HTW Hybrid Models

Modeling

ALM

EXAM

Author

Thomas Gorman

Published

March 16, 2024

Code

pacman::p_load(dplyr,purrr,tidyr,ggplot2, data.table, here, patchwork, conflicted, 
               stringr,future,furrr, knitr, reactable, flextable,ggstance, htmltools,kableExtra,ggdist)
conflict_prefer_all("dplyr", quiet = TRUE)
options(scipen = 999)
walk(c("Display_Functions","fun_alm","fun_indv_fit","fun_model"), ~ source(here::here(paste0("Functions/", .x, ".R"))))

E1

Code

ds <- readRDS(here::here("data/e1_md_11-06-23.rds"))  |> as.data.table()
nbins <- 3

fd <- readRDS(here("data/e1_08-21-23.rds"))
test <- fd |> filter(expMode2 == "Test") 
testAvg <- test %>% group_by(id, condit, vb, bandInt,bandType,tOrder) %>%
  summarise(nHits=sum(dist==0),vx=mean(vx),dist=mean(dist),sdist=mean(sdist),n=n(),Percent_Hit=nHits/n)

trainAvg <- fd |> filter(expMode2 == "Train") |> group_by(id) |> 
  mutate(tr=trial,x=vb,Block=case_when(expMode2=="Train" ~ cut(tr,breaks=seq(1,max(tr), length.out=nbins+1),include.lowest=TRUE,labels=FALSE),
                                         expMode2=="Test" ~ 4)) |> 
  group_by(id,condit,vb,x,Block) |> 
  summarise(dist=mean(dist),y=mean(vx))

input_layer <<- output_layer <<-  c(100,350,600,800,1000,1200)
ids2 <- c(1,66,36)

file_name <- "e1_hybrid_n_iter_250_ntry_200_0637"

ind_fits <- map(list.files(here(paste0('data/abc_reject/'),file_name),full.names=TRUE), readRDS)
ind_fits_df <- ind_fits |> map(~list(dat=.x[[1]], Model = .x[["Model"]], Fit_Method=.x[["Fit_Method"]]))
ind_fits_df <- ind_fits_df |> map(~rbindlist(.x$dat) |> mutate(Model = .x$Model, Fit_Method = .x$Fit_Method)) |> rbindlist() 


process_folder <- function(folder_name) {
  ind_fits <- map(list.files(here(paste0('data/abc_reject/'), folder_name), 
                             full.names = TRUE), readRDS)
  ind_fits_df <- ind_fits |> 
    map(~list(dat = .x[[1]], Model = .x[["Model"]], Fit_Method = .x[["Fit_Method"]],tolM=.x$tolM,ar=.x$min_accept_rate,
              t=.x$ri$elapsed)) |> 
    map(~rbindlist(.x$dat) |> 
           mutate(Model = .x$Model, Fit_Method = .x$Fit_Method,
                 exp = str_extract(folder_name, "^e\\d"),
                 n_int = as.numeric(str_extract(folder_name, "(?<=_n_iter_)\\d+")),
                 ntry = as.numeric(str_extract(folder_name, "(?<=_ntry_)\\d+")),
                 tolM=.x$tolM,ar=.x$ar,
                 run_name = folder_name,
                 min = round(.x$t/60,0))) |> 
    rbindlist()
  return(ind_fits_df)
}

# folder_names <- c("e1_hybrid_n_iter_250_ntry_200_0637",
# "e1_hybrid_n_iter_60_ntry_150_2335",
# "e1_hybrid_n_iter_200_ntry_300_0741",
# "e1_hybrid_n_iter_150_ntry_150_4749","e1_hybrid_n_iter_100_ntry_100_1435","e1_hybrid_n_iter_400_ntry_300_3744")

folder_names <- list.files(here("data/abc_reject"),pattern="e1_hyb*")



all_fits <- map_df(folder_names, process_folder)

all_fits |> 
  group_by(exp,n_int,ntry,tolM,ar,condit,Fit_Method) |> 
  summarise(min=first(min),me=mean(mean_error),
            w=median(weight_exam),sd_w=sd(weight_exam),c=median(c),
            lr=median(lr),n=n_distinct(id)) |>
   mutate(across(c(me, w, lr, sd_w), ~round(., 2))) |>
  arrange(condit,Fit_Method,me) |> kable(caption="E1 Fit Comparisons") |> 
  kable_styling(full_width = F) |>
  column_spec(8,bold=T,border_left=T)

E1 Fit Comparisons
exp	n_int	ntry	tolM	ar	condit	Fit_Method	min	me	w	sd_w	c	lr	n
e1	150	2500	0.65	0.10	Constant	Test	847	214.73	0.80	0.25	0.0000403	2.37	80
e1	100	2500	0.65	0.10	Constant	Test	209	214.87	0.80	0.25	0.0000402	2.37	80
e1	200	1500	0.70	0.10	Constant	Test	402	215.79	0.80	0.25	0.0000411	2.37	80
e1	50	900	0.80	0.12	Constant	Test	55	217.65	0.79	0.25	0.0000433	2.32	80
e1	90	900	0.80	0.09	Constant	Test	101	217.77	0.79	0.25	0.0000439	2.31	80
e1	150	500	0.95	0.10	Constant	Test	112	220.00	0.79	0.25	0.0000500	2.30	80
e1	70	500	0.85	0.10	Constant	Test	177	220.15	0.79	0.24	0.0000486	2.30	80
e1	400	300	0.85	0.04	Constant	Test	337	220.33	0.79	0.24	0.0000491	2.28	80
e1	50	500	0.85	0.05	Constant	Test	201	220.67	0.79	0.24	0.0000472	2.28	80
e1	200	300	0.90	0.10	Constant	Test	129	220.96	0.79	0.24	0.0000525	2.28	80
e1	49	800	0.95	0.18	Constant	Test	286	221.49	0.79	0.25	0.0000513	2.31	80
e1	61	500	0.95	0.10	Constant	Test	41	221.57	0.79	0.24	0.0000552	2.30	80
e1	50	900	0.95	0.10	Constant	Test	37	221.77	0.79	0.25	0.0000524	2.33	80
e1	250	200	0.85	0.10	Constant	Test	176	222.09	0.78	0.24	0.0000594	2.27	80
e1	52	600	0.95	0.18	Constant	Test	199	222.28	0.79	0.25	0.0000584	2.32	80
e1	65	300	0.95	0.10	Constant	Test	29	223.14	0.79	0.25	0.0000602	2.26	80
e1	100	100	0.85	0.01	Constant	Test	372	223.15	0.79	0.25	0.0000567	2.24	80
e1	150	150	0.85	0.10	Constant	Test	72	223.85	0.78	0.24	0.0000647	2.27	80
e1	51	300	0.95	0.15	Constant	Test	149	223.96	0.79	0.25	0.0000643	2.31	80
e1	60	150	0.85	0.10	Constant	Test	24	224.74	0.78	0.25	0.0000670	2.28	80
e1	105	100	0.85	0.05	Constant	Test	228	225.38	0.78	0.25	0.0000644	2.24	80
e1	100	100	0.90	0.10	Constant	Test	162	226.26	0.79	0.25	0.0000798	2.25	80
e1	100	2500	0.65	0.10	Constant	Test_Train	317	233.60	0.79	0.24	0.0000381	0.72	80
e1	200	1500	0.70	0.10	Constant	Test_Train	466	234.43	0.79	0.24	0.0000388	0.72	80
e1	90	900	0.80	0.09	Constant	Test_Train	120	235.49	0.78	0.25	0.0000388	0.73	80
e1	50	900	0.80	0.12	Constant	Test_Train	71	235.70	0.77	0.25	0.0000388	0.71	80
e1	70	500	0.85	0.10	Constant	Test_Train	173	237.28	0.77	0.25	0.0000401	0.76	80
e1	150	500	0.95	0.10	Constant	Test_Train	138	237.33	0.78	0.25	0.0000400	0.75	80
e1	50	500	0.85	0.05	Constant	Test_Train	217	237.64	0.78	0.25	0.0000401	0.77	80
e1	400	300	0.85	0.04	Constant	Test_Train	475	237.70	0.77	0.25	0.0000409	0.79	80
e1	49	800	0.95	0.18	Constant	Test_Train	263	238.08	0.78	0.25	0.0000400	0.79	80
e1	200	300	0.90	0.10	Constant	Test_Train	153	238.15	0.77	0.25	0.0000413	0.82	80
e1	50	900	0.95	0.10	Constant	Test_Train	38	238.19	0.78	0.25	0.0000398	0.79	80
e1	52	600	0.95	0.18	Constant	Test_Train	210	238.88	0.77	0.25	0.0000395	0.82	80
e1	250	200	0.85	0.10	Constant	Test_Train	183	239.01	0.77	0.25	0.0000429	0.83	80
e1	61	500	0.95	0.10	Constant	Test_Train	46	239.02	0.77	0.25	0.0000410	0.81	80
e1	100	100	0.85	0.01	Constant	Test_Train	465	239.10	0.77	0.25	0.0000429	0.83	80
e1	65	300	0.95	0.10	Constant	Test_Train	28	239.92	0.77	0.25	0.0000423	0.82	80
e1	150	150	0.85	0.10	Constant	Test_Train	89	240.04	0.77	0.25	0.0000453	0.84	80
e1	60	150	0.85	0.10	Constant	Test_Train	17	240.55	0.76	0.25	0.0000466	0.83	80
e1	51	300	0.95	0.15	Constant	Test_Train	137	240.77	0.77	0.25	0.0000424	0.87	80
e1	105	100	0.85	0.05	Constant	Test_Train	212	241.24	0.76	0.26	0.0000480	0.91	80
e1	100	100	0.90	0.10	Constant	Test_Train	194	242.04	0.76	0.25	0.0000519	0.94	80
e1	100	100	0.90	0.10	Constant	Train	30	211.21	0.50	0.29	0.0000284	2.00	135
e1	200	300	0.90	0.10	Constant	Train	146	220.24	0.50	0.29	0.0001014	2.08	80
e1	70	500	0.85	0.10	Constant	Train	84	220.80	0.48	0.29	0.0001000	2.09	80
e1	50	500	0.85	0.05	Constant	Train	91	221.54	0.50	0.29	0.0000965	2.08	80
e1	60	150	0.85	0.10	Constant	Train	32	221.76	0.50	0.29	0.0000750	1.38	80
e1	52	600	0.95	0.18	Constant	Train	84	224.05	0.50	0.29	0.0001002	2.01	80
e1	150	2500	0.65	0.10	Varied	Test	847	217.30	0.72	0.27	0.0000745	3.16	76
e1	100	2500	0.65	0.10	Varied	Test	209	217.57	0.72	0.27	0.0000807	3.11	76
e1	200	1500	0.70	0.10	Varied	Test	402	219.21	0.71	0.27	0.0000771	3.09	76
e1	90	900	0.80	0.09	Varied	Test	101	222.25	0.71	0.27	0.0000799	3.03	76
e1	50	900	0.80	0.12	Varied	Test	55	222.79	0.71	0.27	0.0000727	2.98	76
e1	70	500	0.85	0.10	Varied	Test	177	225.34	0.70	0.27	0.0000810	2.87	76
e1	150	500	0.95	0.10	Varied	Test	112	225.47	0.70	0.27	0.0000810	2.89	76
e1	50	900	0.95	0.10	Varied	Test	37	226.30	0.70	0.27	0.0000811	2.89	76
e1	400	300	0.85	0.04	Varied	Test	337	226.51	0.69	0.27	0.0000845	2.84	76
e1	50	500	0.85	0.05	Varied	Test	201	226.53	0.70	0.27	0.0000860	2.86	76
e1	49	800	0.95	0.18	Varied	Test	286	226.86	0.69	0.27	0.0000783	2.95	76
e1	61	500	0.95	0.10	Varied	Test	41	227.08	0.69	0.27	0.0000855	2.81	76
e1	200	300	0.90	0.10	Varied	Test	129	227.66	0.69	0.27	0.0000844	2.82	76
e1	52	600	0.95	0.18	Varied	Test	199	227.96	0.69	0.27	0.0000787	2.84	76
e1	250	200	0.85	0.10	Varied	Test	176	229.83	0.68	0.27	0.0000894	2.71	76
e1	100	100	0.85	0.01	Varied	Test	372	230.20	0.68	0.27	0.0000877	2.71	76
e1	65	300	0.95	0.10	Varied	Test	29	231.00	0.67	0.26	0.0000874	2.71	76
e1	51	300	0.95	0.15	Varied	Test	149	231.32	0.68	0.27	0.0000827	2.78	76
e1	150	150	0.85	0.10	Varied	Test	72	231.85	0.68	0.27	0.0000879	2.57	76
e1	60	150	0.85	0.10	Varied	Test	24	234.05	0.66	0.27	0.0000959	2.54	76
e1	105	100	0.85	0.05	Varied	Test	228	234.81	0.67	0.27	0.0000966	2.51	76
e1	100	100	0.90	0.10	Varied	Test	162	235.56	0.66	0.27	0.0000958	2.48	76
e1	100	2500	0.65	0.10	Varied	Test_Train	317	290.40	0.54	0.27	0.0000764	0.99	76
e1	200	1500	0.70	0.10	Varied	Test_Train	466	291.65	0.54	0.27	0.0000776	0.84	76
e1	90	900	0.80	0.09	Varied	Test_Train	120	293.35	0.55	0.27	0.0000747	0.78	76
e1	50	900	0.80	0.12	Varied	Test_Train	71	293.77	0.56	0.27	0.0000750	0.76	76
e1	70	500	0.85	0.10	Varied	Test_Train	173	295.56	0.55	0.27	0.0000752	0.73	76
e1	49	800	0.95	0.18	Varied	Test_Train	263	295.63	0.55	0.26	0.0000721	0.78	76
e1	50	500	0.85	0.05	Varied	Test_Train	217	295.74	0.56	0.27	0.0000708	0.71	76
e1	50	900	0.95	0.10	Varied	Test_Train	38	295.80	0.55	0.26	0.0000748	0.79	76
e1	150	500	0.95	0.10	Varied	Test_Train	138	295.84	0.55	0.27	0.0000732	0.75	76
e1	52	600	0.95	0.18	Varied	Test_Train	210	296.14	0.56	0.26	0.0000705	0.76	76
e1	400	300	0.85	0.04	Varied	Test_Train	475	296.24	0.56	0.26	0.0000720	0.74	76
e1	61	500	0.95	0.10	Varied	Test_Train	46	296.78	0.56	0.27	0.0000710	0.78	76
e1	200	300	0.90	0.10	Varied	Test_Train	153	296.83	0.56	0.27	0.0000713	0.72	76
e1	100	100	0.85	0.01	Varied	Test_Train	465	297.79	0.55	0.27	0.0000734	0.75	76
e1	250	200	0.85	0.10	Varied	Test_Train	183	298.19	0.55	0.27	0.0000712	0.71	76
e1	65	300	0.95	0.10	Varied	Test_Train	28	298.71	0.56	0.26	0.0000735	0.74	76
e1	150	150	0.85	0.10	Varied	Test_Train	89	299.31	0.56	0.26	0.0000700	0.71	76
e1	51	300	0.95	0.15	Varied	Test_Train	137	299.59	0.55	0.26	0.0000738	0.72	76
e1	60	150	0.85	0.10	Varied	Test_Train	17	300.04	0.56	0.26	0.0000687	0.69	76
e1	105	100	0.85	0.05	Varied	Test_Train	212	300.51	0.56	0.26	0.0000690	0.69	76
e1	100	100	0.90	0.10	Varied	Test_Train	194	301.28	0.56	0.27	0.0000667	0.70	76
e1	100	100	0.90	0.10	Varied	Train	30	265.62	0.50	0.29	0.0000226	0.92	131
e1	70	500	0.85	0.10	Varied	Train	84	312.06	0.49	0.29	0.0000266	1.51	76
e1	50	500	0.85	0.05	Varied	Train	91	312.34	0.51	0.28	0.0000261	1.38	76
e1	200	300	0.90	0.10	Varied	Train	146	313.11	0.50	0.29	0.0000260	1.40	76
e1	52	600	0.95	0.18	Varied	Train	84	314.91	0.49	0.29	0.0000267	1.40	76
e1	60	150	0.85	0.10	Varied	Train	32	315.65	0.49	0.29	0.0000259	1.14	76

Code

all_fits |> 
  filter(id %in% unique(all_fits$id)[1:2]) |>
  group_by(id,exp,n_int,ntry,condit,Fit_Method) |> 
  mutate(rank=rank(mean_error)) |> filter(rank<n_int*.90) |>
  group_by(id,exp,n_int,ntry,condit,Fit_Method) |> 
  summarise(me=mean(mean_error),w=median(weight_exam),sd_w=sd(weight_exam),c=median(c),lr=median(lr)) |>
  arrange(id,condit,Fit_Method,me) |> 
    kbl() |> kable_styling(full_width = F)

id	exp	n_int	ntry	condit	Fit_Method	me	w	sd_w	c	lr
1	e1	100	2500	Varied	Test	223.5378	0.8182305	0.2035620	0.0000019	5.2816922
1	e1	150	2500	Varied	Test	223.5554	0.8585789	0.1787275	0.0000022	5.1924980
1	e1	50	900	Varied	Test	223.6162	0.7473703	0.1897786	0.0000019	5.2693800
1	e1	150	500	Varied	Test	224.6697	0.8149536	0.1970840	0.0000020	5.1997612
1	e1	200	1500	Varied	Test	225.4042	0.7649556	0.2123507	0.0000019	5.2531997
1	e1	90	900	Varied	Test	226.6014	0.7998805	0.2525561	0.0000020	5.1910494
1	e1	100	100	Varied	Test	228.0284	0.8293703	0.2262617	0.0000023	5.1435685
1	e1	70	500	Varied	Test	230.9324	0.8355013	0.2819217	0.0000020	5.1279383
1	e1	400	300	Varied	Test	231.2286	0.8104574	0.2420737	0.0000023	5.1076418
1	e1	200	300	Varied	Test	232.2169	0.8051852	0.2395735	0.0000023	5.0557794
1	e1	50	500	Varied	Test	232.5206	0.8222143	0.2435945	0.0000021	5.0075482
1	e1	52	600	Varied	Test	233.0113	0.8626639	0.1970062	0.0000830	2.6120010
1	e1	49	800	Varied	Test	233.2076	0.8204706	0.1984141	0.0000023	5.0773731
1	e1	250	200	Varied	Test	233.5362	0.7986295	0.2476705	0.0000021	5.0901121
1	e1	61	500	Varied	Test	233.7241	0.8264059	0.2336978	0.0000023	5.0624427
1	e1	60	150	Varied	Test	235.2363	0.7849864	0.2606166	0.0000024	5.0587135
1	e1	65	300	Varied	Test	236.8494	0.7526598	0.2464510	0.0000021	5.1046996
1	e1	51	300	Varied	Test	236.8643	0.8326864	0.2732785	0.0000668	4.6356408
1	e1	150	150	Varied	Test	238.8422	0.8057119	0.2588631	0.0000046	4.8535159
1	e1	105	100	Varied	Test	240.9037	0.7794698	0.2547383	0.0001021	1.8109660
1	e1	100	2500	Varied	Test_Train	291.1208	0.4982487	0.2149724	0.0000627	0.1050512
1	e1	90	900	Varied	Test_Train	291.8641	0.4633852	0.1755527	0.0000643	0.1239611
1	e1	200	1500	Varied	Test_Train	292.0506	0.4656865	0.1840448	0.0000650	0.1342331
1	e1	50	900	Varied	Test_Train	292.7664	0.4828479	0.2012370	0.0000584	0.0823038
1	e1	150	500	Varied	Test_Train	293.0012	0.4539042	0.1576420	0.0000639	0.1014355
1	e1	70	500	Varied	Test_Train	294.7952	0.4768128	0.1804256	0.0000711	0.1343180
1	e1	400	300	Varied	Test_Train	294.8046	0.4961300	0.1948949	0.0000650	0.1460962
1	e1	50	500	Varied	Test_Train	294.8834	0.4763972	0.1708669	0.0000729	0.1072454
1	e1	49	800	Varied	Test_Train	295.0383	0.4812279	0.1352303	0.0000736	0.1449286
1	e1	100	100	Varied	Test_Train	295.4118	0.4917216	0.1817776	0.0000666	0.1636531
1	e1	52	600	Varied	Test_Train	295.4456	0.4399127	0.1818114	0.0000624	0.0987222
1	e1	200	300	Varied	Test_Train	295.9199	0.4870067	0.2079155	0.0000664	0.1371202
1	e1	250	200	Varied	Test_Train	296.5507	0.4909003	0.1756808	0.0000672	0.1867037
1	e1	105	100	Varied	Test_Train	297.3157	0.4826085	0.1881216	0.0000674	0.2043142
1	e1	51	300	Varied	Test_Train	297.3525	0.4793596	0.2287758	0.0000671	0.2355805
1	e1	61	500	Varied	Test_Train	297.3594	0.4408326	0.1632677	0.0000699	0.1775518
1	e1	60	150	Varied	Test_Train	297.5422	0.5341098	0.1860066	0.0000555	0.1508889
1	e1	150	150	Varied	Test_Train	298.1399	0.5021449	0.1864520	0.0000740	0.1710864
1	e1	65	300	Varied	Test_Train	300.3247	0.4955346	0.1788877	0.0000656	0.2270260
1	e1	200	300	Varied	Train	301.0392	0.4510923	0.2769286	0.0000187	8.0239746
1	e1	70	500	Varied	Train	301.9224	0.5754325	0.2973605	0.0000418	7.9297719
1	e1	50	500	Varied	Train	302.6788	0.5817460	0.2974083	0.0000424	8.1026574
1	e1	60	150	Varied	Train	305.4690	0.4887447	0.2976968	0.0000476	7.9030316
1	e1	52	600	Varied	Train	306.0869	0.4761011	0.3128097	0.0000496	7.8344141
1	e1	100	100	Varied	Train	306.6696	0.5408164	0.2728925	0.0000464	7.8161855
2	e1	100	2500	Varied	Test	369.8400	0.5863439	0.2981269	0.0000348	4.0355545
2	e1	150	2500	Varied	Test	369.8738	0.5306497	0.3050162	0.0000338	5.0224378
2	e1	150	500	Varied	Test	370.1960	0.5485465	0.2928199	0.0000335	5.0353815
2	e1	50	900	Varied	Test	370.5906	0.4917389	0.2895874	0.0000317	5.1695962
2	e1	200	1500	Varied	Test	371.4073	0.4989873	0.2888450	0.0000336	5.0766927
2	e1	90	900	Varied	Test	374.1310	0.5857368	0.2691902	0.0000345	4.9535572
2	e1	100	100	Varied	Test	374.7765	0.5144509	0.2778470	0.0000341	4.7317420
2	e1	49	800	Varied	Test	376.1818	0.4814195	0.3264464	0.0000352	4.4422502
2	e1	61	500	Varied	Test	376.4798	0.6001896	0.2867825	0.0000353	4.0377088
2	e1	400	300	Varied	Test	377.7010	0.4745654	0.2921842	0.0000345	4.0564093
2	e1	50	500	Varied	Test	377.7084	0.4700296	0.2752379	0.0000385	4.0132306
2	e1	200	300	Varied	Test	377.9897	0.4644357	0.3013704	0.0000351	4.1080690
2	e1	70	500	Varied	Test	378.6938	0.5585721	0.3168538	0.0000356	4.0292617
2	e1	51	300	Varied	Test	380.1584	0.4959267	0.2805370	0.0000381	4.0102128
2	e1	65	300	Varied	Test	381.2897	0.4558618	0.3250864	0.0000353	4.0229159
2	e1	52	600	Varied	Test	381.8625	0.6456892	0.3142880	0.0000380	3.9758295
2	e1	250	200	Varied	Test	382.6800	0.5574183	0.2947822	0.0000368	4.0376555
2	e1	105	100	Varied	Test	386.5428	0.4663750	0.3017272	0.0000374	4.1107081
2	e1	150	150	Varied	Test	386.6443	0.5275578	0.3094980	0.0000363	4.0164842
2	e1	60	150	Varied	Test	387.5823	0.4715924	0.3332044	0.0000386	4.0013609
2	e1	50	900	Varied	Test_Train	428.1180	0.4978363	0.2970980	0.0000385	4.9800328
2	e1	150	500	Varied	Test_Train	429.5278	0.4630116	0.2978593	0.0000375	4.9235301
2	e1	100	2500	Varied	Test_Train	429.7416	0.4441688	0.2806442	0.0000385	4.8920111
2	e1	200	1500	Varied	Test_Train	432.2028	0.4870858	0.2860881	0.0000373	4.8960848
2	e1	49	800	Varied	Test_Train	436.2951	0.3925758	0.2483927	0.0000358	4.9014996
2	e1	100	100	Varied	Test_Train	438.4416	0.4760461	0.3051119	0.0000351	4.8056798
2	e1	90	900	Varied	Test_Train	439.2872	0.5689515	0.3051232	0.0000354	4.9116788
2	e1	52	600	Varied	Test_Train	441.2831	0.5517423	0.3000017	0.0000357	4.6911650
2	e1	61	500	Varied	Test_Train	441.9809	0.4955319	0.2817430	0.0000361	4.9272707
2	e1	70	500	Varied	Test_Train	442.1006	0.6065832	0.3080914	0.0000364	4.9111936
2	e1	65	300	Varied	Test_Train	443.6291	0.4970180	0.2848160	0.0000341	4.8086849
2	e1	400	300	Varied	Test_Train	444.0730	0.5335898	0.2966844	0.0000351	4.9310083
2	e1	50	500	Varied	Test_Train	444.9344	0.5975558	0.2789440	0.0000351	4.7531000
2	e1	200	300	Varied	Test_Train	445.7221	0.4626835	0.2983041	0.0000358	4.9009381
2	e1	51	300	Varied	Test_Train	448.7055	0.3941551	0.2908179	0.0000349	4.9279146
2	e1	250	200	Varied	Test_Train	451.1523	0.4805013	0.2915747	0.0000357	4.8348896
2	e1	60	150	Varied	Test_Train	455.9307	0.4307338	0.2992679	0.0000349	4.9789964
2	e1	150	150	Varied	Test_Train	456.8058	0.6124759	0.3025009	0.0000359	4.7546298
2	e1	105	100	Varied	Test_Train	457.5066	0.5686353	0.2888873	0.0000361	4.7358573
2	e1	200	300	Varied	Train	445.4326	0.4992835	0.2903908	0.0000143	1.2852145
2	e1	50	500	Varied	Train	446.0390	0.4863962	0.2996348	0.0000139	1.2527575
2	e1	52	600	Varied	Train	446.2512	0.4010491	0.2681370	0.0000138	1.2558990
2	e1	70	500	Varied	Train	446.4922	0.4650381	0.2920932	0.0000139	1.2373839
2	e1	60	150	Varied	Train	447.9030	0.5259795	0.2935894	0.0000131	1.1056549
2	e1	100	100	Varied	Train	448.5521	0.4988236	0.2767507	0.0000149	1.0656699

Code

k = all_fits |> 
    filter(id %in% unique(all_fits$id)[1])  |>
  group_by(Fit_Method) |> arrange(mean_error)

id_mdif <- all_fits |> 
    group_by(id,condit,Fit_Method,run_name) |> 
  mutate(rank=rank(mean_error)) |>
  summarize(n=n(),all_me=mean(mean_error),
            p50=mean(mean_error[rank<n*.50]),
         top10 = mean(mean_error[rank<10])) 

id_wdif <- all_fits |> 
    group_by(id,condit,Fit_Method,run_name) |> 
  mutate(rank=rank(mean_error)) |>
  summarize(n=n(),all_we=mean(weight_exam),
            p50=mean(weight_exam[rank<n*.50]),
         top10 = mean(weight_exam[rank<10])) |>
  mutate(dif=all_we-top10,adif=abs(dif)) |> arrange(adif)

id_wdifAll <- all_fits |> 
    group_by(id,condit,Fit_Method) |> 
  mutate(rank=rank(mean_error)) |>
  summarize(n=n(),all_we=median(weight_exam),
            p50=median(weight_exam[rank<n*.50]),
         top10 = median(weight_exam[rank<30])) |>
  mutate(dif=all_we-top10,adif=abs(dif)) |> arrange(-adif)

Code

# ind_fits_df |> group_by(id,condit,Fit_Method) |> summarise(w=median(weight_exam),me=mean(mean_error))
# 
# ind_fits_df |> group_by(id,condit,Fit_Method) |> summarise(we=median(weight_exam),me=mean(mean_error)) |>
#   group_by(condit,Fit_Method) |> summarise(w=mean(we),sd_w=sd(we),me=mean(me)) 
# 
# ind_fits_df |> 
#   group_by(condit,Fit_Method) |> 
#   summarise(w=median(weight_exam),sd_w=sd(weight_exam),me=mean(mean_error)) 

{all_fits |> 
   filter(Fit_Method=="Test") |>
    ggplot(aes(x=weight_exam,fill=condit)) +geom_density(alpha=.5) } /{

all_fits |> 
   filter(Fit_Method=="Test") |>
    ggplot(aes(x=weight_exam,fill=condit)) +geom_density(alpha=.5) +
  facet_wrap(~run_name)
    }

Code

all_fits |> ggplot(aes(x=condit,y=weight_exam,col=condit)) + stat_pointinterval() +
  facet_wrap(~Fit_Method)

Code

all_fits |> ggplot(aes(x=run_name,y=weight_exam,col=condit)) + stat_pointinterval() +
  facet_wrap(condit~Fit_Method)

Code

all_fits |> 
   filter(id %in% unique(all_fits$id)[1:50], Fit_Method=="Test") |>
    ggplot(aes(x=weight_exam,fill=condit)) +
  geom_density() +
  geom_vline(xintercept = .5,linetype="dashed") +
  ggh4x::facet_wrap2(~id+condit, scales="free_y")

Code

all_fits |> 
   filter(id %in% unique(all_fits$id)[1:50], Fit_Method=="Test_Train") |>
    ggplot(aes(x=weight_exam,fill=condit)) +
  geom_density() +
  geom_vline(xintercept = .5,linetype="dashed") +
  ggh4x::facet_wrap2(~id+condit, scales="free_y")

Code

all_fits |> 
    group_by(id,run_name,condit,Fit_Method) |> 
  mutate(rank=rank(mean_error)) |> 
  filter(rank<n_int*.50) |>
   filter(id %in% unique(all_fits$id)[1:50], Fit_Method=="Test") |>
    ggplot(aes(x=weight_exam,fill=condit)) +
  geom_density() +
  geom_vline(xintercept = .5,linetype="dashed") +
  ggh4x::facet_wrap2(~id+condit, scales="free_y")

Code

all_fits |> 
  group_by(id,condit,Fit_Method) |>
  mutate(we_med=median(weight_exam),
         Best_Model=case_when(we_med>.5 ~"EXAM",we_med<.5 ~"ALM")) |>
  filter(Fit_Method=="Test") |>
  ungroup() |>
  mutate(id=reorder(id,we_med,decreasing = TRUE)) |>
  ggplot(aes(x=weight_exam,y=id,col=Best_Model)) + 
  geom_vline(xintercept = .5,linetype="dashed") +
  stat_pointinterval() + 
  ggh4x::facet_grid2(~condit,axes="all",scales="free_y", independent = "y")+
  theme_minimal() +
  theme(legend.position="top")

Code

all_fits |> 
  filter(Fit_Method=="Test") |>
    group_by(id,run_name,condit,Fit_Method) |> 
  mutate(rank=rank(mean_error)) |> 
  filter(rank<n_int*.50) |>
  group_by(id,condit,Fit_Method,run_name) |>
  mutate(we_med=median(weight_exam),
         Best_Model=case_when(we_med>.5 ~"EXAM",we_med<.5 ~"ALM")) |>
  ungroup() |>
  group_by(run_name) |>
    mutate(id=reorder(id,we_med,decreasing = FALSE)) |>
  ggplot(aes(x=weight_exam,y=id,col=Best_Model)) + 
  geom_vline(xintercept = .5,linetype="dashed") +
  stat_pointinterval() + 
  ggh4x::facet_grid2(run_name~condit,axes="all",scales="free_y", independent = "y")+
  theme_minimal() +
  theme(legend.position="top")

Code

all_fits |> ggplot(aes(x=condit,y=weight_exam,col=run_name)) + 
  stat_pointinterval(position=position_dodge(.5)) +
  facet_wrap(condit~Fit_Method)

Code

all_fits |> 
   filter(Fit_Method=="Test") |>
    ggplot(aes(x=weight_exam,fill=condit)) +geom_density(alpha=.5)

Code

all_fits |> 
   filter(id %in% unique(all_fits$id)[1:50], Fit_Method=="Test") |>
  ggplot(aes(x=run_name,y=weight_exam,col=run_name)) + 
  stat_pointinterval(position=position_dodge(.5)) +
  ggh4x::facet_wrap2(~id+condit) + 
  theme(axis.text.x = element_blank())

Code

all_fits |> 
   filter(id %in% unique(all_fits$id)[1:50], Fit_Method=="Test_Train") |>
  ggplot(aes(x=run_name,y=weight_exam,col=run_name)) + 
  stat_pointinterval(position=position_dodge(.5)) +
  ggh4x::facet_wrap2(~id+condit) + 
  theme(axis.text.x = element_blank())

E2

Code

# folder_names <- c("e2_hybrid_n_iter_60_ntry_150_3658","e2_hybrid_n_iter_250_ntry_200_0540","e2_hybrid_n_iter_150_ntry_150_2904","e2_hybrid_n_iter_100_ntry_100_3211")

folder_names <- list.files(here("data/abc_reject"),pattern="e2_hyb*")


all_fits <- map_df(folder_names, process_folder)

all_fits |> 
  group_by(exp,n_int,ntry,tolM,ar,condit,Fit_Method) |> 
  summarise(min=first(min),me=mean(mean_error),
            w=median(weight_exam),sd_w=sd(weight_exam),c=median(c),
            lr=median(lr),n=n_distinct(id)) |>
   mutate(across(c(me, w, lr, sd_w), ~round(., 2))) |>
  arrange(condit,Fit_Method,me) |> kable(caption="E2 Fit Comparisons") |> 
  kable_styling(full_width = F) |>
  column_spec(8,bold=T,border_left=T)

E2 Fit Comparisons
exp	n_int	ntry	tolM	ar	condit	Fit_Method	min	me	w	sd_w	c	lr	n
e2	100	2500	0.65	0.10	Constant	Test	171	221.67	0.89	0.21	0.0000892	1.82	55
e2	200	1500	0.70	0.10	Constant	Test	268	222.80	0.88	0.21	0.0000989	1.82	55
e2	90	900	0.80	0.09	Constant	Test	68	224.39	0.87	0.22	0.0001410	1.84	55
e2	50	900	0.80	0.12	Constant	Test	33	225.12	0.87	0.22	0.0001354	1.82	55
e2	70	500	0.85	0.10	Constant	Test	87	227.01	0.86	0.22	0.0001926	1.86	55
e2	150	500	0.95	0.10	Constant	Test	81	227.14	0.86	0.22	0.0001649	1.82	55
e2	400	300	0.85	0.04	Constant	Test	249	227.22	0.86	0.22	0.0001926	1.85	55
e2	50	500	0.85	0.05	Constant	Test	108	227.28	0.86	0.22	0.0001984	1.84	55
e2	50	900	0.95	0.10	Constant	Test	32	228.36	0.86	0.23	0.0001890	1.84	55
e2	52	600	0.95	0.18	Constant	Test	118	228.78	0.86	0.22	0.0002007	1.87	55
e2	61	500	0.95	0.10	Constant	Test	24	228.91	0.85	0.22	0.0002036	1.89	55
e2	49	800	0.95	0.18	Constant	Test	139	228.94	0.86	0.22	0.0001620	1.84	55
e2	250	200	0.85	0.10	Constant	Test	74	229.10	0.86	0.22	0.0002492	1.88	55
e2	100	100	0.85	0.01	Constant	Test	225	229.42	0.86	0.22	0.0002304	1.85	55
e2	65	300	0.95	0.10	Constant	Test	15	230.22	0.86	0.22	0.0002432	1.88	55
e2	150	150	0.85	0.10	Constant	Test	57	230.44	0.85	0.23	0.0002646	1.87	55
e2	51	300	0.95	0.15	Constant	Test	81	231.20	0.84	0.23	0.0002761	1.87	55
e2	60	150	0.85	0.10	Constant	Test	12	231.32	0.84	0.23	0.0003111	1.90	55
e2	105	100	0.85	0.05	Constant	Test	126	232.04	0.85	0.22	0.0003055	1.87	55
e2	100	100	0.85	0.10	Constant	Test	92	232.82	0.84	0.22	0.0003746	1.94	55
e2	100	2500	0.65	0.10	Constant	Test_Train	172	221.92	0.84	0.24	0.0000962	0.38	55
e2	200	1500	0.70	0.10	Constant	Test_Train	301	222.68	0.84	0.24	0.0001096	0.42	55
e2	90	900	0.80	0.09	Constant	Test_Train	62	223.95	0.84	0.24	0.0001247	0.48	55
e2	50	900	0.80	0.12	Constant	Test_Train	39	224.60	0.84	0.23	0.0001343	0.53	55
e2	400	300	0.85	0.04	Constant	Test_Train	191	225.77	0.83	0.24	0.0001481	0.69	55
e2	70	500	0.85	0.10	Constant	Test_Train	95	225.87	0.82	0.24	0.0001523	0.67	55
e2	50	500	0.85	0.05	Constant	Test_Train	118	225.89	0.83	0.24	0.0001450	0.64	55
e2	150	500	0.95	0.10	Constant	Test_Train	74	225.96	0.82	0.24	0.0001472	0.66	55
e2	100	100	0.85	0.01	Constant	Test_Train	236	227.13	0.82	0.24	0.0001532	0.74	55
e2	250	200	0.85	0.10	Constant	Test_Train	83	227.25	0.82	0.24	0.0001663	0.88	55
e2	49	800	0.95	0.18	Constant	Test_Train	152	227.38	0.83	0.24	0.0001500	0.76	55
e2	50	900	0.95	0.10	Constant	Test_Train	31	227.49	0.81	0.25	0.0001571	0.69	55
e2	52	600	0.95	0.18	Constant	Test_Train	130	227.65	0.82	0.24	0.0001712	0.80	55
e2	61	500	0.95	0.10	Constant	Test_Train	23	227.79	0.81	0.24	0.0001700	0.89	55
e2	150	150	0.85	0.10	Constant	Test_Train	52	227.94	0.82	0.24	0.0001829	1.03	55
e2	105	100	0.85	0.05	Constant	Test_Train	124	228.84	0.82	0.24	0.0002131	1.22	55
e2	51	300	0.95	0.15	Constant	Test_Train	90	228.86	0.82	0.24	0.0001958	1.20	55
e2	60	150	0.85	0.10	Constant	Test_Train	15	228.86	0.81	0.24	0.0002271	1.66	55
e2	65	300	0.95	0.10	Constant	Test_Train	18	229.04	0.81	0.24	0.0001718	0.96	55
e2	100	100	0.85	0.10	Constant	Test_Train	91	229.41	0.82	0.24	0.0002158	1.61	55
e2	50	500	0.85	0.05	Constant	Train	36	193.84	0.50	0.29	0.0000033	2.43	55
e2	70	500	0.85	0.10	Constant	Train	28	193.88	0.50	0.29	0.0000030	2.40	55
e2	100	100	0.85	0.10	Constant	Train	31	194.16	0.49	0.29	0.0000028	2.33	55
e2	60	150	0.85	0.10	Constant	Train	5	194.66	0.51	0.29	0.0000032	2.35	55
e2	52	600	0.95	0.18	Constant	Train	31	196.66	0.50	0.29	0.0000037	2.45	55
e2	100	2500	0.65	0.10	Varied	Test	171	208.73	0.87	0.26	0.0001168	1.95	55
e2	200	1500	0.70	0.10	Varied	Test	268	209.95	0.86	0.26	0.0001098	1.95	55
e2	90	900	0.80	0.09	Varied	Test	68	211.50	0.85	0.27	0.0001103	1.89	55
e2	50	900	0.80	0.12	Varied	Test	33	212.52	0.85	0.26	0.0000967	1.87	55
e2	150	500	0.95	0.10	Varied	Test	81	214.26	0.84	0.26	0.0001105	1.88	55
e2	50	500	0.85	0.05	Varied	Test	108	214.52	0.85	0.27	0.0000934	1.83	55
e2	400	300	0.85	0.04	Varied	Test	249	214.62	0.84	0.26	0.0001073	1.87	55
e2	70	500	0.85	0.10	Varied	Test	87	214.71	0.84	0.27	0.0001021	1.88	55
e2	50	900	0.95	0.10	Varied	Test	32	215.29	0.84	0.27	0.0000970	1.90	55
e2	52	600	0.95	0.18	Varied	Test	118	215.97	0.84	0.26	0.0001130	1.93	55
e2	49	800	0.95	0.18	Varied	Test	139	216.30	0.84	0.26	0.0001029	1.90	55
e2	61	500	0.95	0.10	Varied	Test	24	216.51	0.84	0.27	0.0001072	1.91	55
e2	250	200	0.85	0.10	Varied	Test	74	216.66	0.83	0.26	0.0001020	1.87	55
e2	100	100	0.85	0.01	Varied	Test	225	217.90	0.82	0.27	0.0000965	1.85	55
e2	150	150	0.85	0.10	Varied	Test	57	217.98	0.83	0.27	0.0001039	1.85	55
e2	65	300	0.95	0.10	Varied	Test	15	218.08	0.83	0.27	0.0000977	1.90	55
e2	51	300	0.95	0.15	Varied	Test	81	219.08	0.83	0.27	0.0001295	1.85	55
e2	60	150	0.85	0.10	Varied	Test	12	219.11	0.83	0.26	0.0001021	1.83	55
e2	105	100	0.85	0.05	Varied	Test	126	219.59	0.82	0.26	0.0001012	1.82	55
e2	100	100	0.85	0.10	Varied	Test	92	220.43	0.81	0.26	0.0001067	1.81	55
e2	100	2500	0.65	0.10	Varied	Test_Train	172	219.33	0.89	0.30	0.0000330	0.60	55
e2	200	1500	0.70	0.10	Varied	Test_Train	301	220.15	0.88	0.29	0.0000330	0.62	55
e2	50	900	0.80	0.12	Varied	Test_Train	39	221.27	0.87	0.29	0.0000316	0.62	55
e2	90	900	0.80	0.09	Varied	Test_Train	62	221.29	0.86	0.29	0.0000305	0.64	55
e2	70	500	0.85	0.10	Varied	Test_Train	95	222.78	0.85	0.29	0.0000291	0.67	55
e2	150	500	0.95	0.10	Varied	Test_Train	74	223.03	0.85	0.28	0.0000292	0.67	55
e2	50	500	0.85	0.05	Varied	Test_Train	118	223.40	0.85	0.28	0.0000289	0.67	55
e2	400	300	0.85	0.04	Varied	Test_Train	191	223.41	0.84	0.28	0.0000281	0.68	55
e2	50	900	0.95	0.10	Varied	Test_Train	31	223.58	0.86	0.29	0.0000299	0.71	55
e2	61	500	0.95	0.10	Varied	Test_Train	23	223.86	0.84	0.28	0.0000292	0.70	55
e2	49	800	0.95	0.18	Varied	Test_Train	152	224.02	0.85	0.28	0.0000329	0.69	55
e2	52	600	0.95	0.18	Varied	Test_Train	130	224.57	0.85	0.28	0.0000307	0.68	55
e2	250	200	0.85	0.10	Varied	Test_Train	83	224.70	0.82	0.28	0.0000269	0.71	55
e2	100	100	0.85	0.01	Varied	Test_Train	236	224.85	0.83	0.28	0.0000274	0.70	55
e2	65	300	0.95	0.10	Varied	Test_Train	18	225.46	0.83	0.28	0.0000273	0.72	55
e2	150	150	0.85	0.10	Varied	Test_Train	52	225.70	0.81	0.28	0.0000255	0.73	55
e2	51	300	0.95	0.15	Varied	Test_Train	90	226.28	0.83	0.28	0.0000276	0.74	55
e2	105	100	0.85	0.05	Varied	Test_Train	124	226.63	0.80	0.28	0.0000251	0.72	55
e2	60	150	0.85	0.10	Varied	Test_Train	15	226.77	0.80	0.28	0.0000256	0.76	55
e2	100	100	0.85	0.10	Varied	Test_Train	91	227.37	0.79	0.27	0.0000253	0.77	55
e2	70	500	0.85	0.10	Varied	Train	28	192.05	0.50	0.29	0.0000191	0.68	55
e2	50	500	0.85	0.05	Varied	Train	36	193.06	0.50	0.29	0.0000194	0.69	55
e2	100	100	0.85	0.10	Varied	Train	31	193.62	0.51	0.29	0.0000190	0.70	55
e2	60	150	0.85	0.10	Varied	Train	5	193.70	0.50	0.29	0.0000190	0.70	55
e2	52	600	0.95	0.18	Varied	Train	31	195.83	0.51	0.29	0.0000198	0.71	55

Code

{all_fits |> 
   filter(Fit_Method=="Test") |>
    ggplot(aes(x=weight_exam,fill=condit)) +geom_density(alpha=.5) } /{

all_fits |> 
   filter(Fit_Method=="Test") |>
    ggplot(aes(x=weight_exam,fill=condit)) +geom_density(alpha=.5) +
  facet_wrap(~run_name)
    }

Code

all_fits |> ggplot(aes(x=condit,y=weight_exam,col=condit)) + stat_pointinterval() +
  facet_wrap(~Fit_Method)

Code

all_fits |> ggplot(aes(x=run_name,y=weight_exam,col=condit)) + stat_pointinterval() +
  facet_wrap(condit~Fit_Method)

Code

all_fits |> 
   filter(id %in% unique(all_fits$id)[1:50], Fit_Method=="Test") |>
    ggplot(aes(x=weight_exam,fill=condit)) +
  geom_density() +
  geom_vline(xintercept = .5,linetype="dashed") +
  ggh4x::facet_wrap2(~id+condit, scales="free_y")

Code

all_fits |> 
   filter(id %in% unique(all_fits$id)[1:50], Fit_Method=="Test_Train") |>
    ggplot(aes(x=weight_exam,fill=condit)) +
  geom_density() +
  geom_vline(xintercept = .5,linetype="dashed") +
  ggh4x::facet_wrap2(~id+condit, scales="free_y")

Code

all_fits |> 
   filter(id %in% unique(all_fits$id)[1:50], Fit_Method=="Test") |>
  ggplot(aes(x=run_name,y=weight_exam,col=run_name)) + 
  stat_pointinterval(position=position_dodge(.5)) +
  ggh4x::facet_wrap2(~id+condit) + 
  theme(axis.text.x = element_blank())

Code

all_fits |> 
   filter(id %in% unique(all_fits$id)[1:50], Fit_Method=="Test_Train") |>
  ggplot(aes(x=run_name,y=weight_exam,col=run_name)) + 
  stat_pointinterval(position=position_dodge(.5)) +
  ggh4x::facet_wrap2(~id+condit) + 
  theme(axis.text.x = element_blank())

Code

all_fits |> 
  group_by(id,condit,Fit_Method) |>
  mutate(we_med=median(weight_exam),
         Best_Model=case_when(we_med>.5 ~"EXAM",we_med<.5 ~"ALM")) |>
  filter(Fit_Method=="Test") |>
  ungroup() |>
  mutate(id=reorder(id,we_med,decreasing = TRUE)) |>
  ggplot(aes(x=weight_exam,y=id,col=Best_Model)) + 
  geom_vline(xintercept = .5,linetype="dashed") +
  stat_pointinterval() + 
  ggh4x::facet_grid2(~condit,axes="all",scales="free_y", independent = "y")+
  theme_minimal() +
  theme(legend.position="top")

E3

Code

# folder_names <- c("e3_hybrid_n_iter_150_ntry_150_1757","e3_hybrid_n_iter_250_ntry_200_4300","e3_hybrid_n_iter_60_ntry_150_0946","e3_hybrid_n_iter_100_ntry_100_0545")
testE3 <- readRDS(here("data/e3_08-04-23.rds")) |> filter(expMode2 == "Test") 
e3Sbjs <- testE3 |> group_by(id,condit,bandOrder) |> summarise(n=n())

folder_names <- list.files(here("data/abc_reject"),pattern="e3_hyb*")

all_fits <- map_df(folder_names, process_folder) |> 
  left_join(e3Sbjs,by=c("id","condit")) 


all_fits |> 
  group_by(exp,n_int,ntry,tolM,ar,condit,Fit_Method) |> 
  summarise(min=first(min),me=mean(mean_error),
            w=median(weight_exam),sd_w=sd(weight_exam),c=median(c),
            lr=median(lr),n=n_distinct(id)) |>
   mutate(across(c(me, w, lr, sd_w), ~round(., 2))) |>
  arrange(condit,Fit_Method,me) |> kable(caption="E3 Fit Comparisons") |> 
  kable_styling(full_width = F) |>
  column_spec(8,bold=T,border_left=T)

E3 Fit Comparisons
exp	n_int	ntry	tolM	ar	condit	Fit_Method	min	me	w	sd_w	c	lr	n
e3	100	2500	0.65	0.10	Constant	Test	246	213.60	0.75	0.27	0.0000413	2.08	110
e3	200	1500	0.70	0.10	Constant	Test	348	214.91	0.74	0.27	0.0000417	2.08	110
e3	90	900	0.80	0.09	Constant	Test	92	217.05	0.73	0.27	0.0000413	2.04	110
e3	50	900	0.80	0.12	Constant	Test	53	217.53	0.73	0.27	0.0000406	2.02	110
e3	70	500	0.85	0.10	Constant	Test	186	219.42	0.72	0.27	0.0000397	2.06	110
e3	150	500	0.95	0.10	Constant	Test	96	219.56	0.72	0.27	0.0000397	2.04	110
e3	50	500	0.85	0.05	Constant	Test	312	220.20	0.72	0.27	0.0000394	2.06	110
e3	400	300	0.85	0.04	Constant	Test	353	220.22	0.72	0.27	0.0000403	2.04	110
e3	50	900	0.95	0.10	Constant	Test	37	220.50	0.72	0.27	0.0000387	2.10	110
e3	49	800	0.95	0.18	Constant	Test	303	221.29	0.72	0.27	0.0000390	2.10	110
e3	52	600	0.95	0.18	Constant	Test	228	221.74	0.72	0.28	0.0000387	2.04	110
e3	61	500	0.95	0.10	Constant	Test	36	221.78	0.71	0.27	0.0000381	2.02	110
e3	250	200	0.85	0.10	Constant	Test	184	222.77	0.72	0.28	0.0000395	2.06	110
e3	100	100	0.85	0.01	Constant	Test	804	223.08	0.71	0.27	0.0000407	2.08	110
e3	65	300	0.95	0.10	Constant	Test	31	224.28	0.71	0.27	0.0000391	2.03	110
e3	150	150	0.85	0.10	Constant	Test	92	224.50	0.72	0.28	0.0000396	2.03	110
e3	60	150	0.85	0.10	Constant	Test	38	225.40	0.72	0.28	0.0000373	2.07	110
e3	51	300	0.95	0.15	Constant	Test	162	225.64	0.72	0.27	0.0000380	2.01	110
e3	105	100	0.85	0.05	Constant	Test	351	225.99	0.71	0.28	0.0000388	2.04	110
e3	100	100	0.85	0.10	Constant	Test	202	227.29	0.72	0.28	0.0000383	2.07	110
e3	100	2500	0.65	0.10	Constant	Test_Train	244	230.55	0.73	0.27	0.0000384	0.60	110
e3	200	1500	0.70	0.10	Constant	Test_Train	480	231.40	0.71	0.27	0.0000372	0.60	110
e3	90	900	0.80	0.09	Constant	Test_Train	94	232.37	0.70	0.27	0.0000369	0.62	110
e3	50	900	0.80	0.12	Constant	Test_Train	58	232.77	0.70	0.28	0.0000365	0.63	110
e3	70	500	0.85	0.10	Constant	Test_Train	186	233.90	0.68	0.27	0.0000364	0.67	110
e3	50	500	0.85	0.05	Constant	Test_Train	224	234.15	0.68	0.28	0.0000352	0.69	110
e3	400	300	0.85	0.04	Constant	Test_Train	613	234.32	0.69	0.27	0.0000358	0.71	110
e3	150	500	0.95	0.10	Constant	Test_Train	106	234.35	0.69	0.27	0.0000358	0.70	110
e3	50	900	0.95	0.10	Constant	Test_Train	40	235.16	0.69	0.28	0.0000359	0.73	110
e3	49	800	0.95	0.18	Constant	Test_Train	263	235.38	0.69	0.28	0.0000358	0.70	110
e3	100	100	0.85	0.01	Constant	Test_Train	645	235.54	0.69	0.28	0.0000357	0.76	110
e3	250	200	0.85	0.10	Constant	Test_Train	289	235.70	0.68	0.28	0.0000351	0.77	110
e3	52	600	0.95	0.18	Constant	Test_Train	227	235.74	0.68	0.28	0.0000358	0.77	110
e3	61	500	0.95	0.10	Constant	Test_Train	48	236.01	0.68	0.28	0.0000355	0.74	110
e3	65	300	0.95	0.10	Constant	Test_Train	30	236.55	0.68	0.27	0.0000347	0.77	110
e3	150	150	0.85	0.10	Constant	Test_Train	92	236.60	0.67	0.28	0.0000345	0.81	110
e3	51	300	0.95	0.15	Constant	Test_Train	152	236.95	0.68	0.28	0.0000352	0.79	110
e3	60	150	0.85	0.10	Constant	Test_Train	23	237.40	0.67	0.28	0.0000340	0.85	110
e3	105	100	0.85	0.05	Constant	Test_Train	284	237.51	0.67	0.28	0.0000348	0.86	110
e3	100	100	0.85	0.10	Constant	Test_Train	167	237.92	0.67	0.28	0.0000341	0.86	110
e3	70	500	0.85	0.10	Constant	Train	70	213.90	0.50	0.29	0.0000236	1.56	110
e3	100	100	0.85	0.10	Constant	Train	77	214.41	0.50	0.29	0.0000195	1.48	110
e3	50	500	0.85	0.05	Constant	Train	78	214.84	0.50	0.29	0.0000225	1.57	110
e3	60	150	0.85	0.10	Constant	Train	16	215.59	0.51	0.29	0.0000208	1.48	110
e3	52	600	0.95	0.18	Constant	Train	74	216.80	0.50	0.29	0.0000286	1.65	110
e3	100	2500	0.65	0.10	Varied	Test	246	194.62	0.81	0.27	0.0000966	2.19	85
e3	200	1500	0.70	0.10	Varied	Test	348	196.04	0.80	0.26	0.0000974	2.19	85
e3	90	900	0.80	0.09	Varied	Test	92	197.98	0.79	0.26	0.0000914	2.18	85
e3	50	900	0.80	0.12	Varied	Test	53	198.23	0.78	0.27	0.0000913	2.23	85
e3	70	500	0.85	0.10	Varied	Test	186	200.61	0.78	0.27	0.0000958	2.15	85
e3	150	500	0.95	0.10	Varied	Test	96	200.72	0.78	0.26	0.0000980	2.18	85
e3	50	500	0.85	0.05	Varied	Test	312	200.88	0.78	0.27	0.0000973	2.20	85
e3	50	900	0.95	0.10	Varied	Test	37	201.12	0.78	0.27	0.0001014	2.15	85
e3	400	300	0.85	0.04	Varied	Test	353	201.41	0.78	0.26	0.0000998	2.17	85
e3	49	800	0.95	0.18	Varied	Test	303	201.72	0.78	0.27	0.0000950	2.25	85
e3	52	600	0.95	0.18	Varied	Test	228	202.47	0.77	0.27	0.0001007	2.18	85
e3	61	500	0.95	0.10	Varied	Test	36	203.09	0.77	0.26	0.0001052	2.20	85
e3	250	200	0.85	0.10	Varied	Test	184	203.72	0.77	0.27	0.0000987	2.15	85
e3	100	100	0.85	0.01	Varied	Test	804	204.11	0.77	0.26	0.0000951	2.16	85
e3	65	300	0.95	0.10	Varied	Test	31	204.23	0.77	0.26	0.0000958	2.17	85
e3	150	150	0.85	0.10	Varied	Test	92	205.29	0.76	0.26	0.0000935	2.15	85
e3	51	300	0.95	0.15	Varied	Test	162	205.69	0.76	0.26	0.0000900	2.15	85
e3	60	150	0.85	0.10	Varied	Test	38	206.85	0.77	0.27	0.0001061	2.13	85
e3	105	100	0.85	0.05	Varied	Test	351	207.63	0.76	0.26	0.0001001	2.13	85
e3	100	100	0.85	0.10	Varied	Test	202	207.79	0.76	0.26	0.0001005	2.15	85
e3	100	2500	0.65	0.10	Varied	Test_Train	244	236.31	0.83	0.28	0.0000443	0.72	85
e3	200	1500	0.70	0.10	Varied	Test_Train	480	236.94	0.82	0.28	0.0000421	0.74	85
e3	90	900	0.80	0.09	Varied	Test_Train	94	238.09	0.81	0.28	0.0000396	0.74	85
e3	50	900	0.80	0.12	Varied	Test_Train	58	238.16	0.81	0.28	0.0000393	0.72	85
e3	50	900	0.95	0.10	Varied	Test_Train	40	239.45	0.80	0.28	0.0000381	0.78	85
e3	70	500	0.85	0.10	Varied	Test_Train	186	239.45	0.79	0.28	0.0000375	0.76	85
e3	50	500	0.85	0.05	Varied	Test_Train	224	239.63	0.79	0.28	0.0000378	0.76	85
e3	150	500	0.95	0.10	Varied	Test_Train	106	239.73	0.79	0.28	0.0000383	0.80	85
e3	400	300	0.85	0.04	Varied	Test_Train	613	240.07	0.78	0.28	0.0000370	0.79	85
e3	49	800	0.95	0.18	Varied	Test_Train	263	240.16	0.79	0.28	0.0000398	0.78	85
e3	52	600	0.95	0.18	Varied	Test_Train	227	240.47	0.79	0.28	0.0000393	0.78	85
e3	61	500	0.95	0.10	Varied	Test_Train	48	240.80	0.78	0.27	0.0000377	0.81	85
e3	250	200	0.85	0.10	Varied	Test_Train	289	241.42	0.77	0.28	0.0000352	0.81	85
e3	100	100	0.85	0.01	Varied	Test_Train	645	241.44	0.77	0.28	0.0000360	0.80	85
e3	51	300	0.95	0.15	Varied	Test_Train	152	242.07	0.77	0.28	0.0000362	0.83	85
e3	65	300	0.95	0.10	Varied	Test_Train	30	242.42	0.78	0.27	0.0000372	0.83	85
e3	150	150	0.85	0.10	Varied	Test_Train	92	242.42	0.76	0.27	0.0000347	0.81	85
e3	60	150	0.85	0.10	Varied	Test_Train	23	242.93	0.77	0.27	0.0000355	0.83	85
e3	105	100	0.85	0.05	Varied	Test_Train	284	243.66	0.76	0.27	0.0000342	0.85	85
e3	100	100	0.85	0.10	Varied	Test_Train	167	244.18	0.74	0.28	0.0000342	0.85	85
e3	70	500	0.85	0.10	Varied	Train	70	231.82	0.50	0.29	0.0000210	0.82	85
e3	50	500	0.85	0.05	Varied	Train	78	232.06	0.51	0.29	0.0000213	0.82	85
e3	60	150	0.85	0.10	Varied	Train	16	233.78	0.50	0.29	0.0000218	0.80	85
e3	52	600	0.95	0.18	Varied	Train	74	234.14	0.50	0.29	0.0000219	0.81	85
e3	100	100	0.85	0.10	Varied	Train	77	234.53	0.50	0.29	0.0000211	0.81	85

Code

all_fits |> 
  group_by(exp,n_int,ntry,tolM,ar,condit,Fit_Method) |> 
  summarise(min=first(min),me=mean(mean_error),
            w=median(weight_exam),sd_w=sd(weight_exam),c=median(c),
            lr=median(lr),n=n_distinct(id)) |>
   mutate(across(c(me, w, lr, sd_w), ~round(., 2))) |>
  arrange(condit,Fit_Method,me) |> kable(caption="E2 Fit Comparisons") |> 
  kable_styling(full_width = F) |>
  column_spec(8,bold=T,border_left=T)

E2 Fit Comparisons
exp	n_int	ntry	tolM	ar	condit	Fit_Method	min	me	w	sd_w	c	lr	n
e3	100	2500	0.65	0.10	Constant	Test	246	213.60	0.75	0.27	0.0000413	2.08	110
e3	200	1500	0.70	0.10	Constant	Test	348	214.91	0.74	0.27	0.0000417	2.08	110
e3	90	900	0.80	0.09	Constant	Test	92	217.05	0.73	0.27	0.0000413	2.04	110
e3	50	900	0.80	0.12	Constant	Test	53	217.53	0.73	0.27	0.0000406	2.02	110
e3	70	500	0.85	0.10	Constant	Test	186	219.42	0.72	0.27	0.0000397	2.06	110
e3	150	500	0.95	0.10	Constant	Test	96	219.56	0.72	0.27	0.0000397	2.04	110
e3	50	500	0.85	0.05	Constant	Test	312	220.20	0.72	0.27	0.0000394	2.06	110
e3	400	300	0.85	0.04	Constant	Test	353	220.22	0.72	0.27	0.0000403	2.04	110
e3	50	900	0.95	0.10	Constant	Test	37	220.50	0.72	0.27	0.0000387	2.10	110
e3	49	800	0.95	0.18	Constant	Test	303	221.29	0.72	0.27	0.0000390	2.10	110
e3	52	600	0.95	0.18	Constant	Test	228	221.74	0.72	0.28	0.0000387	2.04	110
e3	61	500	0.95	0.10	Constant	Test	36	221.78	0.71	0.27	0.0000381	2.02	110
e3	250	200	0.85	0.10	Constant	Test	184	222.77	0.72	0.28	0.0000395	2.06	110
e3	100	100	0.85	0.01	Constant	Test	804	223.08	0.71	0.27	0.0000407	2.08	110
e3	65	300	0.95	0.10	Constant	Test	31	224.28	0.71	0.27	0.0000391	2.03	110
e3	150	150	0.85	0.10	Constant	Test	92	224.50	0.72	0.28	0.0000396	2.03	110
e3	60	150	0.85	0.10	Constant	Test	38	225.40	0.72	0.28	0.0000373	2.07	110
e3	51	300	0.95	0.15	Constant	Test	162	225.64	0.72	0.27	0.0000380	2.01	110
e3	105	100	0.85	0.05	Constant	Test	351	225.99	0.71	0.28	0.0000388	2.04	110
e3	100	100	0.85	0.10	Constant	Test	202	227.29	0.72	0.28	0.0000383	2.07	110
e3	100	2500	0.65	0.10	Constant	Test_Train	244	230.55	0.73	0.27	0.0000384	0.60	110
e3	200	1500	0.70	0.10	Constant	Test_Train	480	231.40	0.71	0.27	0.0000372	0.60	110
e3	90	900	0.80	0.09	Constant	Test_Train	94	232.37	0.70	0.27	0.0000369	0.62	110
e3	50	900	0.80	0.12	Constant	Test_Train	58	232.77	0.70	0.28	0.0000365	0.63	110
e3	70	500	0.85	0.10	Constant	Test_Train	186	233.90	0.68	0.27	0.0000364	0.67	110
e3	50	500	0.85	0.05	Constant	Test_Train	224	234.15	0.68	0.28	0.0000352	0.69	110
e3	400	300	0.85	0.04	Constant	Test_Train	613	234.32	0.69	0.27	0.0000358	0.71	110
e3	150	500	0.95	0.10	Constant	Test_Train	106	234.35	0.69	0.27	0.0000358	0.70	110
e3	50	900	0.95	0.10	Constant	Test_Train	40	235.16	0.69	0.28	0.0000359	0.73	110
e3	49	800	0.95	0.18	Constant	Test_Train	263	235.38	0.69	0.28	0.0000358	0.70	110
e3	100	100	0.85	0.01	Constant	Test_Train	645	235.54	0.69	0.28	0.0000357	0.76	110
e3	250	200	0.85	0.10	Constant	Test_Train	289	235.70	0.68	0.28	0.0000351	0.77	110
e3	52	600	0.95	0.18	Constant	Test_Train	227	235.74	0.68	0.28	0.0000358	0.77	110
e3	61	500	0.95	0.10	Constant	Test_Train	48	236.01	0.68	0.28	0.0000355	0.74	110
e3	65	300	0.95	0.10	Constant	Test_Train	30	236.55	0.68	0.27	0.0000347	0.77	110
e3	150	150	0.85	0.10	Constant	Test_Train	92	236.60	0.67	0.28	0.0000345	0.81	110
e3	51	300	0.95	0.15	Constant	Test_Train	152	236.95	0.68	0.28	0.0000352	0.79	110
e3	60	150	0.85	0.10	Constant	Test_Train	23	237.40	0.67	0.28	0.0000340	0.85	110
e3	105	100	0.85	0.05	Constant	Test_Train	284	237.51	0.67	0.28	0.0000348	0.86	110
e3	100	100	0.85	0.10	Constant	Test_Train	167	237.92	0.67	0.28	0.0000341	0.86	110
e3	70	500	0.85	0.10	Constant	Train	70	213.90	0.50	0.29	0.0000236	1.56	110
e3	100	100	0.85	0.10	Constant	Train	77	214.41	0.50	0.29	0.0000195	1.48	110
e3	50	500	0.85	0.05	Constant	Train	78	214.84	0.50	0.29	0.0000225	1.57	110
e3	60	150	0.85	0.10	Constant	Train	16	215.59	0.51	0.29	0.0000208	1.48	110
e3	52	600	0.95	0.18	Constant	Train	74	216.80	0.50	0.29	0.0000286	1.65	110
e3	100	2500	0.65	0.10	Varied	Test	246	194.62	0.81	0.27	0.0000966	2.19	85
e3	200	1500	0.70	0.10	Varied	Test	348	196.04	0.80	0.26	0.0000974	2.19	85
e3	90	900	0.80	0.09	Varied	Test	92	197.98	0.79	0.26	0.0000914	2.18	85
e3	50	900	0.80	0.12	Varied	Test	53	198.23	0.78	0.27	0.0000913	2.23	85
e3	70	500	0.85	0.10	Varied	Test	186	200.61	0.78	0.27	0.0000958	2.15	85
e3	150	500	0.95	0.10	Varied	Test	96	200.72	0.78	0.26	0.0000980	2.18	85
e3	50	500	0.85	0.05	Varied	Test	312	200.88	0.78	0.27	0.0000973	2.20	85
e3	50	900	0.95	0.10	Varied	Test	37	201.12	0.78	0.27	0.0001014	2.15	85
e3	400	300	0.85	0.04	Varied	Test	353	201.41	0.78	0.26	0.0000998	2.17	85
e3	49	800	0.95	0.18	Varied	Test	303	201.72	0.78	0.27	0.0000950	2.25	85
e3	52	600	0.95	0.18	Varied	Test	228	202.47	0.77	0.27	0.0001007	2.18	85
e3	61	500	0.95	0.10	Varied	Test	36	203.09	0.77	0.26	0.0001052	2.20	85
e3	250	200	0.85	0.10	Varied	Test	184	203.72	0.77	0.27	0.0000987	2.15	85
e3	100	100	0.85	0.01	Varied	Test	804	204.11	0.77	0.26	0.0000951	2.16	85
e3	65	300	0.95	0.10	Varied	Test	31	204.23	0.77	0.26	0.0000958	2.17	85
e3	150	150	0.85	0.10	Varied	Test	92	205.29	0.76	0.26	0.0000935	2.15	85
e3	51	300	0.95	0.15	Varied	Test	162	205.69	0.76	0.26	0.0000900	2.15	85
e3	60	150	0.85	0.10	Varied	Test	38	206.85	0.77	0.27	0.0001061	2.13	85
e3	105	100	0.85	0.05	Varied	Test	351	207.63	0.76	0.26	0.0001001	2.13	85
e3	100	100	0.85	0.10	Varied	Test	202	207.79	0.76	0.26	0.0001005	2.15	85
e3	100	2500	0.65	0.10	Varied	Test_Train	244	236.31	0.83	0.28	0.0000443	0.72	85
e3	200	1500	0.70	0.10	Varied	Test_Train	480	236.94	0.82	0.28	0.0000421	0.74	85
e3	90	900	0.80	0.09	Varied	Test_Train	94	238.09	0.81	0.28	0.0000396	0.74	85
e3	50	900	0.80	0.12	Varied	Test_Train	58	238.16	0.81	0.28	0.0000393	0.72	85
e3	50	900	0.95	0.10	Varied	Test_Train	40	239.45	0.80	0.28	0.0000381	0.78	85
e3	70	500	0.85	0.10	Varied	Test_Train	186	239.45	0.79	0.28	0.0000375	0.76	85
e3	50	500	0.85	0.05	Varied	Test_Train	224	239.63	0.79	0.28	0.0000378	0.76	85
e3	150	500	0.95	0.10	Varied	Test_Train	106	239.73	0.79	0.28	0.0000383	0.80	85
e3	400	300	0.85	0.04	Varied	Test_Train	613	240.07	0.78	0.28	0.0000370	0.79	85
e3	49	800	0.95	0.18	Varied	Test_Train	263	240.16	0.79	0.28	0.0000398	0.78	85
e3	52	600	0.95	0.18	Varied	Test_Train	227	240.47	0.79	0.28	0.0000393	0.78	85
e3	61	500	0.95	0.10	Varied	Test_Train	48	240.80	0.78	0.27	0.0000377	0.81	85
e3	250	200	0.85	0.10	Varied	Test_Train	289	241.42	0.77	0.28	0.0000352	0.81	85
e3	100	100	0.85	0.01	Varied	Test_Train	645	241.44	0.77	0.28	0.0000360	0.80	85
e3	51	300	0.95	0.15	Varied	Test_Train	152	242.07	0.77	0.28	0.0000362	0.83	85
e3	65	300	0.95	0.10	Varied	Test_Train	30	242.42	0.78	0.27	0.0000372	0.83	85
e3	150	150	0.85	0.10	Varied	Test_Train	92	242.42	0.76	0.27	0.0000347	0.81	85
e3	60	150	0.85	0.10	Varied	Test_Train	23	242.93	0.77	0.27	0.0000355	0.83	85
e3	105	100	0.85	0.05	Varied	Test_Train	284	243.66	0.76	0.27	0.0000342	0.85	85
e3	100	100	0.85	0.10	Varied	Test_Train	167	244.18	0.74	0.28	0.0000342	0.85	85
e3	70	500	0.85	0.10	Varied	Train	70	231.82	0.50	0.29	0.0000210	0.82	85
e3	50	500	0.85	0.05	Varied	Train	78	232.06	0.51	0.29	0.0000213	0.82	85
e3	60	150	0.85	0.10	Varied	Train	16	233.78	0.50	0.29	0.0000218	0.80	85
e3	52	600	0.95	0.18	Varied	Train	74	234.14	0.50	0.29	0.0000219	0.81	85
e3	100	100	0.85	0.10	Varied	Train	77	234.53	0.50	0.29	0.0000211	0.81	85

Code

{all_fits |> 
   filter(Fit_Method=="Test") |>
    ggplot(aes(x=weight_exam,fill=condit)) +geom_density(alpha=.5) } /{

all_fits |> 
   filter(Fit_Method=="Test") |>
    ggplot(aes(x=weight_exam,fill=condit)) +geom_density(alpha=.5) +
  facet_wrap(~run_name)
    }

Code

all_fits |> 
   filter(Fit_Method=="Test") |>
    ggplot(aes(x=weight_exam,fill=condit)) +geom_density(alpha=.5) + facet_wrap(~bandOrder)

Code

all_fits |> ggplot(aes(x=condit,y=weight_exam,col=condit)) + stat_pointinterval() +
  facet_wrap(bandOrder~Fit_Method)

Code

all_fits |> ggplot(aes(x=run_name,y=weight_exam,col=condit)) + stat_pointinterval() +
  facet_wrap(condit~Fit_Method)

Code

all_fits |> 
   filter(id %in% unique(all_fits$id)[1:50], Fit_Method=="Test") |>
    ggplot(aes(x=weight_exam,fill=condit)) +
  geom_density() +
  geom_vline(xintercept = .5,linetype="dashed") +
  ggh4x::facet_wrap2(~id+condit, scales="free_y")

Code

all_fits |> 
   filter(id %in% unique(all_fits$id)[1:50], Fit_Method=="Test_Train") |>
    ggplot(aes(x=weight_exam,fill=condit)) +
  geom_density() +
  geom_vline(xintercept = .5,linetype="dashed") +
  ggh4x::facet_wrap2(~id+condit, scales="free_y")

Code

all_fits |> 
   filter(id %in% unique(all_fits$id)[1:50], Fit_Method=="Test") |>
  ggplot(aes(x=run_name,y=weight_exam,col=run_name)) + 
  stat_pointinterval(position=position_dodge(.5)) +
  ggh4x::facet_wrap2(~id+condit) + 
  theme(axis.text.x = element_blank())

Code

all_fits |> 
   filter(id %in% unique(all_fits$id)[1:50], Fit_Method=="Test_Train") |>
  ggplot(aes(x=run_name,y=weight_exam,col=run_name)) + 
  stat_pointinterval(position=position_dodge(.5)) +
  ggh4x::facet_wrap2(~id+condit) + 
  theme(axis.text.x = element_blank())

Code

{all_fits |> 
  group_by(id,condit,Fit_Method) |>
  mutate(we_med=median(weight_exam),
         Best_Model=case_when(we_med>.5 ~"EXAM",we_med<.5 ~"ALM")) |>
  filter(Fit_Method=="Test") |>
  ungroup() |>
  mutate(id=reorder(id,we_med,decreasing = TRUE)) |>
  ggplot(aes(x=weight_exam,y=id,col=Best_Model)) + 
  geom_vline(xintercept = .5,linetype="dashed") +
  stat_pointinterval() + 
  ggh4x::facet_grid2(~condit,axes="all",scales="free_y", independent = "y")+
  theme_minimal() +
  theme(legend.position="top")} / 
  {all_fits |> 
  group_by(id,condit,Fit_Method,bandOrder) |>
  mutate(we_med=median(weight_exam),
         Best_Model=case_when(we_med>.5 ~"EXAM",we_med<.5 ~"ALM")) |>
  filter(Fit_Method=="Test") |>
  ungroup() |>
  mutate(id=reorder(id,we_med,decreasing = TRUE)) |>
  ggplot(aes(x=weight_exam,y=id,col=Best_Model)) + 
  geom_vline(xintercept = .5,linetype="dashed") +
  stat_pointinterval() + 
  ggh4x::facet_grid2(~bandOrder+condit,axes="all",scales="free_y", independent = "y")+
  theme_minimal() +
  theme(legend.position="top")}