Clusterviz

R/clusterviz.R

@@ -5,36 +5,35 @@ parser <- ArgumentParser(description="Create plots for aggregated cluster data")
 parser$add_argument("cluster_file", help = "file to visualise")
 parser$add_argument("-i", "--img-path", dest = "img_path", default = "../img/", help = "path to store plots in; default: ../img/")
 parser$add_argument("-p", "--postfix", dest = "postfix", default = "_plot", help = "postfix for files, default: _plot")
-parser$add_argument("-v", "--virtualenv", dest = "virtualenv", default = "../venv/", help = "path of virtualenv; default: ../venv/; '-' for none")
 parser$add_argument("--width", dest = "width", default = 40, help = "width (cm), default 40", type = "double")
 parser$add_argument("--height", dest = "height", default = 25, help = "height (cm), default 25", type = "double")
+parser$add_argument("--pkl", dest = "csv", help = "load from pickle instead of csv", action = "store_false")
 
 args <- parser$parse_args()
 
-library(reticulate)
-library(ggplot2)
-library(dplyr)
-library(tidyr)
-library(TSA)
-library(forecast)
+print(args)
+
+library(ggplot2, warn.conflicts = FALSE, quietly = TRUE)
+library(dplyr, warn.conflicts = FALSE, quietly = TRUE)
+library(tidyr, warn.conflicts = FALSE, quietly = TRUE)
+library(TSA, warn.conflicts = FALSE, quietly = TRUE)
+library(forecast, warn.conflicts = FALSE, quietly = TRUE)
 theme_set(theme_bw())
 
-if (args$virtualenv != "-") {
-    print(args$virtualenv)
-    use_virtualenv(args$virtualenv)
+if (args$csv) {
+    aggdf <- read.csv(args$cluster_file, header = TRUE, stringsAsFactors = FALSE)
+    aggdf$read_time <- as.POSIXct(aggdf$read_time)
+} else {
+    library(reticulate, warn.conflicts = FALSE, quietly = TRUE)
+    p <- import("pandas")
+    aggdf <- p$read_pickle(args$cluster_file)
 }
-
-p <- import("pandas")
-aggdf <- p$read_pickle(args$cluster_file)
 aggdf$cluster <- factor(aggdf$cluster)
-str(aggdf)
 clusters = levels(aggdf$cluster)
 cbp <- c('#0173b2', '#de8f05', '#029e73', '#d55e00', '#cc78bc', '#ca9161', '#fbafe4', '#949494', '#ece133', '#56b4e9', '#0173b2', 
          '#de8f05', '#029e73', '#d55e00', '#cc78bc', '#ca9161', '#fbafe4', '#949494', '#ece133', '#56b4e9', '#0173b2', '#de8f05', 
          '#029e73', '#d55e00', '#cc78bc', '#ca9161', '#fbafe4', '#949494', '#ece133', '#56b4e9', '#0173b2', '#de8f05', '#029e73', 
-         '#d55e00', '#cc78bc', '#ca9161', '#fbafe4', '#949494', '#ece133', '#56b4e9')[1:length(clusters)
-
-ggplot(aggdf, aes(y = kwh_tot_mean, x = cluster)) + geom_boxplot()
+         '#d55e00', '#cc78bc', '#ca9161', '#fbafe4', '#949494', '#ece133', '#56b4e9')[1:length(clusters)]
 
 facall <- ggplot(aggdf, aes(x = read_time, y = kwh_tot_mean, color = cluster, fill = cluster)) + 
     geom_line(size = 1.5) + geom_ribbon(aes(ymin = kwh_tot_CI_low, ymax = kwh_tot_CI_high), alpha = 0.2, color = NA) +

README.md

@@ -202,3 +202,8 @@ To install the packages needed for these scripts you should run the following fr
 ```R
 install.packages(c("dplyr", "tidyr", "ggplot2", "forecast", "TSA", "reticulate", "caTools", "scales", "argparse"))
 ```
+
+### `clusterviz.R`
+
+This script visualises the patterns of each cluster in an aggregated file.
+