302b111a905

[]

no_license

ebardelli/Introduction-to-Statistical-Learning-Labs

9333a4d38042bae2ef951a3ed43a04a9360084fc

be4dfd9d41d7eb46630a12eea7eaebb2d53b9fcc

refs/heads/master

2022-12-04T11:19:30.314256

2020-08-01T19:22:37

278,771,282

0

null

2020-07-11T02:17:13

2020-07-11T02:17:12

null

UTF-8

R

false

4,981

r

Chapter 6.2 - Ridge and Lasso Regression.R

#This script corresponds with Chapter 6 of Inroduction to Statistical Learning #Author : William Morgan #6.6 Lab 2: Ridge Regression and the Lasso library(glmnet) #glmnet() function for ridge/lasso library(ISLR) #Hitters data ##### SECTION 1: Ridge Regression ##### #Prepare the Hitters data as done in the previous lab hitters <- Hitters hitters <- na.omit(hitters) attach(hitters) colnames(hitters) <- tolower(colnames(hitters)) #Before continuing, take a look at the glmnet() vignette to get an idea of what we need beforehand #x - input matrix containing rows of observations #y - response variable matrix #lambda - a user created decreasing sequence of lambda values #Lambda will be a vector of length 100, ranging from 10^10 to 10^-2 x <- model.matrix(salary~., hitters)[,-1] y <- salary grid <-10^seq(10,-2,length=100) ridge.mod <- glmnet(x,y, alpha=0, lambda=grid) ### Pause for Analysis ### #Because we supplied a 100 lambdas, we've got a coefficient matrix that is 20x100 #We expect that the coefficients of models with larger lambdas will be much smaller than those with smaller lambdas #Remember, the sequence is decreasing so the further along the sequence the larger the coefficients #Find the 50th and 100th lambdas, along with the coefficients associated with that model and their l_2 norm (excluding the intercept) ridge.mod$lambda[50] coef(ridge.mod)[,50] sqrt(sum(coef(ridge.mod)[-1,50])^2) ridge.mod$lambda[100] coef(ridge.mod)[,100] sqrt(sum(coef(ridge.mod)[-1,100])^2) #Just for fun, let's practice writing a function to extract this information for any lambda shrinkage_coef <- function(glmnet_mod, ld) { coef_names <- names(coef(glmnet_mod)[,50]) return(list(print(paste("The Lambda value is:",glmnet_mod$lambda[ld])), print(paste("The coefficient for",coef_names,"is:", coef(glmnet_mod)[,ld])), print(paste("The l_2 norm of these coefficients is:", sqrt(sum(coef(glmnet_mod)[-1,ld])^2))) )) } #Now that we've got a practice run down, let's split our sample to do some testing set.seed(1) train <- sample(1:nrow(x), nrow(x)/2) test <- (-train) y.test <- y[test] ### Important Note ### #There are two common ways to randomly split a data set: #You can produce a random logical vector (TRUE/FALSE) and select observations corresponding to TRUE for the training data #Alternatively, randomlychoose a subset of numbers between 1 and n, which can then be used as the indices for the training data #We do (and have been doing) the former in previous labs; this lab makes use of the latter #Fit a ridge regression on the training set and evaluate its MSE on the test, using lambda = 4 ridge_mod <- glmnet(x[train,], y[train], alpha=0, lambda=grid, thresh=1e-12) ridge_pred <- predict(ridge_mod, s=4, newx=x[test,]) #s option sets the lambda value, newx specifies new observations used to make predicitons mean((ridge_pred - y.test)^2) #MSE #Compare to the test MSE when lambda is extremely large (coefficients are approximately 0) ridge_pred <- predict(ridge_mod, s=1e10, newx=x[test,]) mean((ridge.pred - y.test)^2) #Finally, let's check if the ridge regression gives us better results than the least sqaures option (lambda = 0) ridge_pred <- predict(ridge_mod, s=0, newx=x[test,], exact = T) #the exact option allow us to specify that lambda is exactly 0, instead of searching for the smalles value of lambda in "grid" mean((ridge_pred - y.test)^2) #Instead of arbitrarily choosing a lambda value a priori, it is better to use cross-validation to find the best lambda #This can be done using cv.glmnet(), which conducts 10-fold CV and can be increased to n-folds with the option nfolds set.seed(1) cv_out <- cv.glmnet(x[train,], y[train], alpha=0) plot(cv_out) best_lam <- cv_out$lambda.min #What is the test MSE associated with this lambda? ridge_pred <- predict(ridge_mod, s=best_lam, newx=x[test,]) mean((ridge_pred - y.test)^2) #FINALLY we can run ridge regression on the entire data set, now that we have found the best value for our tuning parameter out <- glmnet(x,y,alpha=0) predict(out, type = "coefficients", s=best_lam)[1:20,] ##### End of 6.2.1 ##### ##### SECTION 2: The Lasso ##### #Fit the lasso model and observe how some of the coefficients are exactly 0 lasso_mod <- glmnet(x[train,], y[train], alpha = 1, lambda = grid) plot(lasso_mod) #Perform CV and compute test errors set.seed(1) cv_out <- cv.glmnet(x[train,], y[train],alpha=1) plot(cv_out) best_lam <- cv_out$lambda.min lasso_pred <- predict(lasso_mod, s=best_lam, newx=x[test,]) mean((lasso_pred - y.test)^2) #Fit the lasso over the entire data set out <- glmnet(x,y,alpha=1, lambda = grid) lasso_coef <- predict(out, type = "coefficients", s=best_lam)[1:20,] lasso_coef lasso_coef[lasso_coef!=0] ### Pause for analysis ### #The test MSE for the lasso is very similar to the ridge regression, but it does have a minor advantage: #12 of the 19 coefficients in the lasso model are exactly 0 ##### End of 6.2.2 #####

14ffe6da5eae92789e116f0af726ee16b4efecef

e56da52eb0eaccad038b8027c0a753d9eb2ff19e

/tests/testthat/test-split_analysis.R

aba5bd299c03dc5342db47cea8a55f62e36db1cb

[]

no_license

ms609/TreeTools

fb1b656968aba57ab975ba1b88a3ddf465155235

3a2dfdef2e01d98bf1b58c8ee057350238a02b06

refs/heads/master

2023-08-31T10:02:01.031912

2023-08-18T12:21:10

215,972,277

16

5

null

2023-08-16T16:04:19

2019-10-18T08:02:40

R

UTF-8

R

false

529

r

test-split_analysis.R

test_that("TipsInSplits() family", { test <- TipsInSplits(BalancedTree(letters[1:5])) expect_identical(test, c("7" = 3L, "8" = 2L, "9" = 2L)[names(test)]) expect_equal(TipsInSplits(PectinateTree(7), smallest = TRUE), c("10" = 2, "11" = 3, "12" = 3, "13" = 2)) expect_identical(15:2, TipsInSplits(PectinateTree(17), keep.names = FALSE)) test <- SplitImbalance(BalancedTree(7)) expectation <- c("9" = 1L, "10" = 3L, "11" = 3L, "12" = 1L, "13" = 3L) expect_identical(test, expectation[names(test)]) })

da9337f12be36190dfbfd2cb6b3e706ea5e9235c

672e6732d5d81ecb3b24bc47011439ef59a728b5

/Rstudio/LAB3_data_mining/lab3_exercises.R

4170bcb457148c260149831fc4698b2d2026aad7

[]

no_license

sslowik/Mgr_Inf

4b4349a5e09a2354eefecf464fe53333d8365758

ddc65c67b2b2b6990f4591adeaa33b03bc34faaf

refs/heads/master

2022-12-01T14:33:39.213225

2021-01-19T20:07:57

156,598,963

2

null

2022-11-24T09:31:00

2018-11-07T19:40:16

HTML

UTF-8

R

false

3,466

r

lab3_exercises.R

dirty.iris <- read.csv("dirty_iris.csv", header=TRUE, sep=",") install.packages("editrules") library(editrules) a = subset(dirty.iris, is.finite(Sepal.Length) & is.finite(Sepal.Width) & is.finite(Petal.Length) & is.finite(Petal.Width)) nrow(a)[1] 95 E <- editset(c("Sepal.Length <= 30", "Species %in% c('setosa','versicolor','virginica')")) E <- editset(c("Sepal.Length <= 30", "Species %in% c('setosa','versicolor','virginica')", "Sepal.Length>0", "Sepal.Width>0", "Petal.Length > 0", "Petal.Width>0", "Petal.Length >= 2* Petal.Width", "Sepal.Length > Petal.Length")) E ve <- violatedEdits(E, dirty.iris) ; ve summary(ve) plot(ve) #zadanie 2. install.packages("deducorrect") library(deducorrect) #rules.txt: # if (!is.na(Petal.Width) & Petal.Width != 'Inf' & Petal.Width <= 0){ # Petal.Width <- NA # } #if (!is.na(Petal.Length) & Petal.Length <= 0){ # Petal.Length <- NA #} #if (!is.na(Sepal.Width) & Sepal.Width <= 0){ # Sepal.Width <- NA #} #if (!is.na(Sepal.Length) & (Sepal.Length <= 0 | Sepal.Length > 30)){ # Sepal.Length <- NA #} #if (!is.na(Petal.Width) & !is.na(Petal.Length) & 2*Petal.Width >= Petal.Length){ # Petal.Length <- NA #} #if (!is.na(Petal.Length) & !is.na(Sepal.Length) & Petal.Length >= Sepal.Length) { # Sepal.Length <- NA #} R <- correctionRules("rules.txt") corrected.dirty.iris <- correctWithRules(R, dirty.iris) iris_corrected <- corrected.dirty.iris$corrected iris_corrected # zadanie 3. #a) install.packages("Hmisc") library(Hmisc) cbind.data.frame(Sepal.Length=impute(corrected$Sepal.Length, mean), Sepal.Width=impute(corrected$Sepal.Width, mean), Petal.Length=impute(corrected$Petal.Length, mean), Petal.Width=impute(corrected$Petal.Width, mean), corrected$Species) #b) install.packages("VIM") library(VIM) clean.iris.knn <- kNN(corrected) clean.iris.knn.2 <- kNN(corrected)[1:5] #zadanie.4. # a) iris # b) log iris.log <- cbind.data.frame(Sepal.Length=log(iris$Sepal.Length), Sepal.Width=log(iris$Sepal.Width), Petal.Length=log(iris$Petal.Length), Petal.Width=log(iris$Petal.Width), Speciesiris$Species) #c) iris.log.scale <- cbind.data.frame(Sepal.Length=scale(iris.log$Sepal.Length), Sepal.Width=scale(iris.log$Sepal.Width), Petal.Length=scale(iris.log$Petal.Length), Petal.Width=scale(iris.log$Petal.Width), Species=iris.log$Species) sd(iris.log.scale$Petal.Length) = 1 mean(iris.log.scale$Petal.Length) ?0 #zadanie 5. #a) iris.log.scale <- subset(iris.log.scale, select = -c(Species)) # or iris.log.scale <- iris.log.scale[,-5] #b) iris.pca <- prcomp(iris.log.scale) #c) iris.pca # Sdev1 = 1.7124583 # Sdev2 = 0.9523797 # Sdev3 = 0.3647029 # Sdev4 = 0.1656840 #d) iris.predict <- predict(iris.pca) iris.predict <- subset(iris.predict, select = -c(PC3)) iris.predict <- subset(iris.predict, select = -c(PC4)) iris.predict <- cbind.data.frame(iris.predict, Species=iris.log$Species) #zadanie.6. plot(iris.predict$PC1, iris.predict$PC2, type="p", col="red", xlab="PC1", ylab="PC2") points(iris.predict$PC1[iris.predict$Species=="versicolor"], iris.predict$PC2[iris.predict$Species=="versicolor"], type="p", col="red") points(iris.predict$PC1[iris.predict$Species=="virginica"], iris.predict$PC2[iris.predict$Species=="virginica"], type="p", col="blue") legend("topleft", c("setosa","versicolor", "virginica"), col=c("red","blue", "green"), lty=1:1)

2547f9c9e3cbe4bf978a45b67e3fee397d3da703

9565039b0bb21e9e84dd98b619fc2684d195c405

/USCriminals.R

de812bfe70faedf9c270f7bef51bc26b8450731a

[]

no_license

Xiaoxi-X-G/Propensity

f8620e9b8f070e7ef08dee49aca6a01667ddfb0f

c615102c8fbf41fc32c62801faaad551af6a6556

refs/heads/master

2021-01-13T15:56:06.991008

2016-12-18T22:27:28

76,810,351

0

null

UTF-8

R

false

712

r

USCriminals.R

rm(list = ls()) RScriptPath<-"C:/gxx/r/project/USCriminal/" Column.type <- c("POSIXct", #Dates "factor", # Category "character", # Description "factor", # DayofWeek "factor", #PdDistrict "factor",#Resolution "factor", #Add "numeric", "numeric") DataRaw <- read.csv(paste(RScriptPath, "train.csv", sep=''), na.strings = c("", "NA"), colClasses = Column.type) DataRaw$Dates <- format(DataRaw$Dates, "%Y-%m-%d") ##### Visu ##### barplot(table(DataRaw$DayOfWeek)) barplot(table(DataRaw$PdDistrict)) barplot(table(DataRaw$Category))

c5245891a47dd631ea6118879282bf1e26cede8d

9bbdcb3936c5063edf237fe550fba4f5bf0a9b49

/man/cpPolyShapeNew.Rd

66e649c8c905e0d41f9de48a0e0710e16adfb864

[ "MIT" ]

permissive

coolbutuseless/chipmunkcore

b2281f89683e0b9268f26967496f560ea1b5bb99

97cc78ad3a68192f9c99cee93203510e20151dde

refs/heads/master

2022-12-10T17:56:15.459688

2020-09-08T22:40:10

288,990,789

17

1

null

UTF-8

R

false

true

818

rd

cpPolyShapeNew.Rd

% Generated by roxygen2: do not edit by hand % Please edit documentation in R/cpPolyShape.R \name{cpPolyShapeNew} \alias{cpPolyShapeNew} \title{Allocate and initialize a polygon shape with rounded corners. A convex hull will be created from the vertexes.} \usage{ cpPolyShapeNew(body, count, verts, transform, radius) } \arguments{ \item{body}{[\code{cpBody *}]} \item{count}{[\code{int}]} \item{verts}{[\code{cpVect *}]} \item{transform}{[\code{cpTransform *}]} \item{radius}{[\code{cpFloat}]} } \value{ [\code{cpShape *}] } \description{ Allocate and initialize a polygon shape with rounded corners. A convex hull will be created from the vertexes. } \details{ C function prototype: \code{CP_EXPORT cpShape* cpPolyShapeNew(cpBody *body, int count, const cpVect *verts, cpTransform transform, cpFloat radius);} }

8496bd027da4ae0216a64a6098b6cfce37eb6093

a6a2e430afe20b8b347c959933d7131639b8f818

/scripts/diff_exp_chip_seq.R

1adbe4786618bbebda2f5c6f9b2f4afa8aa36817

[]

no_license

fcadete/TRF2_siRNA

e1ee0bbcbaa3a885dff77427579beb6c53e51bcf

419d987ea1fc5b1c099b0658b21d01dcd99f507c

refs/heads/master

2020-04-01T17:29:06.315585

2019-01-15T10:07:02

153,432,728

0

null

UTF-8

R

false

18,733

r

diff_exp_chip_seq.R

library("DESeq2") library("Repitools") library("tximport") library("EnsDb.Hsapiens.v86") library("tidyverse") library("parallel") pdf("diff_expressed_encode_analysis.pdf", width = 10) samples <- read_tsv("sample_info.txt") samples$TimePoint <- as.character(samples$TimePoint) files <- file.path("salmon_on_hg19_output", samples$Filename, "quant.sf") names(files) <- samples$Filename tx2gene <- values(transcripts(EnsDb.Hsapiens.v86))[, c("tx_id", "gene_id")] txi <- tximport(files, type = "salmon", tx2gene = tx2gene, ignoreTxVersion = TRUE) ddsTxi <- DESeqDataSetFromTximport(txi, colData = samples, design = ~ siRNA + TimePoint + siRNA:TimePoint) ddsTxi <- DESeq(ddsTxi) res <- results(ddsTxi, contrast = c("siRNA", "control", "TRF2")) #Names of the genes siRNATRF2.TimePoint48.results <- results(ddsTxi, name = "siRNATRF2.TimePoint48") unique(genes(EnsDb.Hsapiens.v86, filter = GeneIdFilter(rownames( siRNATRF2.TimePoint48.results[which(siRNATRF2.TimePoint48.results$padj < 0.1),])))$symbol) # Get and plot the distance that these genes have to the nearest end of their respective chromosomes selected_genes_48h <- genes(EnsDb.Hsapiens.v86, filter = GeneIdFilter(rownames( siRNATRF2.TimePoint48.results[which(siRNATRF2.TimePoint48.results$padj < 0.1),]))) seqlevels(selected_genes_48h) <- paste0("chr", seqlevels(selected_genes_48h)) # Number of diff-expressed genes in the siRNATRF2.TimePoint96 interaction siRNATRF2.TimePoint96.results <- results(ddsTxi, name = "siRNATRF2.TimePoint96") unique(genes(EnsDb.Hsapiens.v86, filter = GeneIdFilter(rownames( siRNATRF2.TimePoint96.results[which(siRNATRF2.TimePoint96.results$padj < 0.1),])))$symbol) # Get and plot the distance that these genes have to the nearest end of their respective chromosomes selected_genes_96h <- genes(EnsDb.Hsapiens.v86, filter = GeneIdFilter(rownames( siRNATRF2.TimePoint96.results[which(siRNATRF2.TimePoint96.results$padj < 0.1),]))) seqlevels(selected_genes_96h) <- paste0("chr", seqlevels(selected_genes_96h)) all_genes <- genes(EnsDb.Hsapiens.v86) seqlevels(all_genes) <- paste0("chr", seqlevels(all_genes)) expressed_genes <- genes(EnsDb.Hsapiens.v86, filter = GeneIdFilter(rownames( siRNATRF2.TimePoint96.results[which(siRNATRF2.TimePoint96.results$baseMean > 20),]))) seqlevels(expressed_genes) <- paste0("chr", seqlevels(expressed_genes)) extraCols_narrowPeak <- c(signalValue = "numeric", pValue = "numeric", qValue = "numeric", peak = "integer") hela_encode_metadata <- read_tsv("metadata.tsv", col_names = TRUE) interesting_data <- hela_encode_metadata %>% filter(`Output type` == "optimal idr thresholded peaks", `File format` == "bed narrowPeak", `File Status` == "released", `Assembly` == "GRCh38") overlap_proportions <- as.data.frame(do.call(rbind, lapply(interesting_data$`File accession`, function(file_accession) { print(file_accession) this_bed <- rtracklayer::import(paste0("encode_files/", file_accession, ".bed.gz"), format = "BED", extraCols = extraCols_narrowPeak) all_genes_bed_overlaps <- countOverlaps(all_genes, this_bed) > 0 all_genes_bed_overlap_proportion <- sum(all_genes_bed_overlaps == TRUE) / length(all_genes_bed_overlaps) expressed_genes_bed_overlaps <- countOverlaps(expressed_genes, this_bed) > 0 expressed_genes_bed_overlap_proportion <- sum(expressed_genes_bed_overlaps == TRUE) / length(expressed_genes_bed_overlaps) selected_genes_48h_bed_overlaps <- countOverlaps(selected_genes_48h, this_bed) > 0 selected_genes_48h_bed_overlap_proportion <- sum(selected_genes_48h_bed_overlaps == TRUE) / length(selected_genes_48h_bed_overlaps) selected_genes_96h_bed_overlaps <- countOverlaps(selected_genes_96h, this_bed) > 0 selected_genes_96h_bed_overlap_proportion <- sum(selected_genes_96h_bed_overlaps == TRUE) / length(selected_genes_96h_bed_overlaps) all_genes_bed_overlaps_promoters <- countOverlaps(promoters(all_genes), this_bed) > 0 all_genes_bed_overlap_promoter_proportion <- sum(all_genes_bed_overlaps_promoters == TRUE) / length(all_genes_bed_overlaps_promoters) expressed_genes_bed_overlaps_promoters <- countOverlaps(promoters(expressed_genes), this_bed) > 0 expressed_genes_bed_overlap_promoter_proportion <- sum(expressed_genes_bed_overlaps_promoters == TRUE) / length(expressed_genes_bed_overlaps_promoters) selected_genes_48h_bed_overlaps_promoters <- countOverlaps(promoters(selected_genes_48h), this_bed) > 0 selected_genes_48h_bed_overlap_promoter_proportion <- sum(selected_genes_48h_bed_overlaps_promoters == TRUE) / length(selected_genes_48h_bed_overlaps_promoters) selected_genes_96h_bed_overlaps_promoters <- countOverlaps(promoters(selected_genes_96h), this_bed) > 0 selected_genes_96h_bed_overlap_promoter_proportion <- sum(selected_genes_96h_bed_overlaps_promoters == TRUE) / length(selected_genes_96h_bed_overlaps_promoters) return(rbind(data.frame(`File accession` = file_accession, group = "all", all_genes = all_genes_bed_overlap_proportion, expressed_genes = expressed_genes_bed_overlap_proportion, selected = selected_genes_48h_bed_overlap_proportion, time_point = "48h"), data.frame(`File accession` = file_accession, group = "all", all_genes = all_genes_bed_overlap_proportion, expressed_genes = expressed_genes_bed_overlap_proportion, selected = selected_genes_96h_bed_overlap_proportion, time_point = "96h"), data.frame(`File accession` = file_accession, group = "promoters", all_genes = all_genes_bed_overlap_promoter_proportion, expressed_genes = expressed_genes_bed_overlap_promoter_proportion, selected = selected_genes_48h_bed_overlap_promoter_proportion, time_point = "48h"), data.frame(`File accession` = file_accession, group = "promoters", all_genes = all_genes_bed_overlap_promoter_proportion, expressed_genes = expressed_genes_bed_overlap_promoter_proportion, selected = selected_genes_96h_bed_overlap_promoter_proportion, time_point = "96h"))) } )), stringsAsFactors = FALSE) overlap_proportions <- left_join(overlap_proportions, select(interesting_data, `File accession`, `Experiment target`, `Experiment accession`), by = c(File.accession = "File accession")) overlap_proportions_hg38 <- overlap_proportions %>% mutate(all_genes = as.numeric(all_genes), expressed_genes = as.numeric(expressed_genes), selected = as.numeric(selected)) p <- ggplot(overlap_proportions_hg38, aes(x = `Experiment target`, y = time_point, fill = log2(selected / all_genes))) + geom_tile() + coord_equal() + scale_fill_gradient2(low = "darkblue", high = "gold", mid = "white") + facet_grid(group ~ .) + theme_bw() + theme(axis.text.x = element_text(angle = 45, hjust = 1)) print(p) p <- ggplot(overlap_proportions_hg38, aes(x = `Experiment target`, y = time_point, fill = log2(selected / expressed_genes))) + geom_tile() + coord_equal() + scale_fill_gradient2(low = "darkblue", high = "gold", mid = "white") + facet_grid(group ~ .) + theme_bw() + theme(axis.text.x = element_text(angle = 45, hjust = 1)) print(p) chain <- rtracklayer::import.chain("hg19ToHg38.over.chain") interesting_data <- hela_encode_metadata %>% filter(`Output type` == "optimal idr thresholded peaks", `File format` == "bed narrowPeak", `File Status` == "released", `Assembly` == "hg19") overlap_proportions <- as.data.frame(do.call(rbind, lapply(interesting_data$`File accession`, function(file_accession) { print(file_accession) this_bed <- rtracklayer::import(paste0("encode_files/", file_accession, ".bed.gz"), format = "BED", extraCols = extraCols_narrowPeak) this_bed <- unlist(rtracklayer::liftOver(this_bed, chain)) all_genes_bed_overlaps <- countOverlaps(all_genes, this_bed) > 0 all_genes_bed_overlap_proportion <- sum(all_genes_bed_overlaps == TRUE) / length(all_genes_bed_overlaps) expressed_genes_bed_overlaps <- countOverlaps(expressed_genes, this_bed) > 0 expressed_genes_bed_overlap_proportion <- sum(expressed_genes_bed_overlaps == TRUE) / length(expressed_genes_bed_overlaps) selected_genes_48h_bed_overlaps <- countOverlaps(selected_genes_48h, this_bed) > 0 selected_genes_48h_bed_overlap_proportion <- sum(selected_genes_48h_bed_overlaps == TRUE) / length(selected_genes_48h_bed_overlaps) selected_genes_96h_bed_overlaps <- countOverlaps(selected_genes_96h, this_bed) > 0 selected_genes_96h_bed_overlap_proportion <- sum(selected_genes_96h_bed_overlaps == TRUE) / length(selected_genes_96h_bed_overlaps) all_genes_bed_overlaps_promoters <- countOverlaps(promoters(all_genes), this_bed) > 0 all_genes_bed_overlap_promoter_proportion <- sum(all_genes_bed_overlaps_promoters == TRUE) / length(all_genes_bed_overlaps_promoters) expressed_genes_bed_overlaps_promoters <- countOverlaps(promoters(expressed_genes), this_bed) > 0 expressed_genes_bed_overlap_promoter_proportion <- sum(expressed_genes_bed_overlaps_promoters == TRUE) / length(expressed_genes_bed_overlaps_promoters) selected_genes_48h_bed_overlaps_promoters <- countOverlaps(promoters(selected_genes_48h), this_bed) > 0 selected_genes_48h_bed_overlap_promoter_proportion <- sum(selected_genes_48h_bed_overlaps_promoters == TRUE) / length(selected_genes_48h_bed_overlaps_promoters) selected_genes_96h_bed_overlaps_promoters <- countOverlaps(promoters(selected_genes_96h), this_bed) > 0 selected_genes_96h_bed_overlap_promoter_proportion <- sum(selected_genes_96h_bed_overlaps_promoters == TRUE) / length(selected_genes_96h_bed_overlaps_promoters) return(rbind(data.frame(`File accession` = file_accession, group = "all", all_genes = all_genes_bed_overlap_proportion, expressed_genes = expressed_genes_bed_overlap_proportion, selected = selected_genes_48h_bed_overlap_proportion, time_point = "48h"), data.frame(`File accession` = file_accession, group = "all", all_genes = all_genes_bed_overlap_proportion, expressed_genes = expressed_genes_bed_overlap_proportion, selected = selected_genes_96h_bed_overlap_proportion, time_point = "96h"), data.frame(`File accession` = file_accession, group = "promoters", all_genes = all_genes_bed_overlap_promoter_proportion, expressed_genes = expressed_genes_bed_overlap_promoter_proportion, selected = selected_genes_48h_bed_overlap_promoter_proportion, time_point = "48h"), data.frame(`File accession` = file_accession, group = "promoters", all_genes = all_genes_bed_overlap_promoter_proportion, expressed_genes = expressed_genes_bed_overlap_promoter_proportion, selected = selected_genes_96h_bed_overlap_promoter_proportion, time_point = "96h"))) } )), stringsAsFactors = FALSE) overlap_proportions <- left_join(overlap_proportions, select(interesting_data, `File accession`, `Experiment target`, `Experiment accession`), by = c(File.accession = "File accession")) overlap_proportions_hg19 <- overlap_proportions %>% mutate(all_genes = as.numeric(all_genes), expressed_genes = as.numeric(expressed_genes), selected = as.numeric(selected)) p <- ggplot(overlap_proportions_hg19, aes(x = `Experiment target`, y = time_point, fill = log2(selected / all_genes))) + geom_tile() + coord_equal() + scale_fill_gradient2(low = "darkblue", high = "gold", mid = "white") + facet_grid(group ~ .) + theme_bw() + theme(axis.text.x = element_text(angle = 45, hjust = 1)) print(p) p <- ggplot(overlap_proportions_hg19, aes(x = `Experiment target`, y = time_point, fill = log2(selected / expressed_genes))) + geom_tile() + coord_equal() + scale_fill_gradient2(low = "darkblue", high = "gold", mid = "white") + facet_grid(group ~ .) + theme_bw() + theme(axis.text.x = element_text(angle = 45, hjust = 1)) print(p) wig_files_GRCh38 <- hela_encode_metadata %>% filter(`File format`== "bigWig", `Output type` == "fold change over control", `Assembly` == "GRCh38", `File Status` == "released", `Biological replicate(s)` == "1, 2") signal_windows <- mclapply(wig_files_GRCh38$`File accession`, function(file_accession) { print(file_accession) this_wig <- rtracklayer::import(paste0("encode_files/", file_accession, ".bigWig"), format = "bigWig", as = "RleList") selected_genes_48h_TSS <- promoters(selected_genes_48h) selected_genes_48h_views <- Views(this_wig, GRangesList(lapply(names(this_wig), function(x) selected_genes_48h_TSS[seqnames(selected_genes_48h_TSS) == x]))) selected_genes_48h_views <- selected_genes_48h_views[unlist(lapply(selected_genes_48h_views, length)) > 0] selected_genes_48h_matrix <- do.call(rbind, lapply(selected_genes_48h_views, as.matrix)) selected_genes_48h_frame <- data.frame(reshape2::melt(selected_genes_48h_matrix), group = "48h_genes", file_accession) selected_genes_96h_TSS <- promoters(selected_genes_96h) selected_genes_96h_views <- Views(this_wig, GRangesList(lapply(names(this_wig), function(x) selected_genes_96h_TSS[seqnames(selected_genes_96h_TSS) == x]))) selected_genes_96h_views <- selected_genes_96h_views[unlist(lapply(selected_genes_96h_views, length)) > 0] selected_genes_96h_matrix <- do.call(rbind, lapply(selected_genes_96h_views, as.matrix)) selected_genes_96h_frame <- data.frame(reshape2::melt(selected_genes_96h_matrix), group = "96h_genes", file_accession) expressed_genes_TSS <- promoters(expressed_genes) expressed_genes_views <- Views(this_wig, GRangesList(lapply(names(this_wig), function(x) expressed_genes_TSS[seqnames(expressed_genes_TSS) == x]))) expressed_genes_views <- expressed_genes_views[unlist(lapply(expressed_genes_views, length)) > 0] expressed_genes_matrix <- do.call(rbind, lapply(expressed_genes_views, as.matrix)) expressed_genes_frame <- data.frame(reshape2::melt(expressed_genes_matrix), group = "Expressed", file_accession) return(rbind(selected_genes_48h_frame, selected_genes_96h_frame, expressed_genes_frame)) }, mc.cores = 12 ) signal_windows <- do.call(rbind, signal_windows) signal_windows <- left_join(signal_windows, select(hela_encode_metadata, `File accession`, `Experiment target`), by = c("file_accession" = "File accession")) signal_windows <- signal_windows %>% ungroup() %>% mutate(diff_expressed = ifelse(group != "Expressed", TRUE, FALSE)) save(signal_windows, file="signal_windows.Rdata") p <- signal_windows %>% group_by(group, `Experiment target`, Var2) %>% summarise(mean_per_pos = mean(value)) %>% ggplot(aes(x = Var2 - 2000, y = mean_per_pos, colour = group)) + geom_line() + facet_wrap(~ `Experiment target`, scales = "free_y") + theme_bw() + theme(axis.text.x = element_text(angle = 45, hjust = 1)) print(p) p <- signal_windows %>% group_by(diff_expressed, `Experiment target`, Var2) %>% summarise(mean_per_pos = mean(value)) %>% ggplot(aes(x = Var2 - 2000, y = mean_per_pos, colour = diff_expressed)) + geom_line() + facet_wrap(~ `Experiment target`, scales = "free_y") + theme_bw() + theme(axis.text.x = element_text(angle = 45, hjust = 1)) print(p) dev.off()

ea207d04f0a35d34355e24b066ca7efb44589c78

41dfdb38b9437a7cf9965a4aaf002894c973c4c4

/data/rivers/eauAM/plot-PHYSICO-CHIMIE.R

abfd1f5fa21cfca079835cc0075863429d55e8ad

[]

no_license

jpgattuso/PointB-git

2f71f1c639b184c56d2c37a5c6fecc2c8bac5a20

f5acd1c2d82d59a7e95ad7153ad87ee4002399ec

refs/heads/master

2022-09-16T21:55:21.963042

2022-09-08T18:48:42

154,498,482

0

5

null

2022-09-08T18:48:43

2018-10-24T12:36:44

JavaScript

UTF-8

R

false

1,074

r

plot-PHYSICO-CHIMIE.R

Sys.setenv(TZ = "UTC") filein <- "eau-sup-AlpesMaritimes-PHYSICO-CHIMIE.RData" filepdf <- "eau-sup-AlpesMaritimes-PHYSICO-CHIMIE.pdf" load(filein) str(eau) for(n in names(eau)) { if(n != "value") { cat("--------------->", n, "\n") cat(levels(factor(eau[[n]])), sep = ";") cat("\n") } } dates <- as.POSIXct(strptime(eau$date, format = "%d/%m/%y")) range(dates) stations <- levels(factor(eau[["station"]])) params <- levels(factor(eau[["parameter"]])) str(stations) pdf(file = filepdf, width = 8, height = 11) for(param in params) { cat("----------------> plot", param, "\n") par(mfrow = c(4, 3), cex.lab = 0.7, cex.main = 0.7) for(station in stations) { i <- eau$station == station & eau$parameter == param if(!all(!i)) { cat(station, "", sep = ";") dtes <- dates[i] values <- eau$value[i] # Traitement des valeurs "<xxx"; ici mise à zéro si le signe "<" est rencontré values[grep("<", values)] <- "0" values <- as.numeric(values) plot(dtes, values, xlab = "", ylab = param, main = station) } } cat("\n") dev.flush() } graphics.off()

c858c39e97f4979ed593e78737e8069ef3dde2d5

b008570d05edad60a3a1675d0053e6b8911c00ef

/Lab04/Lab04.R

7e890f7e516440abb3bd33a44393eee4d30236a1

[]

no_license

brejai/CompeBioLabsAndHomework

328c80cdd1a95b972e8adcfec9f9c97bbe38e870

4d89ed47297508ff93686392cc8e7f41821db597

refs/heads/master

2020-12-31T07:09:37.637498

2017-04-21T15:46:40

80,553,263

0

null

UTF-8

R

false

961

r

Lab04.R

PiggyBank <- 10 Allowance <- 5 Gum <- 2*(1.34) for (i in seq(1,8)) { x <- PiggyBank + Allowance - Gum PiggyBank <- x print(x) } PopSize <- 2000 Shrinkage <- PopSize*(.05) for (i in seq(1,7)) { print(PopSize-Shrinkage) x <- PopSize-Shrinkage PopSize <- x Shrinkage <- PopSize*(.05) } Start <- 2500 n <- rep(2500, 12) for (t in seq(2,12)) { n[t] <- n[t-1] + (0.8 * n[t-1] * (10000-n[t-1])/10000) print(n[t]) } abundance <- n time <- 1:12 plot(time,abundance) n <- 18 data <- rep(0,n) for (i in seq(1,n)) { data[i] <- 3*i } data <- rep(0,n) data[1] <- 1 for (i in seq(2,n)){ data[i] <- 1 + (2*data[i-1]) } n <- 20 Fib <- rep(1,n) for (i in seq (3,n)){ Fib[i] <- Fib[i-1]+Fib[i-2] } CO2Data <- read.csv("compBioSandbox/CompBio_on_git/Labs/Lab04/CO2_data_cut_paste.csv") MetaData <- read.csv("compBioSandbox/CompBio_on_git/Labs/Lab04/MetaData_CO2_emissions.txt") #preallocate a matrix for the data nRows <- dim(CO2Data)[1]

b677a43c6e1b1172e2e7bcdbe1b9cedcf3d6ea63

47c81e91c91d6f321418042a69d5770b5aaadbdf

/tools/r_list_versions.R

dfc46ba5eba01648f9d1f7bc821608f531e1fc72

[ "Apache-2.0" ]

permissive

Kaggle/docker-rstats

f6e4c28638e5f9d33de59bcc56ac296da49f2176

2a42e7619ff99579011ca9cace98ee4604d9c068

refs/heads/main

2023-09-01T11:24:00.881089

2023-08-22T16:43:21

33,904,503

135

103

Apache-2.0

2023-08-29T14:50:52

2015-04-14T01:46:50

R

UTF-8

R

false

169

r

r_list_versions.R

ip <- as.data.frame(installed.packages()[,c(1,3:4)]) ip <- ip[is.na(ip$Priority),1:2,drop=FALSE] write.table(ip, quote=FALSE, sep="==", row.names=FALSE, col.names=FALSE)

050fce9be861fc8106911aaf21ddd28090c6ee05

f044402735a52fa040c5cbc76737c7950406f8b2

/BrCa_Age_Associated_TMA/Packages/biostatUtil/R/plotKMDetail.R

0ba68c9f23515f9661fa0f36281b5cec8265c91b

[]

no_license

BCCRCMO/BrCa_AgeAssociations

5cf34f3b2370c0d5381c34f8e0d2463354c4af5d

48a11c828a38a871f751c996b76b77bc33d5a3c3

refs/heads/master

2023-03-17T14:49:56.817589

2020-03-19T02:18:21

247,175,174

2

1

null

UTF-8

R

false

9,778

r

plotKMDetail.R

#' Plot detailed Kaplan-Meier curves #' #' KM plots with details of event counts. #' #' @param input.data input `data.frame` #' @param surv.formula survival formula to `Surv` #' @param main.text plot title #' @param xlab.text x-axis label #' @param ylab.text y-axis label #' @param line.name name of legend #' @param line.color line colour of survival curves #' @param line.pattern line pattern of survival curves #' @param line.width line width of survival curves #' @param show.test show single or the reference group value (for pairwise #' comparisons). If `"none"`, then no test is show. #' @param single.test.type test to show if specified `show.test = #' "single"`. Possible choices are `"logrank"` (default), #' `"wilcoxon"`, `"taroneware"`, or `"all"`. #' @param round.digits.p.value number of digits for p-value #' @param obs.survyrs show the observed survival years survival rate on KM plot #' @param ten.years.surv.95CI show ten year survival 95\% confidence interval #' @param event.count show the number of events at each time point #' @param legend.pos legend position keyword #' @param file.name name of file to save plot to #' @param file.width width of figure in saved file #' @param file.height height of figure in saved file #' @param grey.scale logical. If `TRUE`, the plot will be in grey scale. #' @param show.single.test.pos position to show single test; defaults to 0.5 if #' `legend.pos = "top"`. Otherwise 0.1 #' @param ... additional arguments to `plot` #' @author Samuel Leung #' @references #' http://courses.nus.edu.sg/course/stacar/internet/st3242/handouts/notes2.pdf #' @seealso [plotKM()] #' @export plotKMDetail <- function(input.data, surv.formula, main.text = "", xlab.text = "", ylab.text = "", line.name, line.color, line.pattern = NULL, line.width = NULL, show.test = "single", single.test.type = "logrank", round.digits.p.value = 4, obs.survyrs, ten.years.surv.95CI, event.count, legend.pos = "bottomleft", file.name = "no.file", file.width = 7, file.height = 7, grey.scale = FALSE, show.single.test.pos, ...) { var.name <- deparse(surv.formula[[3]]) # this should be the biomarker name # the deparse() function is used to make sure var.name is a string log.rank.p.values <- c() wilcox.p.values <- c() tarone.ware.p.values <- c() fit <- survival::survfit(surv.formula, data = input.data) # do not generate a file if "no.file" is specified if (file.name != "no.file" & nchar(file.name) > 4) { file.ext <- tools::file_ext(file.name) if (file.ext == "pdf") { grDevices::cairo_pdf(filename = file.name, width = file.width, height = file.height) # good for unicode character in e.g. line.name } else if (file.ext %in% c("wmf", "emf", "wmz", "emz")) { grDevices::png(filename = file.name, width = file.width, height = file.height) } else if (file.ext == "tiff") { grDevices::tiff(filename = file.name, width = file.width * 100, height = file.height * 100) } else { stop("Extension must be one of 'pdf', 'wmf', 'emf', 'wmz', 'emz', and 'tiff'.") } } # in case some strata do not have any cases which.strata.have.cases <- table(input.data[, var.name]) > 0 # default line width if (is.null(line.width)) { line.width <- 1 } if (grey.scale) { # gray scale plot if (is.null(line.pattern)) { line.pattern <- c(1:length(line.name))[which.strata.have.cases] } graphics::plot(fit, lty = line.pattern, lwd = line.width, main = main.text, xlab = xlab.text, ylab = ylab.text, ...) } else { # color plot if (is.null(line.pattern)) { line.pattern <- 1 } graphics::plot(fit, col = line.color[which.strata.have.cases], lty = line.pattern, lwd = line.width, main = main.text, xlab = xlab.text, ylab = ylab.text, ...) } # Legend 1 if (legend.pos == "top") { x.pos <- diff(range(fit$time, na.rm = TRUE)) / 2 y.pos <- 0.99 # top 1% ... since survival plot always starts at 100% survival } else { x.pos <- legend.pos y.pos <- NULL } l1 <- graphics::legend( x = x.pos, y = y.pos, legend = line.name, lty = line.pattern, lwd = line.width, box.lty = 0, cex = 0.8 ) # there seems to be need for the y-axis adjustment depending on the file.height ... dy <- 0.02 * (file.height - 7) / (12 - 7) # determined empirically if (legend.pos == "top") { y.pos <- l1$rect$top + dy } else { y.pos <- l1$rect$h - dy } # Legend 2 & 3 l2 <- graphics::legend( x = l1$rect$w + l1$rect$left, y = y.pos, legend = ten.years.surv.95CI, title = paste0(obs.survyrs, "yr 95% CI"), title.col = 1, box.lty = 0, cex = 0.8 ) l3 <- graphics::legend( x = l1$rect$w + l1$rect$left + l2$rect$w, y = y.pos, legend = event.count, title = "Events/N", title.col = 1, box.lty = 0, cex = 0.8 ) graphics::box() if (show.test == "single") { log.rank.test <- survival::survdiff(surv.formula, data = input.data, rho = 0) gehan.wilcox.test <- survival::survdiff(surv.formula, data = input.data, rho = 1) tarone.ware.test <- survival::survdiff(surv.formula, data = input.data, rho = 0.5) p.value <- getPval(log.rank.test) log.rank.p.values <- p.value p.value <- round(p.value, digits = round.digits.p.value) gehan.wilcox.p.value <- getPval(gehan.wilcox.test) wilcox.p.values <- gehan.wilcox.p.value gehan.wilcox.p.value <- round(gehan.wilcox.p.value, digits = round.digits.p.value) tarone.ware.p.value <- getPval(tarone.ware.test) tarone.ware.p.values <- tarone.ware.p.value tarone.ware.p.value <- round(tarone.ware.p.value, digits = round.digits.p.value) graphics::text( x = l1$rect$w + l1$rect$left + l2$rect$w + 1.3 * l3$rect$w, y = show.single.test.pos, # position of the test statistics on plot paste0( ifelse(sum(single.test.type %in% c("logrank", "all")) >= 1, paste0("Log-Rank p=", sprintf(paste0("%.", round.digits.p.value, "f"), p.value), "\n"), ""), ifelse(sum(single.test.type %in% c("wilcoxon", "all")) >= 1, paste0("Wilcoxon p=", sprintf(paste0("%.", round.digits.p.value, "f"), gehan.wilcox.p.value), "\n"), ""), ifelse(sum(single.test.type %in% c("taroneware", "all")) >= 1, paste0("Tarone-Ware p=", sprintf(paste0("%.", round.digits.p.value, "f"), tarone.ware.p.value), "\n"), "")), adj = c(0, 0), cex = 0.8) } else if (show.test != "none") { # assume show.test shows the reference group index legend.txt <- c() value.names <- names(table(input.data[, var.name])) for (value.name in value.names) { if (value.name == show.test) { # this is the reference group legend.txt <- c(legend.txt, "reference group") } else { # construct data temp.d <- input.data[input.data[, var.name] == show.test | input.data[, var.name] == value.name, ] if (sum(input.data[, var.name] == value.name, na.rm = TRUE) == 0) { # no case in this group p.value <- NA w.p.value <- NA t.p.value <- NA } else { # calculate log rank p-values p.value <- getPval(survival::survdiff(surv.formula, data = temp.d, rho = 0)) log.rank.p.values <- c(log.rank.p.values, p.value) p.value <- round(p.value, digits = round.digits.p.value) w.p.value <- getPval(survival::survdiff(surv.formula, data = temp.d, rho = 1)) wilcox.p.values <- c(wilcox.p.values, w.p.value) w.p.value <- round(w.p.value, digits = round.digits.p.value) t.p.value <- getPval(survival::survdiff(surv.formula, data = temp.d, rho = 0.5)) tarone.ware.p.values <- c(tarone.ware.p.values, t.p.value) t.p.value <- round(t.p.value, digits = round.digits.p.value) } new.txt <- paste0( ifelse("logrank" %in% single.test.type, paste0(p.value, " / "), ""), ifelse("wilcoxon" %in% single.test.type, paste0(w.p.value, " / "), ""), ifelse("taroneware" %in% single.test.type, t.p.value, "")) if (endsWith(new.txt, " / ")) { new.txt <- substr(new.txt, 0, nchar(new.txt) - 3) } legend.txt <- c(legend.txt, new.txt) } } legend.title <- paste0( ifelse("logrank" %in% single.test.type, "Log-Rank / ", ""), ifelse("wilcoxon" %in% single.test.type, "Wilcoxon / ", ""), ifelse("taroneware" %in% single.test.type, "Tarone-Ware ", "")) if (endsWith(legend.title, " / ")) { legend.title <- substr(legend.title, 0, nchar(legend.title) - 2) } legend.title <- paste0(legend.title, "P-values") l4 <- graphics::legend( x = l1$rect$w + l2$rect$w + l3$rect$w, y = y.pos, #y=l1$rect$h, legend = legend.txt, #text.col=line.color, title = legend.title, title.col = 1, box.lty = 0, cex = 0.8 ) } if (file.name != "no.file") { # do not generate a file if "no.file" is specified grDevices::dev.off() } return(list( "log.rank.p.values" = log.rank.p.values, "wilcox.p.values" = wilcox.p.values )) }

85cf3f8990d0b90fe4c5b40073493ad01bef4c85

f3ef1a4de6101d315afaeadeb3c0192033fd1a57

/covid_notebook.R

834e6f6dd3d0a2a4a6bcab4650a720accd9eb5b6

[]

no_license

increasingemtropy/covid_notebook

6e5a95e5dc9d80ca569ef375b54c8fc9bc475a45

b83678d193ad345da0c02c4a4cc4c29c567bb86b

refs/heads/master

2022-12-22T20:29:56.051606

2020-09-29T13:09:52

293,780,123

0

null

2020-09-29T13:09:53

2020-09-08T10:43:32

Jupyter Notebook

UTF-8

R

false

1,068

r

covid_notebook.R

library(ggplot2) # get up to date coronavirus data coronavirus_data <- read.csv("https://opendata.ecdc.europa.eu/covid19/casedistribution/csv", na.strings = "", fileEncoding = "UTF-8-BOM") data_subset <- subset(coronavirus_data,countriesAndTerritories %in% c("United_Kingdom","Germany","France","Spain","Japan","South_Korea")) # plot the cases vs date for UK vs USA plt <- ggplot(data_subset,aes(x=as.Date(dateRep,format = "%d/%m/%y"),y=Cumulative_number_for_14_days_of_COVID.19_cases_per_100000,color=countriesAndTerritories)) + geom_line(size=1) + coord_cartesian(xlim=c(as.Date('2020-03-15'),Sys.Date())) + labs(title ="14 day total cases per 10k population", x = "Date", y = "14 day case sum per 10k pop")+ theme(legend.position = "top")+ scale_color_brewer(name="",breaks=c("United_Kingdom","Germany","France","Spain","Japan","South_Korea"),labels=c("UK","Germany","France","Spain","Japan","South Korea"),palette = "Dark2") plt filename = paste("coronavirus_plot_",Sys.Date(),".png",sep='') ggsave(filename, width = 8, height = 5)

f86fb4ba757ae4dc026f725b184cebac4197ba68

9b1a760d45e21998b9d3871a1f4dac3a7a90c05a

/man/magplot.Rd

a2ac7276d9424766b6b9703fe91f5042b8276eab

[]

no_license

asgr/magicaxis

ac10b0b054128025976cb6b51003816cbd2157a9

0e3a56587021f8c22f86a3eda87907d8dfbe9e39

refs/heads/master

2023-06-21T11:28:06.031052

2023-06-19T06:30:03

13,343,972

9

4

null

2020-10-22T07:14:05

2013-10-05T11:12:59

R

UTF-8

R

false

11,633

rd

magplot.Rd

\name{magplot} \alias{magplot} %- Also NEED an '\alias' for EACH other topic documented here. \title{ Magically pretty plots } \description{ Makes scientific plots based on magaxis axes. Particularly designed for log plotting. Utilises base plot for the most part, but the axis drawing is replaced by a call to the magaxis fuction. } \usage{ magplot(x, y, z = NULL, log = "", main = "", side = 1:2, majorn = 5, minorn = 'auto', tcl = 0.5, ratio = 0.5, labels = TRUE, unlog = "auto", mgp = c(2,0.5,0), mtline = 2, xlab = '', ylab = '', crunch = TRUE, logpretty = TRUE, prettybase = 10, powbase = 10, hersh = FALSE, family = "sans", frame.plot = TRUE, usepar = FALSE, grid = TRUE, grid.col = 'grey90', grid.lty = 1, grid.lwd = 1, xlim = NULL, ylim = NULL, lwd = 1, axis.lwd = 1, ticks.lwd = axis.lwd, axis.col = 'black', zcol = hcl.colors(21), zstretch = 'lin', dobar = TRUE, ...) } %- maybe also 'usage' for other objects documented here. \arguments{ \item{x}{ The x coordinates of points/lines in the plot. Alternatively, a single plotting structure, function or any R object with a plot method can be provided. } \item{y}{ The y coordinates of points/lines in the plot, optional if x is an appropriate structure. } \item{z}{ The z coordinates for colour scaling of points in the plot. This will be passed through \code{\link{magmap}}, with dots passed as relevant. } \item{log}{ Log axis arguments to be passed to plot. E.g. use 'x', 'y', 'xy' or 'yx' as appropriate. Default '' assumes no logging of any axes. } \item{main}{ Title for the plot. Default is no title. } \item{side}{ The side to be used for axis labelling in the same sense as the base axis function (1=bottom, 2=left, 3=top, 4=right). A vector of multiple entries is allowed. By default, bottom and left axes are drawn (i.e. side 1 and 2). If \option{side}=FALSE then no sides or labels will be drawn. } \item{majorn}{ The target number of major-axis sub-divisions for pretty plotting. If length is 1 and length of side is longer than this value is used for all axes. If length of arguments is longer than 1 then these should tally with the relevant axes in side. Obvious reason for varying this is different pretty labelling between a and y axes. } \item{minorn}{ The exact number of minor-axis divisions (i.e. desired minor ticks + 1) to display in plotting. Auto will produce \code{\link{pretty}} ticks for linear scaling, and powbase-2 minor ticks for logged (this might seem odd, but for base 10 this means ticks at 2/3/4/5/6/7/8/9, which is probably as desired). If set manually, must be greater than 1 to have a visible effect. Minor ticks are always calculated to be equally spaced in linear space, so tick spaces vary when using log plotting. If length is 1 and length of side is longer than this value is used for all axes. If length of arguments is longer than 1 then these should tally with the relevant axes in side. An obvious reason for varying this is different pretty labelling between x and y axes. } \item{tcl}{ The length of major tick marks as a fraction of the height of a line of text. By default these face into the plot (in common with scientific plotting) with a value of 0.5, rather than the R default of -0.5. It is possible to force magaxis to inherit directly from par by setting usepar=TRUE (see below). See \code{\link{par}} for more details. } \item{ratio}{ Ratio of minor to major tick mark lengths. } \item{labels}{ Specifies whether major-axis ticks should be labelled for each axis. If length is 1 and length of side is longer than this value is used for all axes. If length of arguments is longer than 1 then these should tally with the relevant axes in side. Default is to label all axes. } \item{unlog}{ Determines if axis labels should be unlogged. If axis is found to be logged in par('usr') then the minor ticks are automatically log spaced, however "unlog" still controls how the labelling is done: either logged form (FALSE) or exponent form (TRUE). If axis has been explicitly logged (e.g. log10(x)) then this will can produce exponential axis marking/labelling if set to TRUE. This case will also produce log minor tick marks. If length of unlog is 1 and length of side is longer than 1 then the assigned unlog value is used for all axes. If length of arguments is longer than 1 then these should tally with the relevant axes in side. Can also take the text argument 'x', 'y', 'xy' or 'yx', where these refer to which axes have been logged. If left at the default of `auto' then unlog is assumed to be true when the axis in question is logged, and false otherwise. } \item{mgp}{ The margin line (in mex units) for the axis title, axis labels and axis line. This has different (i.e. prettier) defaults than R of c(2,0.5,0) rather than c(3,1,0). This pushes the numbers and labels nearer to the plot compared to the defaults. It is possible to force magaxis to inherit directly from par by setting usepar=TRUE (see below). See \code{\link{par}} for more details. } \item{mtline}{ Number of lines separating axis name from axis. If length 2 then specifies x and y axis separation respectively (else these are the same). } \item{xlab}{ x axis name. } \item{ylab}{ y axis name. } \item{crunch}{ In cases where the scientific text would be written as 1x10^8, should the 1x be removed so it reads 10^8. If length is 1 and length of side is longer then this value is used for all axes. If length of arguments is longer than 1 then these should tally with the relevant axes in side. TRUE by default. } \item{logpretty}{ Should the major-ticks only be located at powers of 10. This changes cases where ticks are placed at 1, 3.1, 10, 31, 100 etc to 1, 10, 100. If length is 1 and length of side is longer then this value is used for all axes. If length of arguments is longer than 1 then these should tally with the relevant axes in side. TRUE by default. } \item{prettybase}{ The unit of repitition desired. By default it is 10, implying a pretty plot is one with marks at 10, 20, 30 etc. If you are plotting degrees then it might be prettier to display 90, 180, 270 etc. In which case prettybase should be set to 90. If log=TRUE then the reference location of 10 is changed, so in the previous example the labels generated would be at 9, 90, 900 etc rather than the deafult of 1, 10, 100 etc. If length is 1 and length of side is longer then this value is used for all axes. If length of arguments is longer than 1 then these should tally with the relevant axes in side. } \item{powbase}{ Set the base to use for logarithmic axes. Default is to use 10. } \item{hersh}{ To determines whether all plot text should be passed using Hershey vector fonts. This applies to the axis labels (which are handled automatically) and the axis names. In the case of axis names the user must be careful to use the correct plot utils escape characters: http://www.gnu.org/software/plotutils/manual/en/html_node/Text-String-Format.html. magaxis will return back to the current plotting family after the function has executed. } \item{family}{ Specifies the plotting family to be used. Allowed options are 'sans' and 'serif'. Depending on whether hersh is TRUE or FALSE these otions are either applied to the Hershey vector fonts (hersh=TRUE) or the default R Helvetica font (hersh=FALSE). magaxis will return back to the current plotting family after the function has executed. } \item{frame.plot}{ Logical indicating whether a box should be drawn around the plot. } \item{usepar}{ Logical indicating whether tcl and mgp should be forced to inherit the global par values. This might be preferred when you want to define global plot settings at the start of a script. } \item{grid}{ Logical indicating whether a background grid should be drawn onto the plotting area. If true this will generate vertical and horiztonal grid lines. For more control (i.e. to only draw horizontal or verical lines) see \code{link{magaxis}}. } \item{grid.col}{ The colour of the grid to be drawn. } \item{grid.lty}{ The line type of the grid to be drawn. } \item{grid.lwd}{ The line width of the grid to be drawn. } \item{xlim}{ Vector; range of data to display. If this is set to NULL (default) then the limits will be estimated from the data dynamically. If length equals 1 then the argument is taken to mean the sigma range to select for plotting and the clipping is done by \code{\link{magclip}}. } \item{ylim}{ Vector; range of data to display. If this is set to NULL (default) then the limits will be estimated from the data dynamically. If length equals 1 then the argument is taken to mean the sigma range to select for plotting and the clipping is done by \code{\link{magclip}}. } \item{lwd}{ The width of plot lines to be drawn. This has different behaviour depending on the plot type. } \item{axis.lwd}{ The line width of the axis to be drawn. This is passed to \option{lwd} argument in \code{\link{axis}}. } \item{ticks.lwd}{ The line width of the ticks to be drawn. This is passed to \option{ticks.lwd} argument in \code{\link{axis}}. } \item{axis.col}{ Colour argument to pass directly to \option{col} in axis. It is a bit clunky to have to specify this, but the option 'col' clashes too much with line and point colours. } \item{zcol}{ Vector; a colour palette to use for \option{z} mapped colours. Must be a vector and not a function.. Only relevant if data has been passed to \option{z} } \item{zstretch}{ Character scalar; \option{z} colour stretch, either linear (lin, default) or logarithmic (log, good for large dynamic ranges). } \item{dobar}{ Logical; should a colour bar be added to the plot? } \item{\dots}{ Further arguments to be passed to: base \code{\link{plot}}; \code{\link{magaxis}} -> \code{\link{axis}}; \code{\link{magmap}} and \code{\link{magbar}} if \option{z} scaling is being used. } } \details{ This is a simple function that just turns off most of the plotting output of base plot, and replaces where possible those present in magaxis. If \option{x} is a data.frame with more than 2 columns then the utility base \code{\link{plot}} data.frame plotting function is used to create a full plotting grid. This ignores \code{\link{magaxis}} settings entirely. Setting \option{xlim} and \option{ylim} } \value{ No output. Run for the side effect of producing nice plotting axes. } \author{ Aaron Robotham } \seealso{ \code{\link{magaxis}}, \code{\link{maglab}}, \code{\link{magerr}}, \code{\link{magmap}}, \code{\link{magrun}} } \examples{ x=10^{1:9} y=1:9 magplot(log10(x),y,unlog='x') magplot(x,y,log='x') #Not ideal to have two decades between major labels: magplot(x,y,log='x',majorn=c(10,5)) magplot(x,y,log='xy',majorn=c(10,5,5,5),side=1:4) #Sometimes it is helpful to focus on where most of the data actually is. #Using a single value for xlim and ylim sigma clips the data to that range. #Here a value of 2 means we only show the inner 2-sigma (2\% to 98\%) range. #The 'auto' option allows magclip to dynamically estimate a clip value. temp=cbind(rt(1e3,1.5),rt(1e3,1.5)) magplot(temp) magplot(temp, xlim=2, ylim=2) magplot(temp, xlim='auto', ylim='auto') #Some astronomy related examples (and how to display the solar symbol): temp=cbind(runif(10,8,12),runif(10,0,5)) magplot(temp[,1:2], xlab=expression(M['\u0298']), ylab=expression(M['\u0298']/Yr), unlog='xy') #With z scaling z=sqrt(9:1) magplot(x, y, z, log='x', position='topleft') } % Add one or more standard keywords, see file 'KEYWORDS' in the % R documentation directory. \keyword{plot} \keyword{axis}% __ONLY ONE__ keyword per line \keyword{log}

169d7c25787eac2b488ef76575988a9b6f622bcb

9bc986ed5c9830e9b28406758b6c5a0e41f1fdcf

/R_DataAnalysis/R/1.lm_basic3.R

9279193c20061774cfabf180bef7390517cca7e5

[]

no_license

Jerrykim91/Bigdata_Analytics

ca9e8c86f494d44336091729c08da541e315a929

fab1f0522ff575537903790facf00f5a5c913262

refs/heads/master

2020-11-25T06:35:34.169050

2020-05-11T01:48:09

228,540,309

2

null

UTF-8

R

false

865

r

1.lm_basic3.R

require(graphics) ## 100 -> 70 : 30 / 80 : 20 # 70 -> 학습/ 30-> 검증(테스트) # 4 : 3 : 3 (학습/검증/테스트) m1<-c() m2<-c() for(i in 1:10){ sam<-sample(1:nrow(mtcars),nrow(mtcars)*0.7) fit2<-lm(mpg~.,data=mtcars[sam,]) # mean((fit2$residuals^2)) # MSE pred<-predict(fit2,mtcars[-sam,]) m1[i]<-mean((pred-mtcars[-sam,1])^2) ##회귀 분석의 경우 모델의 성능지표 MSE / MAPE / MAE index<-abs(fit2$coefficients)[-1] >0.5 var<-names(index)[index==T] fo<-paste0("mpg~",paste(var,collapse = "+")) fit3<-lm(fo,data=mtcars[sam,]) pred2<-predict(fit3,mtcars[-sam,]) m2[i]<-mean((pred2-mtcars[-sam,1])^2) cat("\n",i) ts.plot(cbind(m1,m2),col=c("red","blue")) } mean(m1) mean(m2) ## 다중 선형 회귀분석 ## (변수가 여러개, x와 y가 선형관계가 있다 라는 가정) ts.plot(fit3$residuals)

3297148f6201f79a95ba778c93fe3731e25e364e

162ad14e40fb0ffba7a8b52c83c3a3406d60adc2

/man/get.mat.omega.Rd

3040bf671fed15580787cf525456026bbb751dc6

[]

no_license

guillaumeevin/GWEX

c09c1f53a7c54eebc209b1f4aa5b8484fb59faf6

b1cae5f753a625d5963507b619af34efa2459280

refs/heads/master

2023-01-21T10:01:28.873553

2023-01-16T11:10:16

172,738,929

2

1

null

UTF-8

R

false

true

855

rd

get.mat.omega.Rd

% Generated by roxygen2: do not edit by hand % Please edit documentation in R/GWexPrec_lib.r \name{get.mat.omega} \alias{get.mat.omega} \title{get.mat.omega} \usage{ get.mat.omega(cor.obs, Qtrans.mat, mat.comb, nLag, nChainFit, isParallel) } \arguments{ \item{cor.obs}{matrix p x p of observed correlations between occurrences for all pairs of stations} \item{Qtrans.mat}{transition probabilities, 2 x ncomb matrix} \item{mat.comb}{matrix of logical: ncomb x nlag} \item{nLag}{order of the Markov chain} \item{nChainFit}{length of the simulated chains used during the fitting} \item{isParallel}{logical: indicate computation in parallel or not (easier for debugging)} } \value{ \item{matrix}{omega correlations for all pairs of stations} } \description{ find omega correlation leading to estimates cor between occurrences } \author{ Guillaume Evin }

ee357d4acd9c4873c786776299abeff475585762

eb89d6f54071a8c7bb9fc213b94d1af101121723

/tests/testthat/test-sag-gradients.R

8695c020c6567b34007ecab2017d8f7a9d0d1af5

[]

no_license

tdhock/bigoptim

cfe613c7098589e8179bee7c5805e69bd1e2fdf5

60245e925c9e8eb6ca7c48da9662eb0f2109d15f

refs/heads/master

2020-12-01T13:07:31.289218

2015-06-09T14:26:03

37,132,104

0

null

2015-06-09T12:58:35

2015-06-09T12:58:34

null

UTF-8

R

false

1,726

r

test-sag-gradients.R

context("Gradient tests") ## test parameters eps <- 1e-08 L2regularized.logistic.regression.gradient <- function(X, y, lambda, weight){ stop("TODO compute gradient in R code") } data(covtype.libsvm) test_that("gradient of covtype.libsvm is close to 0", { lambda <- 1 fit <- with(covtype.libsvm, sag_constant(X, y, lambda)) gradient <- with(covtype.libsvm, { L2regularized.logistic.regression.gradient(X, y, lambda, fit$w) }) expect_less_than(sum(abs(gradient)), eps) }) data(rcv1_train) test_that("gradient of rcv1_train is close to 0", { lambda <- 1 fit <- with(rcv1_train, sag_constant(X, y, lambda)) gradient <- with(rcv1_train, { L2regularized.logistic.regression.gradient(X, y, lambda, fit$w) }) expect_less_than(sum(abs(gradient)), eps) }) ## Simulating logistic datasets true_params <- c(1, 2, 3) sample_size <- 1000 sim <- .simulate_logistic(true_params, sample_size, intercept=FALSE) ################################# ## SAG with Constant Step Size ## ################################# test_that("constant step size Sag gradient norm is zero", { ## Fitting SAG pryr::mem_change({ sag_fit <- sag_constant(sim$X, sim$y, lambda=0, maxiter=NROW(sim$X) * 100) }) expect_less_than(norm(sag_fit$d, type="F"), eps) }) ######################### ## SAG with linesearch ## ######################### test_that("linesearch SAG gradient norm is zero", { expect_less_than(sag_ls(), eps) }) ########################################################## ## SAG with line-search and adaptive Lipschitz Sampling ## ########################################################## test_that("linesearch adaptive sag gradient norm is zero", { expect_less_than(sag_adaptive_ls(), eps) })

77b63fd9fad3a836b137661aa6475a17148df3fb

77776736c4b7311499e249e2d5f636e6ecbb4768

/static/slides/penalized-regression.R

2777a5b99466c754857d358b14a6e5d1b08c59d6

[]

no_license

turgeonmaxime/w20-stat7200

69227bed4b8ba30bfbce7c954cf91f5b278b45f8

0f30c8f753dd73756029bd44e8379610f4d9ed96

refs/heads/master

2020-09-29T02:38:56.815904

2020-04-03T22:40:07

226,928,767

2

1

null

UTF-8

R

false

5,983

r

penalized-regression.R

## ----setup, include=FALSE----------------------------------------------------- knitr::opts_chunk$set(cache=FALSE) ## ---- message = FALSE--------------------------------------------------------- library(tidyverse) url <- "https://maxturgeon.ca/w20-stat7200/prostate.csv" prostate <- read_csv(url) # Separate into training and testing sets data_train <- filter(prostate, train == TRUE) %>% dplyr::select(-train) data_test <- filter(prostate, train == FALSE) %>% dplyr::select(-train) ## ----------------------------------------------------------------------------- # OLS model1 <- lm(lpsa ~ ., data = data_train) pred1 <- predict(model1, data_test) mean((data_test$lpsa - pred1)^2) ## ----------------------------------------------------------------------------- # Ridge regression X_train <- model.matrix(lpsa ~ ., data = data_train) Y_train <- data_train$lpsa B_ridge <- solve(crossprod(X_train) + diag(0.7, 9), t(X_train)) %*% Y_train ## ----------------------------------------------------------------------------- X_test <- model.matrix(lpsa ~ ., data = data_test) pred2 <- X_test %*% B_ridge mean((data_test$lpsa - pred2)^2) ## ----------------------------------------------------------------------------- # Compare both estimates head(cbind(coef(model1), B_ridge)) ## ----------------------------------------------------------------------------- mse_df <- purrr::map_df(seq(0, 5, by = 0.1), function(lambda) { B_ridge <- solve(crossprod(X_train) + diag(lambda, 9), t(X_train)) %*% Y_train pred2 <- X_test %*% B_ridge mse <- mean((data_test$lpsa - pred2)^2) return(data.frame(MSE = mse, lambda = lambda)) }) ## ----------------------------------------------------------------------------- ols_mse <- mean((data_test$lpsa - pred1)^2) ggplot(mse_df, aes(lambda, MSE)) + geom_line() + theme_minimal() + geom_hline(yintercept = ols_mse) ## ---- message = FALSE--------------------------------------------------------- library(glmnet) # Fit for multiple values of lambda X_train <- model.matrix(lpsa ~ . - 1, data = data_train) ridge_fit <- glmnet(X_train, data_train$lpsa, alpha = 0, lambda = seq(0, 5, by = 0.1)) ## ---- message = FALSE--------------------------------------------------------- # Plot the value of the coefficients # as a function of lambda plot(ridge_fit, xvar = "lambda") abline(h = 0, lty = 2) ## ----------------------------------------------------------------------------- # Fit lasso regression along the same lambda sequence lasso_fit <- glmnet(X_train, data_train$lpsa, alpha = 1, # For lasso regression lambda = seq(0, 5, by = 0.1)) ## ----------------------------------------------------------------------------- X_test <- model.matrix(lpsa ~ . - 1, data = data_test) lasso_pred <- predict(lasso_fit, newx = X_test) lasso_mse <- apply(lasso_pred, 2, function(col) { mean((data_test$lpsa - col)^2) }) ## ----------------------------------------------------------------------------- lasso_mse_df <- data.frame(MSE = lasso_mse, lambda = seq(0, 5, by = 0.1)) ggplot(mse_df, aes(lambda, MSE)) + geom_line() + theme_minimal() + geom_hline(yintercept = ols_mse) + geom_line(data = lasso_mse_df, colour = 'red') ## ----------------------------------------------------------------------------- # Plot the value of the coefficients # as a function of lambda plot(lasso_fit, xvar = "lambda") abline(h = 0, lty = 2) ## ----------------------------------------------------------------------------- # Where is the min MSE? filter(lasso_mse_df, MSE == min(MSE)) # What are the estimates? coef(lasso_fit, s = 4.9) ## ----------------------------------------------------------------------------- # Take all the data dataset <- dplyr::select(prostate, -train) dim(dataset) ## ---- message = FALSE--------------------------------------------------------- set.seed(7200) library(caret) # 5-fold CV trainIndex <- createFolds(dataset$lpsa, k = 5) str(trainIndex) ## ----------------------------------------------------------------------------- # Define function to compute MSE compute_mse <- function(prediction, actual) { # Recall: the prediction comes in an array apply(prediction, 2, function(col) { mean((actual - col)^2) }) } ## ----------------------------------------------------------------------------- MSEs <- sapply(trainIndex, function(indices){ X_train <- model.matrix(lpsa ~ . - 1, data = dataset[-indices,]) Y_train <- dataset$lpsa[-indices] X_test <- model.matrix(lpsa ~ . - 1, data = dataset[indices,]) lasso_fit <- glmnet(X_train, Y_train, alpha = 1, lambda = seq(0, 5, by = 0.1)) lasso_pred <- predict(lasso_fit, newx = X_test) compute_mse(lasso_pred, dataset$lpsa[indices]) }) ## ----------------------------------------------------------------------------- # Each column is for a different fold dim(MSEs) CV_MSE <- colMeans(MSEs) seq(0, 5, by = 0.1)[which.min(CV_MSE)] ## ----------------------------------------------------------------------------- # What are the estimates? coef(lasso_fit, s = 0.4) ## ----------------------------------------------------------------------------- # Conveniently, glmnet has a function for CV # It also chooses the lambda sequence for you X <- model.matrix(lpsa ~ . -1, data = dataset) lasso_cv_fit <- cv.glmnet(X, dataset$lpsa, alpha = 1, nfolds = 5) ## ----------------------------------------------------------------------------- c("lambda.min" = lasso_cv_fit$lambda.min, "lambda.1se" = lasso_cv_fit$lambda.1se) # What are the estimates? coef(lasso_cv_fit, s = 'lambda.min') # 1 SE rule coef(lasso_cv_fit, s = 'lambda.1se')

d54304e426a7dc1c4101236a6a4fd361ccb0e37b

19984c47727b920e9c32ef30639abfd1d744b8ba

/Lab07/Lab07.R

f917dc322b260148661fbb1735a153cc30136741

[]

no_license

jadi9906/LABS

91665dc0476d561bcf2927c5fd371079e7e03201

8b94e8a55dc1cca154512d8aac8dd1dde9964c3a

refs/heads/master

2020-12-20T12:16:06.560803

2020-05-01T02:10:03

236,072,496

0

null

UTF-8

R

false

2,271

r

Lab07.R

#Lab 07: "Put the FUN in FUNction! :-) #Feb 28, 2020 #Jacob Di Biase # #Problem 1 #The area of a triangle can be calculated as 0.5 * base * height # Write a function named triangleArea that calculates and returns #the area of a triangle when given two arguments (base and height). triangleArea <- function(b,h) { area <- (b + h) * 0.5 return(area) } triangleArea(10,9) #Problem #2 #2A:Write a function named myAbs() that calculates and returns absolute values. myAbs <- function(x) { newx <- sqrt(x * x) return(newx) } myAbs(5) myAbs(-2.3) #2B:Revise your function to make it work on vectors #it already worked on vectors myAbs <- function(x) { newx <- sqrt(x * x) return(newx) } example <- c(1.1, 2, 0, -4.3, 9, -12) myAbs(example) #Problem #3 Fibonacci sequence #write a function that returns a vector of the first n Fibonacci numbers, #Your function should take two arguments: the user's desired value of n and the user's desired starting number # # myfib <- function(x,y){ fib <- rep(0,x) if(y==1){ fib[1] <- 1 fib[2] <- 1} if(y==0){ fib[1] <- 0 fib[2] <- 1} #if functions decide where the sequence begins for (i in 3:x) fib[i] <- fib[i-1] + fib[i-2] #for loop is the same fib sequence as previously done in an earlier lab return(fib) } myfib(10,1) myfib(10,0) #Problem #4 #4A:Write a function that takes two numbers as #its arguments and returns the square of the difference between them. fourthfunction <- function(x,y){ answer <- (x - y) ^ 2 return(answer) } fourthfunction(3,5) fourthfunction(c(2, 4, 6), 4) #Part 4b: Write a function of your own that calculates the average of a vector of numbers. arithmeticmean <- function(x){ XSum <- sum(x) XLength <- length(x) exampleMEAN <- (XSum/XLength) return(exampleMEAN) } arithmeticmean(c(5, 15, 10)) Lab07NewData <- DataForLab07[[1]] arithmeticmean(Lab07NewData) #Part 4c:the sum of squares can be calculated as the sum of the squared deviations from the mean #sum of squares SumofSquares <- function(x){ y <- arithmeticmean(x) newdata <- fourthfunction(x,y) results <- sum(newdata) #used the two previously made functions for mean and position - mean return(results) } Lab07NewData <- DataForLab07[[1]] SumofSquares(Lab07NewData)

eaf01c57c3793d5398e0a45be5022cc372b08c70

a27b79fc527614f1ae9ab192bec123f7ad55ff36

/R/markdown.R

c0ce0b7dd6b736425c5cc40e42b0a14b8dab00c9

[ "MIT" ]

permissive

r-lib/pkgdown

59528c00deab7466f678c48ed6e26227eecf1e6c

c9206802f2888992de92aa41f517ba7812f05331

refs/heads/main

2023-08-29T05:25:38.049588

2023-07-19T14:26:10

3,723,845

443

330

NOASSERTION

2023-09-06T09:08:11

2012-03-15T00:36:24

R

UTF-8

R

false

3,323

r

markdown.R

markdown_text <- function(text, ...) { if (identical(text, NA_character_) || is.null(text)) { return(NULL) } md_path <- withr::local_tempfile() write_lines(text, md_path) markdown_path_html(md_path, ...) } markdown_text_inline <- function(text, where = "<inline>", ...) { html <- markdown_text(text, ...) if (is.null(html)) { return() } children <- xml2::xml_children(xml2::xml_find_first(html, ".//body")) if (length(children) > 1) { abort( sprintf( "Can't use a block element in %s, need an inline element: \n%s", where, text ) ) } paste0(xml2::xml_contents(children), collapse="") } markdown_text_block <- function(text, ...) { html <- markdown_text(text, ...) if (is.null(html)) { return() } children <- xml2::xml_children(xml2::xml_find_first(html, ".//body")) paste0(as.character(children, options = character()), collapse = "") } markdown_body <- function(path, strip_header = FALSE) { xml <- markdown_path_html(path, strip_header = strip_header) # Extract body of html - as.character renders as xml which adds # significant whitespace in tags like pre transformed_path <- withr::local_tempfile() xml %>% xml2::xml_find_first(".//body") %>% xml2::write_html(transformed_path, format = FALSE) lines <- read_lines(transformed_path) lines <- sub("<body>", "", lines, fixed = TRUE) lines <- sub("</body>", "", lines, fixed = TRUE) structure( paste(lines, collapse = "\n"), title = attr(xml, "title") ) } markdown_path_html <- function(path, strip_header = FALSE) { html_path <- withr::local_tempfile() convert_markdown_to_html(path, html_path) xml <- xml2::read_html(html_path, encoding = "UTF-8") if (!inherits(xml, "xml_node")) { return(NULL) } # Capture heading, and optionally remove h1 <- xml2::xml_find_first(xml, ".//h1") title <- xml2::xml_text(h1) if (strip_header) { xml2::xml_remove(h1) } structure(xml, title = title) } markdown_to_html <- function(text, dedent = 4, bs_version = 3) { if (dedent) { text <- gsub(paste0("($|\n)", strrep(" ", dedent)), "\\1", text, perl = TRUE) } md_path <- withr::local_tempfile() html_path <- withr::local_tempfile() write_lines(text, md_path) convert_markdown_to_html(md_path, html_path) html <- xml2::read_html(html_path, encoding = "UTF-8") tweak_page(html, "markdown", list(bs_version = bs_version)) html } convert_markdown_to_html <- function(in_path, out_path, ...) { if (rmarkdown::pandoc_available("2.0")) { from <- "markdown+gfm_auto_identifiers-citations+emoji+autolink_bare_uris" } else if (rmarkdown::pandoc_available("1.12.3")) { from <- "markdown_github-hard_line_breaks+tex_math_dollars+tex_math_single_backslash+header_attributes" } else { if (is_testing()) { testthat::skip("Pandoc not available") } else { abort("Pandoc not available") } } rmarkdown::pandoc_convert( input = in_path, output = out_path, from = from, to = "html", options = purrr::compact(c( if (!rmarkdown::pandoc_available("2.0")) "--smart", if (rmarkdown::pandoc_available("2.0")) c("-t", "html4"), "--indented-code-classes=R", "--section-divs", "--wrap=none", ... )) ) invisible() }

3f19b28aa1ae9be91811b7393cb009eec58bad54

2a6f46cc8b818b8a498433df63f822673ef6d4b0

/LEARNINGPLOT.R

faecd6931cc55dc9c6a0b4085d99780b3948fad2

[]

no_license

goredoc/DissertationFilesFinalVersion

45b882d1166abeed790d8ce4a284bd1437a06314

d2c55e383f937271bb84bb12450dd6c21a9e307e

refs/heads/main

2023-02-19T02:55:57.040135

2021-01-24T18:37:24

326,014,503

0

null

UTF-8

R

false

810

r

LEARNINGPLOT.R

# LEARNINGPLOT learningPlot = function(burnin,i) { burnin = 500 oddsMeans = (1:(length(phi_names) / 2) * 2) - 1 samples = phi[(i*2)-1,,(burnin+1):length(phi[1,1,])] priorMu = mu_mean_vec[i] priorSigma = mu_sd_vec[i] loci = priorMu - 2*priorSigma hici = priorMu + 2*priorSigma losamp= min(samples) hisamp= max(samples) lowx = min(loci,losamp) highx= max(hici,hisamp) x=seq(lowx,highx,.001) y=dnorm(x,priorMu,priorSigma) hist(samples,xlim=c(lowx,highx),freq=FALSE,col="red",main=(paste0("Histogram with Prior Curve, ", phi_names[(i*2)-1])),xlab="Posterior Samples (Post-Burn-In)") arrows(mean(samples),0,mean(samples),10,len=0,lwd=1,col="white") points(y~x,xlim=c(lowx,highx),lty=3,col="gray",type="l",lwd=1) arrows(priorMu,0,priorMu,10,len=0,col="gray",lwd=1,lty=3) }

da380f77007e6f0dff195715fbacad295fe60399

cd951d8c027e1c679bfec654e4850d787c5ba960

/tests/test_bcSeq.R

0fced2e941ad2f47a3ee326f25e4aee784c8e67e

[]

no_license

jl354/bcSeq

ee0666fa402e90bb0bd7ccde0289906b54b9cc87

e97586cf05198876cda8a425e5d55810a2ba4137

refs/heads/master

2023-04-08T16:45:29.366402

2021-04-14T16:54:33

104,780,130

1

0

null

UTF-8

R

false

2,379

r

test_bcSeq.R

library(bcSeq) #devtools::load_all("../") #### Set the seed set.seed(4523) #### Generate barcode lFName <- "./libFile.fasta" bases <- c(rep('A', 4), rep('C',4), rep('G',4), rep('T',4)) numOfBars <- 7 Barcodes <- rep(NA, numOfBars*2) for (i in 1:numOfBars){ Barcodes[2*i-1] <- paste0(">barcode_ID: ", i) Barcodes[2*i] <- paste(sample(bases, length(bases)), collapse = '') } write(Barcodes, lFName) #### Generate reads and phred score rFName <- "./readFile.fastq" numOfReads <- 8 Reads <- rep(NA, numOfReads*4) for (i in 1:numOfReads){ Reads[4*i-3] <- paste0("@read_ID_",i) Reads[4*i-2] <- Barcodes[2*sample(1:numOfBars,1, replace=TRUE, prob=seq(1:numOfBars))] Reads[4*i-1] <- "+" Reads[4*i] <- paste(rawToChar(as.raw( 33+sample(20:30, length(bases),replace=TRUE))), collapse='') } write(Reads, rFName) #### perform alignment ReadFile <- "./readFile.fastq" BarFile <- "./libFile.fasta" outFile <- "./countH.csv" #### with default output for bcSeq_hamming #res <- bcSeq_hamming(ReadFile, BarFile, outFile, misMatch = 2, # tMat = NULL, numThread = 2, count_only = TRUE ) #res <- read.csv(outFile, header=FALSE) #res #### with return of alignment probability matrix to R #outFile <- "./countH2.csv" #res <- bcSeq_hamming(ReadFile, BarFile, outFile, misMatch = 2, # tMat = NULL, numThread = 2, count_only = FALSE ) #res #### with default output for bcSeq_edit outFile <- "./countE.csv" #res <- bcSeq_edit(ReadFile, BarFile, outFile, misMatch = 2, # tMat = NULL, numThread = 2, count_only = TRUE, # gap_left = 2, ext_left = 1, gap_right = 2, ext_right = 1, # pen_max = 7) #res <- read.csv(outFile, header=FALSE) #res #### with return of alignment probability matrix to R #outFile <- "./countE2.csv" #res <- bcSeq_edit(ReadFile, BarFile, outFile, misMatch = 2, # tMat = NULL, numThread = 2, count_only = FALSE, # gap_left = 2, ext_left = 1, gap_right = 2, ext_right = 1, # pen_max = 7) #res #### user-defined probability model comstomizeP <- function(m, x, y) { x * (1 - log(2) + log(1 + m / (m + y) ) ) } outFile = "comstomizeP.csv" #bcSeq_edit(ReadFile, BarFile, outFile, misMatch = 2, # tMat = NULL, numThread = 2, count_only = TRUE, # gap_left = 2, ext_left = 1, gap_right = 2, ext_right = 1, # pen_max = 7, userProb = comstomizeP)

24f584e7ff20291a00138bee5cb59cd2533b9e09

b83cde74005d5d837f0494ce17ff75af290cc12d

/man/techrep_fdata.Rd

33809b3244f203876f2d8bcf570b76cabfe31e42

[]

no_license

clabornd/pmartRdata

eea42713a6af5389a9e2035e14c8db990d03095f

34ea559378f1f0523bfaa09609c98cfa368c103a

refs/heads/master

2020-04-08T21:19:04.424102

2019-11-06T19:12:29

159,738,842

0

null

2018-11-29T23:05:44

null

UTF-8

R

false

true

855

rd

techrep_fdata.Rd

% Generated by roxygen2: do not edit by hand % Please edit documentation in R/data.R \name{techrep_fdata} \alias{techrep_fdata} \title{Peptide-level Technical Replicate Feature Data (f_data)} \format{A data.frame with 64 rows (samples) and 4 columns: \describe{ \item{RunID}{LC-MS run identifier (matches column headers in pep_edata)} \item{FACTOR}{Character string indicating either regular weight (RW), or obese (OB) groups} \item{DILUTION}{Character string indicating dilution of mouse plasma to Shewanella Oneidensis MR-1} \item{TECH_REP}{Character string indicating which technical replicates belong to the same biological sample} }} \source{ See details of \code{\link{pmartRdata}} for relevant grant numbers. } \description{ A dataset containing the technical replicate metadata, including technical replicate sample assignment variable. }

1e4d5330525733e69d5719b7577a85ad4f7c7a80

de882c7604b62c5975274bf0e3027da96a2f7b4d

/man/QTL_res_list.Rd

99d8e8094ea851a8fdb2294f420c548e01e62db4

[]

no_license

vincentgarin/mppR

bc061f2d0284adc6468619e162e8ffd45b641fb3

2fb0286257697f796c2c1b2590e9284ad281a939

refs/heads/master

2023-01-28T18:07:19.180068

2023-01-02T13:27:41

75,842,226

2

1

null

UTF-8

R

false

true

1,043

rd

QTL_res_list.Rd

% Generated by roxygen2: do not edit by hand % Please edit documentation in R/QTL_res_list.R \name{QTL_res_list} \alias{QTL_res_list} \title{List of QTL results} \usage{ QTL_res_list(mppData, MPP_out, trait, Q.eff, VCOV, res_file = c()) } \arguments{ \item{mppData}{An object of class \code{mppData}.} \item{MPP_out}{Output from \code{\link{mpp_proc}}.} \item{trait}{\code{character} indicator to specify the trait name.} \item{Q.eff}{\code{Character} expression indicating the assumption concerning the QTL effect: 'cr', 'par', 'anc', 'biall' or 'MQE'.} \item{VCOV}{\code{Character} expression defining the type of variance covariance structure used.} \item{res_file}{\code{data.frame} to store the QTL effects results. Default, empty file.} } \value{ The results of \code{MPP_out} are appended to \code{res_file}. } \description{ Form a list of QTL results appending QTL effects results obtained during different QTL detection procedure. The results are appended to \code{res_file}. } \examples{ # not yet } \author{ Vincent Garin }

b6c175bbead875f7b13d51ee383d1eba284d3b3d

8dc13f15bc5c086d658b71a67a120b8eba2f388c

/R/vsm_info.R

9c56aa5d1b2ff207ae714548e6185a0232a90cf2

[]

no_license

thomasgredig/quantumPPMS

4499966224e302f27e0a138f6ed74408493204b2

c79cd5027b21446d168d8775a378e95b7dabe7e5

refs/heads/master

2022-05-23T18:35:58.955287

2022-04-09T21:29:55

182,595,380

1

0

null

UTF-8

R

false

2,555

r

vsm_info.R

#' Reads VSM data file header #' #' also returns the PPMS option (VSM,ACMS,LogData,Resistivity) #' #' @param filename filename including path #' @return data frame #' @examples #' filename = vsm.getSampleFiles()[1] #' vsm.info(filename) #' @export vsm.info <- function(filename) { # check if file exists if (!file.exists(filename)) { warning(paste('Cannot find file:',filename)) return() } v = vsm.version(filename) if (v==1.5667) skipLEN = list(30,30,TRUE, cols=c(1,4,3,5,6)) if (v==1.56) skipLEN = list(19,19,TRUE, cols=c(2,4,5,7,8)) if (v==1.0914) skipLEN = list(20,20,TRUE, cols=c(2,3,4,7,8)) if (v==1.2401) skipLEN = list(22,23,FALSE, cols=c(2,3,4,5,6)) if (v==1.36) skipLEN = list(22,23,FALSE, cols=c(2,3,4,5,6)) if (v==1.3702) skipLEN = list(22,23,FALSE, cols=c(2,3,4,5,6)) scan(file = filename, nlines=skipLEN[[1]], what=character(0), sep='\n', quiet = TRUE) -> header d=data.frame() if ((length(header)>0) && (header[1]=='[Header]')) { ppms.option = gsub(' ','',strsplit(header[grep('^BYAPP,',header)],',')[[1]][2]) title = gsub('TITLE,','',header[grep('^TITLE', header)]) # [1] "FILEOPENTIME" "5500334.30" "09/21/2018" "4:50 pm" filedate = as.character(strptime(paste(gsub(',','',strsplit(header[grep('FILEOPENTIME,',header)],' ')[[1]][c(3,4,5)]), collapse=' '), format='%m/%d/%Y %I:%M:%S %p')) # filedate = as.character(strptime(paste(strsplit(header[grep('FILEOPENTIME,',header)],',')[[1]][c(3,4)], collapse=' '), # format='%m/%d/%Y %I:%M %p')) dl.appname = grep('APPNAME',header) appname = gsub(',\\s*','',gsub('INFO','',gsub('APPNAME','',header[dl.appname]))) header = header[-dl.appname] info.str = gsub('^INFO,','',header[grep('INFO',header)]) if (ppms.option == 'ACMS') { attr = info.str[1:4] attr.names = paste0('ACMS.INFO',1:4) } else { attr = gsub('\\s*(.*)[,:][^,]+','\\1',info.str) attr.names = gsub('.*[,:]\\s*([^,]+)','\\1',info.str) if (v==1.5667) { tmp = attr attr = attr.names attr.names = tmp } } d = data.frame(rbind(c(ppms.option, title, filedate, appname, attr)), stringsAsFactors = FALSE) names(d) = c('option','title','file.open.time','AppName', attr.names) # guess the sample name d$sample.name = gsub('.*([A-Z]{2,3}\\d{6,8}[a-zA-Z]{0,2}\\d{0,1}).*','\\1', paste(paste(d, collapse=' == '), filename)) } d }

fa9537401de80d14674fb5418985769c424b8652

24c8edec774d1b2ec1fa7f75af41fec5c036f9a6

/R Programming Programming Assignment 2/cachesolve.R

f28490c8a523dd651ef84817bb01fd76e59ca43f

[]

no_license

Dingkai1996/R-programing-Programming-Assignment-2

f12970d8b29b1590c41d88f1f1f666b320bad383

bd83be78d8a479323c5895b7927369654d3119c0

refs/heads/main

2023-06-24T14:14:13.360753

2021-07-22T21:30:35

384,751,511

0

null

UTF-8

R

false

2,486

r

cachesolve.R

makeCacheMatrix <- function(x=numeric(), y, z){ m <- NULL ##initially set the inverse to NULL if( lapply(length(x), as.numeric) != y*z){ ##check whether it can form matrix or not message("can't make matrix") return() } else if(y != z){ ##check whether the matrix have inverse or not message("can't make the matrix with inverse") return() } else{ set <- function(a, b, c){ ##set matrix in parent env with the desired value, if inverse is already set, get rid of it! x <<- a y <<- b z <<- c m <<- NULL } s <- matrix(x, y, z) ##make the matrix with value get <- function() s ##get the matrix, add it into get setinverse <- function(inverse) m <<- inverse ##set the inverse for the matrix as m getinverse <- function() m ##get the inverse of matrix, add it into getinverse list(set = set, get = get, setinverse = setinverse, getinverse = getinverse) ##given the list variable from the first function, will first check to see if there's already a cached inverse and return ##otherwise will attempt to solve its inverse and set/return it } } cachesolve <- function(s, ...){ m <- s$getinverse() ##check if there's cached value there already if(!is.null(m)){ ##if there is cached inverse already, return it message("getting cached data") return(m) } data <- s$get() ## else, get the matrix from get m <- solve(data, ...) ## make the inverse of matrix s$setinverse(m) ## set the inverse into setinverse m ## show the result of inverse }

1a3fa86c503a8d34f3974fd20164fca33cfb0b90

2d02239b9d095490b34dcf77b891b65263aa9ddc

/lee_ready.R

a9b158b9422e27cdc4a66aa93b1788cfec166768

[]

no_license

julienneves/herding

f629d8858b836e84e1b3253b953cce2baa06c2d8

e6891da7394eb266e1e3f45589d9e7bbbaed173f

refs/heads/master

2021-01-11T00:27:25.605447

2017-01-20T19:37:53

70,556,432

0

null

UTF-8

R

false

1,698

r

lee_ready.R

lee_ready <- function(tqdata, delay = default_delay(tqdata)){ x = getTradeDirection(tqdata) trade = xts(x = x, order.by = time(tqdata), unique = FALSE, tzone = attr(tqdata,'tzone')) trade_split <- split.xts(trade, f= "days") trade_split <- lapply(trade_split, insert_no_trade, delay) trade <- do.call(rbind,trade_split) trade <- merge(trade,xts(x = matrix(rep(NA,9*length(trade)),ncol = 9), order.by = time(trade))) colnames(trade) <- c("x", "prob_x", "beta", "sigma", "prob_x_h", "prob_x_l", "prob_x_n", "prob_v_h", "prob_v_l", "prob_v_n") return(trade) } insert_no_trade <- function(trade, delay = NULL){ time_trade <- time(trade) time_no_trade <- test_inactivity(time_trade, delay) no_trade <- xts(x = rep(0,length(time_no_trade)), order.by = time_no_trade, unique = FALSE, tzone = attr(tqdata,'tzone')) no_trade[time_trade,] <- trade trade <- no_trade return(trade) } test_inactivity <- function(time_trade, delay){ diff_time <- diff(time_trade) if(!all(diff_time<=delay)){ time_trade <- c(time_trade[diff_time>delay]+delay,time_trade) time_trade <- time_trade[order(time_trade)] time_trade <- add_no_trade(time_trade, delay) } return(time_trade) } delay = trade_split <- split.xts(trade, f= "days") default_delay <- function(tqdata) { x = getTradeDirection(tqdata) trade = xts(x = x, order.by = time(tqdata), unique = FALSE, tzone = attr(tqdata,'tzone')) trade_split <- split.xts(trade, f= "days") time_mean <- function(trade) mean(diff(time(trade))) return(mean(sapply(trade_split, time_mean))) }

f8608efe3c48967ac13bce660d74aa90a350a56e

03c99906a94c70e9a13e7714aad996f461f339c1

/R/dsimFun.R

a6097dec490cd9e65001f6e1b24e774ccd3ac0f3

[]

no_license

cran/adiv

6a111f6a1ef39fe302a2f882b9a9d04e7d652c04

d65d6e0301e4611a94a91933299bff1fdc06d96b

refs/heads/master

2022-10-28T08:07:33.352817

2022-10-06T12:40:04

97,764,074

1

null

UTF-8

R

false

2,751

r

dsimFun.R

dsimFun <- function(df, vartype=c("Q","N","M","P"), method=1:5, type=c("similarity", "dissimilarity")){ type <- type[1] if(!type%in% c("dissimilarity", "similarity")) stop("type must be either dissimilarity or similarity") meantype <- method[1] if(!meantype%in%(1:5)) stop("Incorrect definition of method") fun0 <- function(i){ df0 <- as.matrix(df[[i]]) type <- type[1] vartype0 <- vartype[i] if(vartype0=="Q" | vartype0=="N"){ if(type=="dissimilarity") return(daisy(df0, metric = "gower")*ncol(df0)) else return((1-as.matrix(daisy(df0, metric = "gower")))*ncol(df0)) } if(vartype0=="P"){ df0 <- sweep(df0, 1, rowSums(df0), "/") } if(vartype0=="P" | vartype0=="M"){ A <- df0%*%t(df0) B <- diag(A)%*%t(rep(1, nrow(df0))) C <- rep(1, nrow(df0))%*%t(diag(A)) if(meantype==4) S <- A/sqrt(B)/sqrt(C) else if(meantype==3){ S <- 2*A/(B+C) } else if(meantype==1){ S <- A/(2*B+2*C-3*A) } else if(meantype==2){ S <- A/(B+C-A) } else S <- 4*A/(2*A+B+C) rownames(S)<-colnames(S)<-rownames(df0) if(type=="dissimilarity") return(as.dist(1-S)) else return(S) } } if(inherits(df, "ktab")){ listdsim <- lapply(1:length(df$blo), fun0) res <- listdsim[[1]] if(length(listdsim)>1){ for(i in 2:length(listdsim)) res <- res + listdsim[[i]] } nk <- length(vartype[vartype!="Q" & vartype!="N"]) nk <- nk + sum(df$blo[vartype=="Q" | vartype=="N"]) return(res/nk) } else{ df <- as.matrix(df) type <- type[1] vartype <- vartype[1] if(vartype=="Q" | vartype=="N"){ if(type=="dissimilarity") return(daisy(df, metric = "gower")) else return(1-as.matrix(daisy(df, metric = "gower"))) } if(vartype=="P"){ df <- sweep(df, 1, rowSums(df), "/") } if(vartype=="P" | vartype=="M"){ A <- df%*%t(df) B <- diag(A)%*%t(rep(1, nrow(df))) C <- rep(1, nrow(df))%*%t(diag(A)) if(meantype==4) S <- A/sqrt(B)/sqrt(C) else if(meantype==3){ S <- 2*A/(B+C) } else if(meantype==1){ S <- A/(2*B+2*C-3*A) } else if(meantype==2){ S <- A/(B+C-A) } else S <- 4*A/(2*A+B+C) rownames(S)<-colnames(S)<-rownames(df) if(type=="dissimilarity") return(sqrt(1-S)) else return(S) } } }

0c95ef1fdd2981474df8f691cf9bbaeb3446cf00

f7a0f3cbeefdc01fc0f172a47359c0c4610c95a7

/code_active/feature_fit.R

cc0d357036a87727c7dd93ef49d5f3a1e04a830b

[]

no_license

EESI/exploring_thematic_structure

65e77efbb56fea646a9f165eaa94f955f68259ff

06f7ea096c31dbb63b09fc117ee22411e52ab60e

refs/heads/master

2020-08-25T03:30:04.253394

2019-10-23T03:02:18

216,955,082

2

0

null

UTF-8

R

false

6,568

r

feature_fit.R

feature_fit <- function(num_features,out,fit, train_docs,test_docs,train_meta,test_meta, vocab,variable,beta=NULL){ print(out$dx$imp$RF1500DS) if (is.null(beta)){ topic_features <- order(out$dx$imp$RF1500DS$importance[,1],decreasing=TRUE)[1:num_features] fit$beta$logbeta[[3]] <- do.call('rbind',fit$beta$logbeta)[topic_features,] fit$settings$dim$K <- NROW(fit$beta$logbeta[[3]]) }else{ topic_features <- 1:NROW(beta) fit$beta$logbeta[[3]] <- beta fit$settings$dim$K <- NROW(fit$beta$logbeta[[3]]) } cat('\nSelecting topics: ',topic_features,'\n\n') z_covariate <- 3 cat('Exploring training set posterior\n') train_fit <- ctm_frozen(fit,train_docs,vocab, seed=seed_train,max.em.its=500,emtol=1e-5,avg_iters=1, verbose=TRUE,true_doc_content=TRUE, data=train_meta,covariate=z_covariate, parallel=TRUE,nc=25) if(exists('cl')) stopCluster(cl) cat('Exploring testing set posterior\n') test_fit <- ctm_frozen(fit,test_docs,vocab, seed=seed_test,max.em.its=500,emtol=1e-5,avg_iters=1, verbose=TRUE,true_doc_content=TRUE, data=test_meta,covariate=z_covariate, parallel=FALSE,nc=10) if(exists('cl')) stopCluster(cl) K <- train_fit$settings$dim$K load_fit <- FALSE save_fit <- FALSE betas <- beta_prep(train_fit,test_fit,counts,otu_taxa_xavier,vocab, save_fit,save_dir,save_fit_foldername,dupnum,seed_permuted) beta_frozen_ra <- betas$beta_frozen_ra beta_frozen <- betas$beta_frozen beta_meta <- betas$beta_meta beta_otu <- betas$beta_otu out_new <- eval_labels(save_fit,load_fit,train_fit$Z_bar,test_fit$Z_bar,train_meta,test_meta, save_dir,save_fit_foldername,save_coef_filename,beta_type=model, nc=60,test='dx') colnames(train_fit$Z_bar) <- paste0('T',topic_features) names(out_new$dx$lasso) <- paste0('T',topic_features) names(out_new$dx$en1) <- paste0('T',topic_features) names(out_new$dx$en2) <- paste0('T',topic_features) rownames(out_new$dx$imp$RF1500DS$importance) <- paste0('T',topic_features) plot_z_heatmap(train_fit$Z_bar,train_meta,out_new$dx$en2, dist1='jaccard',dist2='jaccard', clust1='ward.D2',clust2='ward.D2', transform='none', main='', rowclust=FALSE,variable=variable) colnames(test_fit$Z_bar) <- paste0('T',topic_features) plot_z_heatmap(test_fit$Z_bar,test_meta,out_new$dx$en2, dist1='jaccard',dist2='jaccard', clust1='ward.D2',clust2='ward.D2', transform='none', main='', rowclust=FALSE,variable) return(list(out=out_new, z_bar_train=train_fit$Z_bar, train_meta=train_meta, z_bar_test=test_fit$Z_bar, test_meta=test_meta)) } library(stm) library(biom) library(readr) library(tidyr) library(dplyr) library(fastICA) library(randomForest) library(stringr) library(kernlab) library(Rcpp) library(parallel) library(foreach) library(ape) library(phyloseq) library(doParallel) library(stm) library(LDAvis) library(caret) library(glmnet) library(ggplot2) library(knitr) library(gridExtra) params <- expand.grid(K=c(25,50,75,125,200), content=c(FALSE,TRUE), variable=c('DIAGNOSIS','ISOLATION_SOURCE')) params <- params %>% filter(!(content == FALSE & variable == 'ISOLATION_SOURCE'), !(content == TRUE & K == 200)) %>% arrange(K,variable) param <- 2 source('~/Dropbox/stm_microbiome/code_active/stm_functions.R') source('~/Dropbox/stm_microbiome/code_active/nav_froz_fxns_3.R') source('~/Dropbox/stm_microbiome/code_active/performance_1.R') source('~/Dropbox/stm_microbiome/code_active/framework.R') load_fit <- TRUE save_fit <- FALSE save_output <- FALSE random_seed <- FALSE K <- params[param,]$K cn_normalize <- TRUE content <- params[param,]$content seq_sim <- 's97' variable <- as.character(params[param,]$variable) dupnum <- NULL prepare_framework(random_seed,K,cn_normalize,content,variable,seq_sim) load_fits(file.path(save_dir,save_fit_foldername,save_fit_filename)) # take combined (K x M where M is the number of features, 2 for dx) b <- 1 train_fit <- fit_frozen[[b]][['train']] test_fit <- fit_frozen[[b]][['test']] K <- train_fit$settings$dim$K model <- str_replace_all(names(fit_frozen)[b],' ','') betas <- beta_prep(train_fit,test_fit,counts,otu_taxa_xavier,vocab, save_fit,save_dir,save_fit_foldername,dupnum,seed_permuted,model) beta_frozen_ra <- betas$beta_frozen_ra beta_frozen <- betas$beta_frozen beta_meta <- betas$beta_meta beta_otu <- betas$beta_otu # evaluate the performance of the combined beta matrix cat('Evaluating predictive performance for',names(fit_frozen)[b],'\n') out <- eval_labels(save_fit,load_fit,train_fit$Z_bar,test_fit$Z_bar,train_meta,test_meta, save_dir,save_fit_foldername,save_coef_filename,beta_type=model, nc=60) # take the top F topics via RF importance (tree = 1500) for predicting the target labels (dx) # and subset the beta matrix with the indexes for these features, yielding an N x F matrix. # Generate the topic assignments (z bar) via the subsetted beta matrix. Then, evaluluate the # predictive performance to color code the topics via the EN output, then plot a heatmap. For # the heatmap, the columns are ordered in terms for decreasing PCDAI score, where CD- gets a 0. ff <- feature_fit(30,out,fit,train_docs,test_docs,train_meta,test_meta,vocab,variable) ff$out$dx$score ff$out$dx$imp$RF1500DS plot_z_heatmap(ff$z_bar_train,ff$train_meta,ff$out$dx$lasso, dist1='jaccard',dist2='jaccard', clust1='ward.D2',clust2='ward.D2', transform='none', main='', rowclust=TRUE,variable='DIAGNOSIS') plot_z_heatmap(ff$z_bar_test,ff$test_meta,ff$out$dx$lasso, dist1='jaccard',dist2='jaccard', clust1='ward.D2',clust2='ward.D2', transform='none', main='', rowclust=TRUE,variable='DIAGNOSIS')

2b17b9ea100c8e7a4ca35d4885c5895e4dc0779b

d359475bd587e8f364ca2ece794f9358fce66a84

/model/r_scripts/functions/get_init.R

c1bc37fc07340a45b3c61de99d7fdbd6103bb04a

[]

no_license

umich-cphds/cov-ind-19

c22cc191bb64b1b2c581e28afaa2260a7583143c

ebb3c0093fff66d39d7a7cf14397efe269731b4a

refs/heads/master

2023-02-18T07:43:37.731747

2023-02-06T17:24:29

249,284,918

15

6

null

2023-02-02T16:17:31

2020-03-22T22:32:42

R

UTF-8

R

false

769

r

get_init.R

get_init <- function(data) { tmp_data <- data %>% filter(date < min_date) %>% dplyr::select(-date) %>% summarize( Confirmed = sum(Confirmed, na.rm = TRUE), Recovered = sum(Recovered, na.rm = TRUE), Deceased = sum(Deceased, na.rm = TRUE) ) %>% as.numeric(as.vector(.)) tmp <- data %>% filter(date >= min_date & date <= max_date) data_initial <- c(tmp_data, tmp %>% filter(date == min_date) %>% dplyr::select(-date) %>% as.numeric(as.vector(.)) ) if (data_initial[1] == 0) {data_initial[1] <- 1} if (data_initial[4] == 0) {data_initial[4] <- 1} # check with Ritwik/Ritoban if this is necessary return(data_initial) }

44b10a057744aec3595f649561413b730ed68a96

ee5573b3b198214f0d5db33015e635182a75e1ca

/binomial/R/bin_distribution.R

22d36f1e3c1754fc4e1d747ba35e85e22384c425

[]

no_license

stat133-sp19/hw-stat133-TheGoldenKyle

e6b6a946963ac9aececeefdfedbc59133a7786d2

e11cb3f4722b1a490acebc548607e08e7861b222

refs/heads/master

2020-04-28T13:28:53.897190

2019-05-03T19:55:32

175,308,057

0

null

UTF-8

R

false

1,022

r

bin_distribution.R

#' @title Binomial Distribution #' @description Computes the binomial distribution of a certain number of success over a given number of trials #' @param trials Number of trials (numeric) #' @param prob Probability of successs (real) #' @return Returns a data.frame of the number of successes and their probabilities #' @export #' @examples #' x <- bin_distribution(trials = 5, prob = 0.5) #' x #' success probability #' 1 0 0.03125 #' 2 1 0.15625 #' 3 2 0.31250 #' 4 3 0.31250 #' 5 4 0.15625 #' 6 5 0.03125 #' #' plot(x) #' Returns a barplot of the distribution #' bin_distribution <- function(trials, prob) { probabilities <- c() for (i in 0:trials) { probabilities <- c(probabilities, bin_probability(i, trials, prob)) } object <- data.frame(success = 0:trials, probability = probabilities) class(object) <- c("bindis", "data.frame") object } #' @export plot.bindis <- function(distribution) { barplot(distribution$prob, names.arg = distribution$success, xlab = "successes", ylab = "probability") }

adaed9489e39b4101783705bc1de7d20762c378a

41e59bef1fe26f89626e6cba7be15ee98e6d80f8

/R/projection_domain.R

b1e7385f79137362d901e4a8a2988d1fe35d6e5f

[]

no_license

ardeeshany/FLAME

32ae0a694a9c4cc3bd946757c1ccdb15c20c5673

e8ba1670778f32988db6b9f57ede425f9f2d8d2e

refs/heads/master

2021-07-05T03:49:04.885156

2017-09-25T00:10:40

103,167,356

2

0

null

UTF-8

R

false

1,238

r

projection_domain.R

#' Representation on the time domain of a function defined by the coefficients of the kernel basis #' #' It computes the pointwise evaluation of a function (or a set of #' functions) on the time domain, given its (their) projection on the kernel basis. #' #' @param y matrix. \code{J} \eqn{\times} \code{N} matrix containing in #' column \eqn{n} the coefficients of the projection of the function \eqn{y_n} #' on the \code{J} eigenfunctions of the kernel. #' @param eigenfun matrix. \code{m} \eqn{\times} \code{J} matrix containing #' in each column the #' point-wise evaluation of the eigenfunctions on the #' kernel #' #' @return \code{N} \eqn{\times} \code{m} matrix containing in the row \eqn{n} the pointwise #' evaluation of the function \eqn{y_n} on the domain \eqn{D} of length \code{m}. #' @export #' #' @examples #' data(SobolevKernel) #' data(simulation) #' projection_domain(Y_matrix, eigenvect) # projection of the data #' # on the time domain seq(0, 1, length = 50) #' ## @A: Here we do not know what the domain D is, but eigenfun is containing the pointwise evaluation ## of the eigenfunctions of kernel and it is enough. projection_domain <- function(y, eigenfun) { y_mat <- eigenfun %*% y return(t(y_mat)) }

0c69409597e8c0fbb28e953cc5a11065f08deaab

faab8fe6f24c90dff7c6a7eb42b955b073dd7c9f

/TLSfinder_GetTif.R

2699134878afe2a776f9a4e8f436b652776dc930

[]

no_license

Famingzhao/TLS-finder

5d638d575ae76174174f4b2fb5b1265a85d5d44a

d4d221441e667161a8328c333f0fdab397aac1cc

refs/heads/master

2023-08-08T04:03:15.098597

2020-07-13T09:33:58

408,386,511

0

null

UTF-8

R

false

696

r

TLSfinder_GetTif.R

# 3.0_TLSfinder_GetTif.R # Simply takes RData count matrices for B-cells saved with GridQuant and print them as tables with '.tif' extenstion so they can be loaded in Fiji # Author: Daniele Tavernari # Please cite: Tavernari et al., 2021 (see full citation on GitHub repository's README) ########## Input OutDir_matrices = "Grid_matrices/" s = "s8B" grid_spacing = 20 #################### #################### Main ##################### load(file = paste0(OutDir_matrices,s,"_tileSizeMicrons",grid_spacing,"_Bcell_meaMat_and_coos.RData")) write.table(meaMat, file = paste0(OutDir_matrices,s,"_tileSizeMicrons",grid_spacing,"_Bcell_meaMatTab.tif"), row.names = F, col.names = F, sep = "\t")

380ba92fcbaf1176279c539cb13f6072be101f0d

da444894ad9f8181b0f89ac026d8e988779f9850

/Ejercicio2.R

faa934ce45ae8927c99b22c3ba7666194be098f7

[]

no_license

romeoaxpuac123/Practica2Semi2

2ca46d63617c5d5230a1fc2f7d4dc112ee6abf24

c4d711fd9df294e8a87624a519864748c2bfd172

refs/heads/master

2022-11-12T22:50:46.436362

2020-06-28T21:05:06

275,077,449

0

null

UTF-8

R

false

500

r

Ejercicio2.R

print("Romeo Axpuac") library(ggplot2) #Guardamos el path del primer archivo archivo <- "C:\\Users\\Bayyron\\Desktop\\Junio2020\\Seminario2\\Laboratorio\\Practica2\\Archivo\\cardio_train.csv" informacion <- read.csv(archivo) #head(informacion) columnasReporte2 <- c("id","weight") datos <- informacion[columnasReporte2] h= hist(datos$weight, main = "HISTOGRAMA Y FRECUENCIAS DE PESOS", col="red", xlab="PESOS",labels = T) lines(c(0,h$mids),c(0,h$counts), type = "b", pch = 20, col = "blue", lwd = 3)

b691153e616246676424b8ac9452cbf0fa22fc47

622eb3dc154d7779473fc161da613002b3ee877c

/scripts/r/xxtemplate-r.r

1eb2f9e7dd9d9218593601632a3f09b6020fadba

[]

no_license

thereisnotime/xxToolbelt

d9e99e2139da31a4d6e034f4c56753527d120157

2d2a311737d4a910816adc328f6cf4414caed61c

refs/heads/main

2023-04-04T10:35:53.270861

2021-04-05T19:06:50

354,077,716

2

0

null

UTF-8

R

false

66

r

xxtemplate-r.r

#!/usr/bin/env Rscript args <- commandArgs() cat(args, sep = "\n")

19233423044d61fc345497c940acb65d4f411cf7

5d5d7785f5ce2ff377ebec29d74382652502c1d8

/R/calculate_IV.R

0b3baba94526c2d571295a22d541a8328568cc0f

[ "MIT" ]

permissive

standardgalactic/wpa

d7256e719732c7c3f067e88d253e600cd1d66a06

b64b562cee59ea737df58a9cd2b3afaec5d9db64

refs/heads/main

2023-08-10T19:11:03.211088

2021-09-08T13:40:35

null

0

null

UTF-8

R

false

5,905

r

calculate_IV.R

# -------------------------------------------------------------------------------------------- # Copyright (c) Microsoft Corporation. All rights reserved. # Licensed under the MIT License. See LICENSE.txt in the project root for license information. # -------------------------------------------------------------------------------------------- #' @title #' Calculate Weight of Evidence (WOE) and Information Value (IV) between a #' single predictor and a single outcome variable. #' #' @description #' Calculates Weight of Evidence (WOE) and Information Value (IV) between a #' single predictor and a single outcome variable. This function implements the #' common Information Value calculations whilst maintaining the minimum reliance #' on external dependencies. Use `map_IV()` for the equivalent of #' `Information::create_infotables()`, which performs calculations for multiple #' predictors and a single outcome variable. #' #' @details #' The approach used mirrors the one used in `Information::create_infotables()`. #' #' @param data Data frame containing the data. #' @param outcome String containing the name of the outcome variable. #' @param predictor String containing the name of the predictor variable. #' @param bins Numeric value representing the number of bins to use. #' #' @import dplyr #' #' @return A data frame is returned as an output. #' calculate_IV <- function(data, outcome, predictor, bins){ pred_var <- data[[predictor]] outc_var <- data[[outcome]] # Check inputs if(sum(is.na(outc_var)) > 0){ stop( glue::glue( "dependent variable {outcome} has missing values in the input training data frame" ) ) } # Compute q q <- stats::quantile( pred_var, probs = c(1:(bins - 1) / bins), na.rm = TRUE, type = 3 ) # Compute cuts cuts <- unique(q) # Compute intervals intervals <- findInterval( pred_var, vec = cuts, rightmost.closed = FALSE) # Compute cut_table cut_table <- table( intervals, outc_var) %>% as.data.frame.matrix() ## get min/max cut_table_2 <- data.frame( var = pred_var, intervals ) %>% group_by(intervals) %>% summarise( min = min(var, na.rm = TRUE) %>% round(digits = 1), max = max(var, na.rm = TRUE) %>% round(digits = 1), n = n(), .groups = "drop" ) %>% mutate(!!sym(predictor) := glue::glue("[{round(min, digits = 1)},{round(max, digits = 1)}]")) %>% mutate(percentage = n / sum(n)) %>% select(!!sym(predictor), intervals, n, percentage) # Create variables that are double cut_table_1 <- as.numeric(cut_table$`1`) cut_table_0 <- as.numeric(cut_table$`0`) # Non-events in group n_non_event <- cut_table_1 * sum(cut_table_0) # t$y_1*sum_y_0 n_yes_event <- cut_table_0 * sum(cut_table_1) # t$y_0*sum_y_1 # Compute WOE cut_table_2$WOE <- ifelse( cut_table$`1` > 0 & cut_table$`0` > 0, # Both positive log(n_non_event / n_yes_event), # % of non-events divided by % of events 0) # Otherwise impute 0 # Compute IV_weight p1 <- cut_table$`1` / sum(cut_table$`1`) p0 <- cut_table$`0` / sum(cut_table$`0`) cut_table_2$IV_weight <- p1 - p0 cut_table_2$IV <- cut_table_2$WOE * cut_table_2$IV_weight cut_table_2 %>% mutate(IV = cumsum(IV)) %>% # Maintain consistency with `Information::create_infotables()` select( !!sym(predictor), N = "n", Percent = "percentage", WOE, IV) } #' @title #' Calculate Weight of Evidence (WOE) and Information Value (IV) between #' multiple predictors and a single outcome variable, returning a list of #' statistics. #' #' @description #' This is a wrapper around `calculate_IV()` to loop through multiple predictors #' and calculate their Weight of Evidence (WOE) and Information Value (IV) with #' respect to an outcome variable. #' #' @details #' The approach used mirrors the one used in `Information::create_infotables()`. #' #' @param data Data frame containing the data. #' @param outcome String containing the name of the outcome variable. #' @param predictors Character vector containing the names of the predictor #' variables. If `NULL` (default) is supplied, all numeric variables in the #' data will be used. #' @param bins Numeric value representing the number of bins to use. Defaults to #' 10. #' #' @import dplyr #' #' @return A list of data frames is returned as an output. The first layer of #' the list contains `Tables` and `Summary`: #' - `Tables` is a list of data frames containing the WOE and cumulative sum #' IV for each predictor. #' - `Summary` is a single data frame containing the IV for all predictors. #' map_IV <- function(data, predictors = NULL, outcome, bins = 10){ if(is.null(predictors)){ predictors <- data %>% select(-!!sym(outcome)) %>% select( where(is.numeric) ) %>% names() } # List of individual tables Tables <- predictors %>% purrr::map(function(pred){ calculate_IV( data = data, outcome = outcome, predictor = pred, bins = bins ) }) %>% purrr::set_names( nm = purrr::map( ., function(df){ names(df)[[1]] } ) ) # Compile Summary Table Summary <- list("df" = Tables, "names" = names(Tables)) %>% purrr::pmap(function(df, names){ IV_final <- df %>% slice(nrow(df)) %>% pull(IV) data.frame( Variable = names, IV = IV_final ) }) %>% bind_rows() %>% arrange(desc(IV)) # Reorder and combine list c( list("Tables" = Tables[Summary$Variable]), # Reordered list("Summary" = Summary) ) }

979b774695439bbe29c8e18a64d1f0c0dfec2efb

1fcfa19b2fdb270e0862990db0cae8e733c1a7f7

/R/UCTSUpload.R

786b2631b0a9adcf483521e2b6cd53b3c5b5bfff

[]

permissive

GreenGrassBlueOcean/DatastreamDSWS2R

28ae2916239c8f2c7a76a42691cf5cff86494b48

1250c37d360e14ee0beaa709e9033ef363257306

refs/heads/master

2022-12-08T07:45:12.499165

2020-06-03T21:05:51

267,281,708

0

Apache-2.0

2020-05-27T09:50:38

2020-05-27T09:50:37

null

UTF-8

R

false

17,145

r

UCTSUpload.R

#' @include common.R #' @include classConstructor.R #' @include wrapper.R #' #' @name dotEncryptPassword #' @title Encrypt the Datastream password #' @description This is a port of the VBA code #' #' @param strPassword the password to be encrypted #' @return an encrypted password #' #' #' @keywords internal #' .EncryptPassword <- function(strPassword=""){ iSeed <- as.raw(199L) # arbitrary number strCrypted <- "" bBytes <- charToRaw(strPassword) for(b in bBytes) { iCryptedByte <- as.raw(xor(b , iSeed)) strCrypted <- paste0(strCrypted, formatC(as.integer(iCryptedByte),digits=3,width=3,flag="0")) # add previous byte, XOR with arbitrary value iSeed <- xor(as.raw((as.integer(iSeed) + as.integer(iCryptedByte)) %% 255L), as.raw(67L)) } return(strCrypted) } #' @name dotgetTimeseries #' @title convert xts timeseries into a string that can be sent to #' the Datastream server #' #' @param Data the xts timeseries to be converted #' @param freq the frequency of the data #' @param digits the number of decimal places to round the data to #' @param NA_VALUE the string to replace NA data with #' #' @return A string of the core data of Data #' #' #' @importFrom zoo zoo index #' @importFrom xts merge.xts .indexwday #' @importFrom stringr str_trim #' @keywords internal #' .getTimeseries <- function(Data, freq, digits, NA_VALUE){ if(ncol(Data) > 1) { # Make sure we are only dealing with a single column xts Data <- Data[,1] } if (freq == "D") { # We have a daily frequency, which means we need to do more work matching up the dates as # Datastream assumes that they are in weekday order. The loaded timeseries might have gaps or weekend # measures # the xts .indexwday gives the day of the week with 0=Sunday and 6=Saturday # We need to make sure there are no blanks in the data startDate <- zoo::index(first(Data)) endDate <- zoo::index(last(Data)) NADates <- seq(from=startDate, to=endDate, by="days") NAData <- zoo(c(NA), order.by=NADates) #merge and fill missing rows with NAs wData <- xts::merge.xts(Data, NAData, fill=NA) # This only picks the weeksdays from the original series wData <- wData[which(xts::.indexwday(wData) %in% 1:5),1] }else{ wData <- Data #If we do not have a daily frequency then we can just load up the datapoints, with the implicit #assumption that they are in the right frequency } sFormattedData <- suppressWarnings(formatC(wData, digits = digits, mode="double", format="f")) sFormattedData <- stringr::str_trim(sFormattedData) #We need to make sure that any missing data is replaced with the # the correct symbol sFormattedData[which(sFormattedData=="NaN")] <- NA_VALUE #Collapse the array into a string sData<-paste0(sFormattedData,collapse=",") sData<-paste0(sData,",") return(sData) } #' @title Upload a UCTS timeseries into Datastream #' #' @description Uploads an xts into a UCTS in the Datastream Database #' @details Note this function does not check to see if there is #' a pre-existing timeseries already in Datastream. It will just overwrite #' any existing UCTS. #' @param tsData - an xts (or timeseries object that can be converted to #' one) to be uploaded. #' @param TSCode The mnemonic of the target UCTS #' @param MGMTGroup Must have managment group. Only the first #' characters will be used. #' @param freq The frequency of the data to be uploaded #' @param seriesName the name of the series #' @param Units Units of the data - can be no more than 12 characters - #' excess will be trimmed to that length #' @param Decimals Number of Decimals in the data - a number between 0 and #' 9 - if outside that range then trimmed #' @param ActPer Whether the values are percentages ("N") or actual #' numbers ("Y") #' @param freqConversion How to do any FX conversions #' @param Alignment Alignment of the data within periods #' @param Carry whether to carry data over missing dates #' @param PrimeCurr the currency of the timeseries #' @param strUsername your Datastream username #' @param strPassword your Datastream Password #' @param strServerName URL of the Datastream server #' @param strServerPage page on the datastream server #' @return TRUE if the upload has been a success, otherwise an error message #' #' @export #' #' @importFrom zoo index #' @importFrom httr POST add_headers content content_type #' @importFrom xts as.xts first last xtsible #' UCTSUpload <- function(tsData, TSCode="", MGMTGroup="ABC", freq = c("D","W","M","Q","Y"), seriesName, Units="", Decimals=2, ActPer=c("N","Y"), freqConversion= c("ACT","SUM","AVG","END"), Alignment=c("1ST","MID","END"), Carry=c("YES","NO","PAD"), PrimeCurr="", strUsername = ifelse(Sys.getenv("DatastreamUsername") != "", Sys.getenv("DatastreamUsername"), options()$Datastream.Username), strPassword = ifelse(Sys.getenv("DatastreamPassword") != "", Sys.getenv("DatastreamPassword"), options()$Datastream.Password), strServerName="http://product.datastream.com", strServerPage="/UCTS/UCTSMaint.asp"){ #Check inputs are valid if(!xtsible(tsData)){ stop(paste0("tsData must be a time-based object and not of class ",class(tsData))) } if(!freq[1] %in% c("D","W","M","Q","Y")){ stop("freq is not an allowed value") } if(!ActPer[1] %in% c("N","Y")){ stop("ActPer is not an allowed value") } if(!freqConversion[1] %in% c("ACT","SUM","AVG","END")){ stop("freqConversion is not an allowed value") } if(!Alignment[1] %in% c("1ST","MID","END")){ stop("Alignment is not an allowed value") } if(!Carry[1] %in% c("YES","NO","PAD")){ stop("Carry is not an allowed value") } # Limit decimals a number in range to the range 0-9 if(!is.numeric(Decimals)) Decimals <- 2L Decimals <- as.integer(Decimals) if(Decimals < 0) Decimals <- 0 if(Decimals > 9) Decimals <- 9 # Trim any excess for units Units <- substr(Units,0,12) # Replace any ISO currency codes with DS codes if(is.null(PrimeCurr)) { PrimeCurr <- "" } if(nchar(PrimeCurr) > 3){ stop("Invalid currency. Should be either 3 digit ISO code or Datastream code") } else if(nchar(PrimeCurr) == 3 ){ # Check ISO code is valid and convert to DS Code dfXRef <- DatastreamDSWS2R::currencyDS2ISO if(PrimeCurr %in% dfXRef$isoCode){ PrimeCurr <- dfXRef$dsCode[which(PrimeCurr == dfXRef$isoCode & dfXRef$primeCode == TRUE)] } else { stop("Invalid currency. Should be an ISO code in table currencyDS2ISO.") } } else if(nchar(PrimeCurr) > 0 ){ # Check DS Code is valid PrimeCurr <- iconv(PrimeCurr, from="utf-8", to = "latin1") dfXRef <- DatastreamDSWS2R::currencyDS2ISO if(!PrimeCurr %in% dfXRef$dsCode){ stop("Invalid currency. Should be an Datastream code in table currencyDS2ISO.") } } # At the moment everything will be a full update, and a hard coded NA value NA_VALUE <- "NA" # Add Start Date for values - make sure it is in DD/MM/YY format #CMC actually the function returns a dd/MM/yyyy format post Y2K # convert to xts object myXtsData <- as.xts(tsData) startDate <- zoo::index(first(myXtsData)) endDate <- zoo::index(last(myXtsData)) # Now create the URL to post the form to dsURL <- paste0(strServerName , strServerPage , "?UserID=" , strUsername) # Create a list of the parameters to be uploaded # We have not included the pair AmendFlag="Y", so all these will be full updates dsParams <- list(CallType = "Upload", TSMnemonic = toupper(TSCode), TSMLM = toupper(MGMTGroup), TSStartDate = format(startDate,format="%d/%m/%Y"), TSEndDate = format(endDate,format="%d/%m/%Y"), TSFrequency = freq[1], TSTitle = seriesName, TSUnits = Units, TSDecPlaces = Decimals, TSAsPerc = ActPer[1], TSFreqConv = freqConversion[1], # Add "Frequency Conversion" TSAlignment = Alignment[1], # Add "Alignment" TSCarryInd = Carry[1], # Add "Carry Indicator" TSPrimeCurr = I(PrimeCurr), # Add "Prime Currency" TSULCurr = "", # no longer use Underlying Currency, but need to pass up a null value as the mainframe is expecting it ForceUpdateFlag1 = "Y", ForceUpdateFlag2 = "Y", # We have ignored some logic in the original UCTS VBA code # AmendFlag = "Y", TSValsStart = format(startDate,format="%d/%m/%Y"), #TODO adjust this date according to the frequency of the data VBA function AdjustDateTo1st NAValue = NA_VALUE, TSValues = .getTimeseries(myXtsData, freq= freq[1], digits=Decimals, NA_VALUE), #Now add the datapoints - the date element of the series is discarded here, with obvious risks UserOption = .EncryptPassword(strPassword) ) # Now post the form # We will give it three tries nLoop <- 1 waitTimeBase <- 2 maxLoop <- 4 retValue <- "" while(nLoop < maxLoop){ retValue <- tryCatch(httr::POST(url = dsURL, body = dsParams, config = httr::add_headers(encoding = "utf-8"), httr::content_type("application/x-www-form-urlencoded; charset=utf-8"), encode = "form"), error = function(e) e) # Break if an error or null if(is.null(retValue)) break if("error" %in% class(retValue)) break # If did not get a time out then break if(httr::status_code(retValue) != 408) break # If not succesful then wait 2 seconds before re-submitting, ie give time for the # server/network to recover. Sys.sleep(waitTimeBase ^ nLoop) nLoop <- nLoop + 1 } if(is.null(retValue)){ return(structure(FALSE, error = "NULL value returned")) } if("error" %in% class(retValue)){ return(structure(FALSE, error = paste("Error ", retValue$message))) } if(httr::http_error(retValue)){ return(structure(FALSE, error = paste("http Error: ", paste0(httr::http_status(retValue, collapse = " : "))))) } myResponse <- content(retValue, as = "text") if(myResponse[1] == "*OK*"){ return(structure(TRUE, error = "")) } else{ return(structure(FALSE, error = paste("*Error* Upload failed after ", nLoop, " attempts with error ", myResponse[1]))) } } #' @title Append a xts to an existing UCTS timeseries in Datastream #' #' @description Uploads and appends an xts into a UCTS in the Datastream Database #' @details This function checks if there is a pre-existing timeseries already in Datastream. #' If there is then it will append the xts onto the existing series. If there are any #' overlapping dates then depending on the setting of overwrite then the new data #' will overwrite the existing data in the UCTS #' #' @param tsData - an xts (or timeseries object that can be converted to #' one) to be uploaded. #' @param TSCode The mnemonic of the target UCTS #' @param MGMTGroup Must have managment group. Only the first #' characters will be used. #' @param freq The frequency of the data to be uploaded #' @param seriesName the name of the series #' @param Units Units of the data - can be no more than 12 characters - #' excess will be trimmed to that length #' @param Decimals Number of Decimals in the data - a number between 0 and #' 9 - if outside that range then trimmed #' @param ActPer Whether the values are percentages ("N") or actual #' numbers ("Y") #' @param freqConversion How to do any FX conversions #' @param Alignment Alignment of the data within periods #' @param Carry whether to carry data over missing dates #' @param PrimeCurr the currency of the timeseries #' @param overwrite if TRUE then existing data in the UCTS will be overwritten #' @param strUsername your Datastream username #' @param strPassword your Datastream Password #' @param strServerName URL of the Datastream server #' @param strServerPage page on the datastream server #' @return TRUE if the upload has been a success, otherwise an error message #' #' @export #' #' @importFrom zoo index #' @importFrom xts as.xts first last xtsible #' UCTSAppend <- function(tsData, TSCode = "", MGMTGroup = "ABC", freq = c("D","W","M","Q","Y"), seriesName, Units = "", Decimals = 2, ActPer = c("N","Y"), freqConversion = c("ACT","SUM","AVG","END"), Alignment = c("1ST","MID","END"), Carry = c("YES","NO","PAD"), PrimeCurr ="", overwrite = TRUE, strUsername = ifelse(Sys.getenv("DatastreamUsername") != "", Sys.getenv("DatastreamUsername"), options()$Datastream.Username), strPassword = ifelse(Sys.getenv("DatastreamPassword") != "", Sys.getenv("DatastreamPassword"), options()$Datastream.Password), strServerName = "http://product.datastream.com", strServerPage = "/UCTS/UCTSMaint.asp"){ #Check inputs are valid - we can also rely on checks in UCTSUpload later if(!xtsible(tsData)){ stop(paste0("tsData must be a time-based object and not of class ", class(tsData))) } tsData <- as.xts(tsData) if(!freq[1] %in% c("D","W","M","Q","Y")){ stop("freq is not an allowed value") } # Get the existing UCTS from Datastream mydsws <- dsws$new() tsExisting <- mydsws$timeSeriesRequest(instrument = TSCode, startDate = as.Date("1950-01-01"), endDate = index(last(tsData)), frequency = freq) if(is.null(tsExisting)){ errMsg <- paste0("Datastream Server Error retrieving existing series\n", paste(mydsws$errorlist, collapse = "\n", sep = "\n")) stop(errMsg) } # In the absence of being able to define start and end dates for UCTS as defined # on http://product.datastream.com/DSWSClient/Docs/SoapApiHelp/EnumDetails.html#DSDateNames # We are going to trim the start and end of the series of any null values # If this is fixed by Datastream or another way is suggested then these lines # could be removed validRows <- which(!is.na(tsExisting)) # Check if any data was found if(length(validRows) != 0){ # There was no existing timeseries # Take the non-null middle segment firstNotNULL <- min(validRows) lastNotNULL <- max(validRows) tsExisting <- tsExisting[firstNotNULL:lastNotNULL, ] # Combine the new data with the existing data if(overwrite){ # append with new data overwriting the old tsData <- xts::make.index.unique(rbind(tsData, tsExisting), drop = TRUE) } else { # append with old data being kept tsData <- xts::make.index.unique(rbind(tsExisting, tsData), drop = TRUE) } } # Upload combined timeseries return(UCTSUpload(tsData = tsData, TSCode = TSCode, MGMTGroup = MGMTGroup, freq = freq, seriesName = seriesName, Units = Units, Decimals = Decimals, ActPer = ActPer, freqConversion = freqConversion, Alignment = Alignment, Carry = Carry, PrimeCurr = PrimeCurr, strUsername = strUsername, strPassword = strPassword, strServerName = strServerName, strServerPage = strServerPage)) }