Revision e22c2d71
Added by Benoit Parmentier almost 9 years ago
| climate/research/oregon/interpolation/global_run_scalingup_assessment_part2.R | ||
|---|---|---|
| 5 | 5 |
#Analyses, figures, tables and data are also produced in the script. |
| 6 | 6 |
#AUTHOR: Benoit Parmentier |
| 7 | 7 |
#CREATED ON: 03/23/2014 |
| 8 |
#MODIFIED ON: 01/02/2016
|
|
| 9 |
#Version: 4
|
|
| 8 |
#MODIFIED ON: 01/03/2016
|
|
| 9 |
#Version: 5
|
|
| 10 | 10 |
#PROJECT: Environmental Layers project |
| 11 | 11 |
#COMMENTS: analyses for run 10 global analyses,all regions 1500x4500km with additional tiles to increase overlap |
| 12 | 12 |
#TODO: |
| ... | ... | |
| 16 | 16 |
|
| 17 | 17 |
################################################################################################# |
| 18 | 18 |
|
| 19 |
|
|
| 20 |
|
|
| 21 | 19 |
#### FUNCTION USED IN SCRIPT |
| 22 | 20 |
|
| 23 | 21 |
#function_analyses_paper1 <-"contribution_of_covariates_paper_interpolation_functions_07182014.R" #first interp paper |
| 24 | 22 |
#function_analyses_paper2 <-"multi_timescales_paper_interpolation_functions_08132014.R" |
| 25 | 23 |
#function_global_run_assessment_part2 <- "global_run_scalingup_assessment_part2_functions_0923015.R" |
| 26 | 24 |
|
| 27 |
|
|
| 28 |
|
|
| 29 |
|
|
| 30 | 25 |
############################################ |
| 31 | 26 |
#### Parameters and constants |
| 32 | 27 |
|
| ... | ... | |
| 45 | 40 |
#parent output dir for the current script analyes |
| 46 | 41 |
#out_dir <- "/nobackup/bparmen1/" #on NEX |
| 47 | 42 |
#in_dir_shp <- "/nobackupp4/aguzman4/climateLayers/output4/subset/shapefiles/" |
| 48 |
in_dir <- "" #PARAM 0 |
|
| 49 |
y_var_name <- "dailyTmax" #PARAM1 |
|
| 50 |
interpolation_method <- c("gam_CAI") #PARAM2
|
|
| 43 |
#in_dir <- "" #PARAM 0
|
|
| 44 |
#y_var_name <- "dailyTmax" #PARAM1
|
|
| 45 |
#interpolation_method <- c("gam_CAI") #PARAM2
|
|
| 51 | 46 |
#out_suffix<-"run10_global_analyses_01282015" |
| 52 | 47 |
#out_suffix <- "output_run10_1000x3000_global_analyses_02102015" |
| 53 |
out_suffix <- "run10_1500x4500_global_analyses_pred_1992_10052015" #PARAM3 |
|
| 54 |
out_dir <- "/data/project/layers/commons/NEX_data/output_run10_1500x4500_global_analyses_pred_1992_10052015" #PARAM4 |
|
| 55 |
create_out_dir_param <- FALSE #PARAM 5 |
|
| 56 |
mosaic_plot <- FALSE #PARAM6 |
|
| 48 |
#out_suffix <- "run10_1500x4500_global_analyses_pred_1992_10052015" #PARAM3
|
|
| 49 |
#out_dir <- "/data/project/layers/commons/NEX_data/output_run10_1500x4500_global_analyses_pred_1992_10052015" #PARAM4
|
|
| 50 |
#create_out_dir_param <- FALSE #PARAM 5
|
|
| 51 |
#mosaic_plot <- FALSE #PARAM6
|
|
| 57 | 52 |
#if daily mosaics NULL then mosaicas all days of the year |
| 58 | 53 |
#day_to_mosaic <- c("19920101","19920102","19920103") #PARAM7
|
| 59 | 54 |
#CRS_WGS84 <- CRS("+proj=longlat +ellps=WGS84 +datum=WGS84 +towgs84=0,0,0") #Station coords WGS84 #CONSTANT1
|
| 60 |
CRS_locs_WGS84<-CRS("+proj=longlat +ellps=WGS84 +datum=WGS84 +towgs84=0,0,0") #Station coords WGS84
|
|
| 61 |
proj_str<- CRS_WGS84 #PARAM 8 #check this parameter |
|
| 62 |
file_format <- ".rst" #PARAM 9 |
|
| 63 |
NA_value <- -9999 #PARAM10 |
|
| 64 |
NA_flag_val <- NA_value |
|
| 65 |
multiple_region <- TRUE #PARAM 12 |
|
| 55 |
#CRS_locs_WGS84<-CRS("+proj=longlat +ellps=WGS84 +datum=WGS84 +towgs84=0,0,0") #Station coords WGS84
|
|
| 56 |
#proj_str<- CRS_WGS84 #PARAM 8 #check this parameter
|
|
| 57 |
#file_format <- ".rst" #PARAM 9
|
|
| 58 |
#NA_value <- -9999 #PARAM10
|
|
| 59 |
#NA_flag_val <- NA_value
|
|
| 60 |
#multiple_region <- TRUE #PARAM 12
|
|
| 66 | 61 |
#region_name <- "world" #PARAM 13 |
| 67 |
countries_shp <-"/data/project/layers/commons/NEX_data/countries.shp" #PARAM 13, copy this on NEX too |
|
| 68 |
plot_region <- TRUE |
|
| 69 |
num_cores <- 6 #PARAM 14 |
|
| 70 |
region_name <- c("reg4") #reference region to merge if necessary, if world all the regions are together #PARAM 16
|
|
| 62 |
#countries_shp <-"/data/project/layers/commons/NEX_data/countries.shp" #PARAM 13, copy this on NEX too
|
|
| 63 |
#plot_region <- TRUE
|
|
| 64 |
#num_cores <- 6 #PARAM 14
|
|
| 65 |
#region_name <- c("reg4") #reference region to merge if necessary, if world all the regions are together #PARAM 16
|
|
| 71 | 66 |
#use previous files produced in step 1a and stored in a data.frame |
| 72 |
df_assessment_files <- "df_assessment_files_reg4_1984_run_global_analyses_pred_12282015.txt" #PARAM 17 |
|
| 73 |
threshold_missing_day <- c(367,365,300,200) #PARM18 |
|
| 67 |
#df_assessment_files <- "df_assessment_files_reg4_1984_run_global_analyses_pred_12282015.txt" #PARAM 17
|
|
| 68 |
#threshold_missing_day <- c(367,365,300,200) #PARM18
|
|
| 74 | 69 |
|
| 75 |
list_param_run_assessment_plottingin_dir <- list(y_var_name, interpolation_method, out_suffix,
|
|
| 76 |
out_dir, create_out_dir_param, mosaic_plot, proj_str, file_format, NA_value, |
|
| 77 |
multiple_region, countries_shp, plot_region, num_cores, |
|
| 78 |
region_name, df_assessment_files, threshold_missing_day) |
|
| 70 |
#list_param_run_assessment_plottingin_dir <- list(in_dir,y_var_name, interpolation_method, out_suffix,
|
|
| 71 |
# out_dir, create_out_dir_param, mosaic_plot, proj_str, file_format, NA_value,
|
|
| 72 |
# multiple_region, countries_shp, plot_region, num_cores,
|
|
| 73 |
# region_name, df_assessment_files, threshold_missing_day)
|
|
| 79 | 74 |
|
| 80 |
names(list_param_run_assessment_plottingin_dir) <- c("y_var_name","interpolation_method","out_suffix",
|
|
| 81 |
"out_dir","create_out_dir_param","mosaic_plot","proj_str","file_format","NA_value", |
|
| 82 |
"multiple_region","countries_shp","plot_region","num_cores", |
|
| 83 |
"region_name","df_assessment_files","threshold_missing_day") |
|
| 75 |
#names(list_param_run_assessment_plottingin_dir) <- c("in_dir","y_var_name","interpolation_method","out_suffix",
|
|
| 76 |
# "out_dir","create_out_dir_param","mosaic_plot","proj_str","file_format","NA_value",
|
|
| 77 |
# "multiple_region","countries_shp","plot_region","num_cores",
|
|
| 78 |
# "region_name","df_assessment_files","threshold_missing_day")
|
|
| 84 | 79 |
|
| 85 |
run_assessment_plotting_prediction_fun(list_param_run_assessment_plottingin_dir) |
|
| 80 |
#run_assessment_plotting_prediction_fun(list_param_run_assessment_plottingin_dir)
|
|
| 86 | 81 |
|
| 87 | 82 |
run_assessment_plotting_prediction_fun <-function(list_param_run_assessment_plotting){
|
| 88 | 83 |
|
| ... | ... | |
| 101 | 96 |
#12) countries_shp #<- "world" #PARAM 13 |
| 102 | 97 |
#13) plot_region #<- TRUE |
| 103 | 98 |
#14) num_cores <- number of cores used # 6 #PARAM 14 |
| 104 |
#16) region_name #<- c("reg4"), world if full assessment #reference region to merge if necessary #PARAM 16
|
|
| 105 |
#18) df_assessment_files #PARAM 16
|
|
| 106 |
#19) threshold_missing_day #PARM18
|
|
| 99 |
#15) region_name #<- c("reg4"), world if full assessment #reference region to merge if necessary #PARAM 16
|
|
| 100 |
#16) df_assessment_files #PARAM 16
|
|
| 101 |
#17) threshold_missing_day #PARM18
|
|
| 107 | 102 |
# |
| 108 | 103 |
|
| 109 | 104 |
### Loading R library and packages |
| ... | ... | |
| 160 | 155 |
|
| 161 | 156 |
####### PARSE INPUT ARGUMENTS/PARAMETERS ##### |
| 162 | 157 |
|
| 163 |
in_dir <- list_param_run_assessment_plotting$in_dir #PARAM 0
|
|
| 164 |
y_var_name <- list_param_run_assessment_plotting$y_var_name #PARAM1
|
|
| 165 |
interpolation_method <- list_param_run_assessment_plotting$interpolation_method #c("gam_CAI") #PARAM2
|
|
| 166 |
out_suffix <- list_param_run_assessment_plotting$out_suffix #PARAM3
|
|
| 167 |
out_dir <- list_param_run_assessment_plotting$out_dir # |
|
| 168 |
create_out_dir_param <- list_param_run_assessment_plotting$create_out_dir_param # FALSE #PARAM 5
|
|
| 169 |
mosaic_plot <- list_param_run_assessment_plotting$mosaic_plot #FALSE #PARAM6
|
|
| 158 |
in_dir <- list_param_run_assessment_plotting$in_dir #PARAM 1
|
|
| 159 |
y_var_name <- list_param_run_assessment_plotting$y_var_name #PARAM2
|
|
| 160 |
interpolation_method <- list_param_run_assessment_plotting$interpolation_method #c("gam_CAI") #PARAM3
|
|
| 161 |
out_suffix <- list_param_run_assessment_plotting$out_suffix #PARAM4
|
|
| 162 |
out_dir <- list_param_run_assessment_plotting$out_dir # PARAM5
|
|
| 163 |
create_out_dir_param <- list_param_run_assessment_plotting$create_out_dir_param # FALSE #PARAM 6
|
|
| 164 |
mosaic_plot <- list_param_run_assessment_plotting$mosaic_plot #FALSE #PARAM7
|
|
| 170 | 165 |
proj_str<- list_param_run_assessment_plotting$proj_str #CRS_WGS84 #PARAM 8 #check this parameter |
| 171 | 166 |
file_format <- list_param_run_assessment_plotting$file_format #".rst" #PARAM 9 |
| 172 | 167 |
NA_value <- list_param_run_assessment_plotting$NA_value #-9999 #PARAM10 |
| 173 |
multiple_region <- list_param_run_assessment_plotting$multiple_region # <- TRUE #PARAM 12
|
|
| 174 |
countries_shp <- list_param_run_assessment_plotting$countries_shp #<- "world" #PARAM 13
|
|
| 175 |
plot_region <- list_param_run_assessment_plotting$plot_region #<- TRUE
|
|
| 168 |
multiple_region <- list_param_run_assessment_plotting$multiple_region # <- TRUE #PARAM 11
|
|
| 169 |
countries_shp <- list_param_run_assessment_plotting$countries_shp #<- "world" #PARAM 12
|
|
| 170 |
plot_region <- list_param_run_assessment_plotting$plot_region # PARAM13
|
|
| 176 | 171 |
num_cores <- list_param_run_assessment_plotting$num_cores # 6 #PARAM 14 |
| 177 |
region_name <- list_param_run_assessment_plotting$region_name #<- "world" #PARAM 13
|
|
| 178 |
df_assessment_files <- list_param_run_assessment_plotting$df_assessment_files #PARAM 16
|
|
| 179 |
threshold_missing_day <- list_param_run_assessment_plotting$threshold_missing_day #PARM18
|
|
| 172 |
region_name <- list_param_run_assessment_plotting$region_name #<- "world" #PARAM 15
|
|
| 173 |
df_assessment_files_name <- list_param_run_assessment_plotting$df_assessment_files_name #PARAM 16
|
|
| 174 |
threshold_missing_day <- list_param_run_assessment_plotting$threshold_missing_day #PARM17
|
|
| 180 | 175 |
|
| 181 | 176 |
NA_flag_val <- NA_value |
| 182 | 177 |
|
| ... | ... | |
| 193 | 188 |
|
| 194 | 189 |
setwd(out_dir) |
| 195 | 190 |
|
| 191 |
list_outfiles <- vector("list", length=20) #collect names of output files
|
|
| 192 |
list_outfiles_names <- vector("list", length=20) #collect names of output files
|
|
| 193 |
counter_fig <- 0 #index of figure to collect outputs |
|
| 194 |
|
|
| 196 | 195 |
#i <- year_predicted |
| 197 | 196 |
###Table 1: Average accuracy metrics |
| 198 | 197 |
###Table 2: daily accuracy metrics for all tiles |
| 199 | 198 |
|
| 199 |
df_assessment_files <- read.table(df_assessment_files_name,stringsAsFactors=F,sep=",") |
|
| 200 | 200 |
#df_assessment_files, note that in_dir indicate the path of the textfiles |
| 201 |
summary_metrics_v <- file.path(in_dir,basename(df_assessment_files$files[1]))
|
|
| 202 |
tb <- file.path(in_dir, basename(df_assessment_files$files[2]))
|
|
| 203 |
tb_s <-file.path(in_dir, basename(df_assessment_files$files[4]))
|
|
| 201 |
summary_metrics_v <- read.table(file.path(in_dir,basename(df_assessment_files$files[1])),sep=",")
|
|
| 202 |
tb <- read.table(file.path(in_dir, basename(df_assessment_files$files[2])),sep=",")
|
|
| 203 |
tb_s <- read.table(file.path(in_dir, basename(df_assessment_files$files[4])),sep=",")
|
|
| 204 | 204 |
|
| 205 |
tb_month_s <- file.path(in_dir,basename(df_assessment_files$files[3]))
|
|
| 206 |
pred_data_month_info <- file.path(in_dir, basename(df_assessment_files$files[10]))
|
|
| 207 |
pred_data_day_info <- file.path(in_dir, basename(df_assessment_files$files[11]))
|
|
| 208 |
df_tile_processed <- file.path(in_dir, basename(df_assessment_files$files[12]))
|
|
| 205 |
tb_month_s <- read.table(file.path(in_dir,basename(df_assessment_files$files[3])),sep=",")
|
|
| 206 |
pred_data_month_info <- read.table(file.path(in_dir, basename(df_assessment_files$files[10])),sep=",")
|
|
| 207 |
pred_data_day_info <- read.table(file.path(in_dir, basename(df_assessment_files$files[11])),sep=",")
|
|
| 208 |
df_tile_processed <- read.table(file.path(in_dir, basename(df_assessment_files$files[12])),sep=",")
|
|
| 209 | 209 |
|
| 210 | 210 |
#add column for tile size later on!!! |
| 211 | 211 |
|
| ... | ... | |
| 219 | 219 |
tb_s$pred_mod <- as.character(tb_s$pred_mod) |
| 220 | 220 |
tb_s$tile_id <- as.character(tb_s$tile_id) |
| 221 | 221 |
|
| 222 |
#multiple regions? |
|
| 222 |
#multiple regions? #this needs to be included in the previous script!!!
|
|
| 223 | 223 |
if(multiple_region==TRUE){
|
| 224 | 224 |
df_tile_processed$reg <- basename(dirname(as.character(df_tile_processed$path_NEX))) |
| 225 | 225 |
tb <- merge(tb,df_tile_processed,by="tile_id") |
| ... | ... | |
| 331 | 331 |
#unique(summaty_metrics$tile_id) |
| 332 | 332 |
#text(lat-shp,) |
| 333 | 333 |
#union(list_shp_reg_files[[1]],list_shp_reg_files[[2]]) |
| 334 |
list_outfiles[[counter_fig+1]] <- paste("Figure1_tile_processed_region_",region_name,"_",out_suffix,".png",sep="")
|
|
| 334 | 335 |
|
| 335 | 336 |
############### |
| 336 | 337 |
### Figure 2: boxplot of average accuracy by model and by tiles |
| 337 | 338 |
|
| 338 |
|
|
| 339 | 339 |
tb$tile_id <- factor(tb$tile_id, levels=unique(tb$tile_id)) |
| 340 | 340 |
model_name <- as.character(unique(tb$pred_mod)) |
| 341 | 341 |
|
| ... | ... | |
| 345 | 345 |
res_pix <- 480 |
| 346 | 346 |
col_mfrow <- 1 |
| 347 | 347 |
row_mfrow <- 1 |
| 348 |
|
|
| 348 |
fig_name <- paste("Figure2a_boxplot_with_oultiers_by_tiles_",model_name[i],"_",out_suffix,".png",sep="")
|
|
| 349 | 349 |
png(filename=paste("Figure2a_boxplot_with_oultiers_by_tiles_",model_name[i],"_",out_suffix,".png",sep=""),
|
| 350 | 350 |
width=col_mfrow*res_pix,height=row_mfrow*res_pix) |
| 351 | 351 |
|
| ... | ... | |
| 353 | 353 |
title(paste("RMSE per ",model_name[i]))
|
| 354 | 354 |
|
| 355 | 355 |
dev.off() |
| 356 |
list_outfiles[[counter_fig+i]] <- fig_filename |
|
| 356 | 357 |
} |
| 357 | 358 |
|
| 358 | 359 |
## Figure 2b |
| 359 | 360 |
#with ylim and removing trailing... |
| 360 |
for(i in 1:length(model_name)){
|
|
| 361 |
for(i in 1:length(model_name)){ #there are two models!!
|
|
| 361 | 362 |
|
| 362 | 363 |
res_pix <- 480 |
| 363 | 364 |
col_mfrow <- 1 |
| 364 | 365 |
row_mfrow <- 1 |
| 366 |
fig_filename <- paste("Figure2b_boxplot_scaling_by_tiles","_",model_name[i],"_",out_suffix,".png",sep="")
|
|
| 365 | 367 |
png(filename=paste("Figure2b_boxplot_scaling_by_tiles","_",model_name[i],"_",out_suffix,".png",sep=""),
|
| 366 | 368 |
width=col_mfrow*res_pix,height=row_mfrow*res_pix) |
| 367 | 369 |
|
| ... | ... | |
| 370 | 372 |
,ylim=c(0,4),outline=FALSE) |
| 371 | 373 |
title(paste("RMSE per ",model_name[i]))
|
| 372 | 374 |
dev.off() |
| 375 |
#we already stored one figure |
|
| 376 |
list_outfiles[[counter_fig+i]] <- fig_filename |
|
| 373 | 377 |
} |
| 378 |
counter_fig <- counter_fig + length(model_name) |
|
| 374 | 379 |
#bwplot(rmse~tile_id, data=subset(tb,tb$pred_mod=="mod1")) |
| 375 |
|
|
| 380 |
|
|
| 376 | 381 |
############### |
| 377 | 382 |
### Figure 3: boxplot of average RMSE by model acrosss all tiles |
| 378 | 383 |
|
| ... | ... | |
| 386 | 391 |
|
| 387 | 392 |
boxplot(rmse~pred_mod,data=tb)#,names=tb$pred_mod) |
| 388 | 393 |
title("RMSE per model over all tiles")
|
| 389 |
|
|
| 390 | 394 |
dev.off() |
| 391 |
|
|
| 395 |
list_outfiles[[counter_fig+1]] <- paste("Figure3a_boxplot_overall_region_with_oultiers_",model_name[i],"_",out_suffix,".png",sep="")
|
|
| 396 |
|
|
| 392 | 397 |
## Figure 3b |
| 393 | 398 |
png(filename=paste("Figure3b_boxplot_overall_region_scaling_",model_name[i],"_",out_suffix,".png",sep=""),
|
| 394 | 399 |
width=col_mfrow*res_pix,height=row_mfrow*res_pix) |
| ... | ... | |
| 397 | 402 |
title("RMSE per model over all tiles")
|
| 398 | 403 |
|
| 399 | 404 |
dev.off() |
| 405 |
list_outfiles[[counter_fig+2]] <- paste("Figure3b_boxplot_overall_region_scaling_",model_name[i],"_",out_suffix,".png",sep="")
|
|
| 406 |
counter_fig <- counter_fig + 2 |
|
| 400 | 407 |
|
| 401 | 408 |
################ |
| 402 | 409 |
### Figure 4: plot predicted tiff for specific date per model |
| ... | ... | |
| 451 | 458 |
res_pix <- 480 |
| 452 | 459 |
col_mfrow <- 1 |
| 453 | 460 |
row_mfrow <- 1 |
| 454 |
|
|
| 461 |
fig_filename <- paste("Figure5_ac_metrics_ranked_",model_name[i],"_",out_suffix,".png",sep="")
|
|
| 462 |
|
|
| 455 | 463 |
png(filename=paste("Figure5_ac_metrics_ranked_",model_name[i],"_",out_suffix,".png",sep=""),
|
| 456 | 464 |
width=col_mfrow*res_pix,height=row_mfrow*res_pix) |
| 457 | 465 |
x<- as.character(df_ac_mod$tile_id) |
| ... | ... | |
| 461 | 469 |
title(paste("RMSE ranked by tile for ",model_name[i],sep=""))
|
| 462 | 470 |
|
| 463 | 471 |
dev.off() |
| 464 |
|
|
| 472 |
list_outfiles[[counter_fig+i]] <- fig_filename |
|
| 465 | 473 |
} |
| 466 | 474 |
|
| 475 |
counter_fig <- counter_fig+length(model_name) |
|
| 476 |
|
|
| 467 | 477 |
###################### |
| 468 | 478 |
### Figure 6: plot map of average RMSE per tile at centroids |
| 469 | 479 |
|
| ... | ... | |
| 482 | 492 |
|
| 483 | 493 |
col_mfrow <- 1 |
| 484 | 494 |
row_mfrow <- 1 |
| 485 |
|
|
| 495 |
fig_filename <- paste("Figure6_ac_metrics_map_centroids_tile_",model_name[i],"_",out_suffix,".png",sep="")
|
|
| 486 | 496 |
png(filename=paste("Figure6_ac_metrics_map_centroids_tile_",model_name[i],"_",out_suffix,".png",sep=""),
|
| 487 | 497 |
width=col_mfrow*res_pix,height=row_mfrow*res_pix) |
| 488 | 498 |
|
| ... | ... | |
| 490 | 500 |
coordinates(ac_mod) <- ac_mod[,c("lon.x","lat.x")] #solve this later
|
| 491 | 501 |
p_shp <- layer(sp.polygons(reg_layer, lwd=1, col='black')) |
| 492 | 502 |
#title("(a) Mean for 1 January")
|
| 493 |
p <- bubble(ac_mod,"rmse",main=paste("Averrage RMSE per tile and by ",model_name[i]))
|
|
| 503 |
p <- bubble(ac_mod,"rmse",main=paste("Average RMSE per tile and by ",model_name[i]))
|
|
| 494 | 504 |
p1 <- p+p_shp |
| 495 | 505 |
print(p1) |
| 496 | 506 |
#plot(ac_mod1,cex=(ac_mod1$rmse1)*2,pch=1,add=T) |
| ... | ... | |
| 501 | 511 |
### Ranking by tile... |
| 502 | 512 |
#df_ac_mod <- |
| 503 | 513 |
list_df_ac_mod[[i]] <- arrange(as.data.frame(ac_mod),desc(rmse))[,c("rmse","mae","tile_id")]
|
| 514 |
list_outfiles[[counter_fig+i]] <- fig_filename |
|
| 504 | 515 |
} |
| 516 |
counter_fig <- counter_fig+length(model_name) |
|
| 505 | 517 |
|
| 518 |
|
|
| 519 |
###################### |
|
| 520 |
### Figure 7: Number of predictions: daily and monthly |
|
| 521 |
|
|
| 522 |
## Figure 7a |
|
| 523 |
|
|
| 506 | 524 |
## Number of tiles with information: |
| 507 | 525 |
sum(df_tile_processed$metrics_v) #26,number of tiles with raster object |
| 508 | 526 |
length(df_tile_processed$metrics_v) #26,number of tiles in the region |
| ... | ... | |
| 535 | 553 |
|
| 536 | 554 |
col_mfrow <- 1 |
| 537 | 555 |
row_mfrow <- 1 |
| 538 |
|
|
| 556 |
fig_filename <- paste("Figure7a_ac_metrics_map_centroids_tile_",model_name[j],"_","missing_day_",threshold_missing_day[i],
|
|
| 557 |
"_",out_suffix,".png",sep="") |
|
| 539 | 558 |
png(filename=paste("Figure7a_ac_metrics_map_centroids_tile_",model_name[j],"_","missing_day_",threshold_missing_day[i],
|
| 540 | 559 |
"_",out_suffix,".png",sep=""), |
| 541 | 560 |
width=col_mfrow*res_pix,height=row_mfrow*res_pix) |
| ... | ... | |
| 548 | 567 |
threshold_missing_day[i])) |
| 549 | 568 |
p1 <- p+p_shp |
| 550 | 569 |
try(print(p1)) #error raised if number of missing values below a threshold does not exist |
| 551 |
|
|
| 552 | 570 |
dev.off() |
| 571 |
|
|
| 572 |
list_outfiles[[counter_fig+i]] <- fig_filename |
|
| 553 | 573 |
} |
| 574 |
counter_fig <- counter_fig+length(threshold_missing_day) |
|
| 554 | 575 |
|
| 555 |
###################### |
|
| 556 |
### Figure 7: Number of predictions: daily and monthly |
|
| 557 |
|
|
| 558 |
## Figure 7a |
|
| 559 |
png(filename=paste("Figure7a_number_daily_predictions_per_models","_",out_suffix,".png",sep=""),
|
|
| 576 |
png(filename=paste("Figure7b_number_daily_predictions_per_models","_",out_suffix,".png",sep=""),
|
|
| 560 | 577 |
width=col_mfrow*res_pix,height=row_mfrow*res_pix) |
| 561 | 578 |
|
| 562 | 579 |
xyplot(n~pred_mod | tile_id,data=subset(as.data.frame(summary_metrics_v), |
| 563 | 580 |
pred_mod!="mod_kr"),type="h") |
| 564 | 581 |
dev.off() |
| 565 | 582 |
|
| 583 |
list_outfiles[[counter_fig+1]] <- paste("Figure7b_number_daily_predictions_per_models","_",out_suffix,".png",sep="")
|
|
| 584 |
counter_fig <- counter_fig + 1 |
|
| 585 |
|
|
| 566 | 586 |
table(tb$pred_mod) |
| 567 | 587 |
table(tb$index_d) |
| 568 | 588 |
table(subset(tb,pred_mod!="mod_kr")) |
| ... | ... | |
| 581 | 601 |
table((subset(test, test$pred_mod=="mod1")$predicted)) |
| 582 | 602 |
|
| 583 | 603 |
#LST_avgm_min <- aggregate(LST~month,data=data_month_all,min) |
| 604 |
png(filename=paste("Figure7c_histogram_number_daily_predictions_per_models","_",out_suffix,".png",sep=""),
|
|
| 605 |
width=col_mfrow*res_pix,height=row_mfrow*res_pix) |
|
| 606 |
|
|
| 584 | 607 |
histogram(test$predicted~test$tile_id) |
| 608 |
dev.off() |
|
| 609 |
|
|
| 610 |
list_outfiles[[counter_fig+1]] <- paste("Figure7c_histogram_number_daily_predictions_per_models","_",out_suffix,".png",sep="")
|
|
| 611 |
counter_fig <- counter_fig + 1 |
|
| 612 |
|
|
| 585 | 613 |
#table(tb) |
| 586 | 614 |
## Figure 7b |
| 587 | 615 |
#png(filename=paste("Figure7b_number_daily_predictions_per_models","_",out_suffix,".png",sep=""),
|
| ... | ... | |
| 614 | 642 |
print(p) |
| 615 | 643 |
dev.off() |
| 616 | 644 |
|
| 645 |
list_outfiles[[counter_fig+1]] <- paste("Figure8a_boxplot_overall_separated_by_region_with_oultiers_",model_name[i],"_",out_suffix,".png",sep="")
|
|
| 646 |
counter_fig <- counter_fig + 1 |
|
| 647 |
|
|
| 617 | 648 |
## Figure 8b |
| 618 | 649 |
png(filename=paste("Figure8b_boxplot_overall_separated_by_region_scaling_",model_name[i],"_",out_suffix,".png",sep=""),
|
| 619 | 650 |
width=col_mfrow*res_pix,height=row_mfrow*res_pix) |
| ... | ... | |
| 625 | 656 |
print(p) |
| 626 | 657 |
dev.off() |
| 627 | 658 |
|
| 659 |
list_outfiles[[counter_fig+1]] <- paste("Figure8b_boxplot_overall_separated_by_region_scaling_",model_name[i],"_",out_suffix,".png",sep="")
|
|
| 660 |
counter_fig <- counter_fig + 1 |
|
| 661 |
|
|
| 628 | 662 |
##################################################### |
| 629 | 663 |
#### Figure 9: plotting boxplot by year and regions ########### |
| 630 | 664 |
|
| 631 |
## Figure 8a
|
|
| 665 |
## Figure 9a
|
|
| 632 | 666 |
res_pix <- 480 |
| 633 | 667 |
col_mfrow <- 1 |
| 634 | 668 |
row_mfrow <- 1 |
| ... | ... | |
| 636 | 670 |
png(filename=paste("Figure9a_boxplot_overall_separated_by_region_year_with_oultiers_",model_name[i],"_",out_suffix,".png",sep=""),
|
| 637 | 671 |
width=col_mfrow*res_pix,height=row_mfrow*res_pix) |
| 638 | 672 |
|
| 639 |
p<- bwplot(rmse~pred_mod | reg, data=tb, |
|
| 673 |
p<- bwplot(rmse~pred_mod | reg + year, data=tb,
|
|
| 640 | 674 |
main="RMSE per model and region over all tiles") |
| 641 | 675 |
print(p) |
| 642 | 676 |
dev.off() |
| 643 | 677 |
|
| 644 |
## Figure 8b
|
|
| 678 |
## Figure 9b
|
|
| 645 | 679 |
png(filename=paste("Figure8b_boxplot_overall_separated_by_region_year_scaling_",model_name[i],"_",out_suffix,".png",sep=""),
|
| 646 | 680 |
width=col_mfrow*res_pix,height=row_mfrow*res_pix) |
| 647 | 681 |
|
| ... | ... | |
| 652 | 686 |
print(p) |
| 653 | 687 |
dev.off() |
| 654 | 688 |
|
| 655 |
} |
|
| 689 |
list_outfiles[[counter_fig+1]] <- paste("Figure9a_boxplot_overall_separated_by_region_year_with_oultiers_",model_name[i],"_",out_suffix,".png",sep="")
|
|
| 690 |
counter_fig <- counter_fig + 1 |
|
| 691 |
|
|
| 692 |
############################################################## |
|
| 693 |
############## Prepare object to return |
|
| 694 |
############## Collect information from assessment ########## |
|
| 695 |
|
|
| 696 |
outfiles_names <- c("summary_metrics_v_names","tb_v_accuracy_name","tb_month_s_name","tb_s_accuracy_name",
|
|
| 697 |
"data_month_s_name","data_day_v_name","data_day_s_name","data_month_v_name", "tb_month_v_name", |
|
| 698 |
"pred_data_month_info_name","pred_data_day_info_name","df_tile_processed_name","df_tiles_all_name", |
|
| 699 |
"df_tiles_all_name") |
|
| 700 |
names(list_outfiles) <- outfiles_names |
|
| 701 |
|
|
| 702 |
#This data.frame contains all the files from the assessment |
|
| 703 |
df_assessment_figures_files <- data.frame(filename=outfiles_names,files=unlist(list_outfiles), |
|
| 704 |
reg=region_name,year=year_predicted) |
|
| 705 |
###Prepare files for copying back? |
|
| 706 |
df_assessment_figures_files_names <- file.path(out_dir,paste("df_assessment_files_",region_name,"_",year_predicted,"_",out_prefix,".txt",sep=""))
|
|
| 707 |
write.table(df_assessment_files, |
|
| 708 |
file=df_assessment_files_name,sep=",") |
|
| 656 | 709 |
|
| 710 |
#df_assessment_figures_files_names |
|
| 711 |
|
|
| 712 |
###################################################### |
|
| 713 |
##### Prepare objet to return #### |
|
| 657 | 714 |
|
| 715 |
#assessment_obj <- list(df_assessment_files, df_assessment_figures_files) |
|
| 716 |
#names(assessment_obj) <- c("df_assessment_files", "df_assessment_figures_files")
|
|
| 717 |
## Prepare list of files to return... |
|
| 718 |
return(df_assessment_figures_files_names) |
|
| 719 |
|
|
| 720 |
} |
|
| 658 | 721 |
|
| 659 | 722 |
##################### END OF SCRIPT ###################### |
| 723 |
|
|
| 724 |
# #comments #figure_no #region #models |
|
| 725 |
# tile processed for the region figure_1 reg4 NA |
|
| 726 |
# boxplot with outlier figure_2a reg4 mod1 |
|
| 727 |
# boxplot with outlier figure_2a reg4 mod_kr |
|
| 728 |
# boxplot scaling by tiles figure_2b reg4 mod1 |
|
| 729 |
# boxplot scaling by tiles figure_2b reg4 mod_kr |
|
| 730 |
# boxplot overall region with outliers figure_3a reg4 NA |
|
| 731 |
# boxplot overall region with scaling figure_3b reg4 NA |
|
| 732 |
# Barplot of metrics ranked by tile Figure_5 |
|
| 733 |
# Barplot of metrics ranked by tile Figure_5 |
|
| 734 |
# Average metrics map centroids Figure_6 |
|
| 735 |
# Average metrics map centroids Figure_6 |
|
| 736 |
# Number of missing day threshold1 map centroids Figure_7a |
|
| 737 |
# Number of missing day threshold1 map centroids Figure_7a |
|
| 738 |
# Number of missing day threshold1 map centroids Figure_7a |
|
| 739 |
# Number of missing day threshold1 map centroids Figure_7a |
|
| 740 |
# number_daily_predictions_per_model Figure_7b |
|
| 741 |
# histogram number_daily_predictions_per_models Figure_7c |
|
| 742 |
# boxplot_overall_separated_by_region_with_oultiers_ Figure 8a |
|
| 743 |
# boxplot_overall_separated_by_region_with_scaling Figure 8b |
|
Also available in: Unified diff
assessment part2 plotting of figures checking input parameters and clean up