e1_Analysis.qmd

---
title: E1 Extras
date: last-modified
categories: [Analysis, R]
page-layout: full
# fig-width: 15
# fig-height: 8
toc: false
---

```{r setup, echo=FALSE, include=FALSE}
pacman::p_load(tidyverse,lme4,emmeans,here,knitr,kableExtra,gt,gghalves,patchwork,ggforce,ggdist, modelsummary)
e1 <- readRDS(here("data/e1_08-21-23.rds"))
quietly( source(here::here("Functions", "packages.R")))

train <- e1 |> filter(expMode %in% c("train"))
trainAvg <- train %>% group_by(id, condit, vb, bandInt,trainStage) %>%
  summarise(nHits=sum(dist==0),vx=mean(vx),dist=mean(dist),sdist=mean(sdist),n=n(),Percent_Hit=nHits/n)
test <- e1 |> filter(expMode %in% c("test-Nf","test-train-nf"))
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)

```


::: {.column-page-right}
```{r}

plot_violin_condit <- function(fcondit, fvb, pvar) {

  full_range <- testAvg |> pull({{pvar}}) |> range(na.rm=TRUE)
  testAvg |>
    filter(condit == fcondit, vb==fvb) |> 
    ggplot(aes(.data[[pvar]])) +
    geom_density(fill = 'dodgerblue4') +
    theme_void() +
    coord_cartesian(xlim = full_range)
}

unique_combinations = testAvg |> ungroup() |>
  distinct(condit, vb)

testAvg |> group_by(condit,vb) |> 
  summarize(mean=mean(dist),sd=sd(dist)) |> 
  mutate(Distribution=vb) |> 
  gt() |>
  tab_spanner(
    label = 'Training Condit',
    columns = -condit
  ) |> 
  cols_label_with(fn = str_to_title) |> 
  fmt_number(decimals = 2) |> 
  cols_align('left', columns = condit) |> 
  text_transform(
    locations = cells_body(columns = 'Distribution'),
    fn = function(column) {
      map2(unique_combinations$condit, unique_combinations$vb,~plot_violin_condit(.x,.y,'vx')) |>
        ggplot_image(height = px(50), aspect_ratio = 3)
    }
  )

vars_to_plot <- c('dist', 'vx')

tables <- map(vars_to_plot, function(pvar) {
  testAvg |>
    group_by(condit, vb) |>
    summarize(mean = mean(.data[[pvar]]), sd = sd(.data[[pvar]])) |>
    mutate(Distribution = vb) |>
    gt() |>
    tab_spanner(
      label = 'Training Condit',
      columns = -condit
    ) |>
    cols_label_with(fn = str_to_title) |>
    fmt_number(decimals = 2) |>
    cols_align('left', columns = condit) |>
    text_transform(
      locations = cells_body(columns = 'Distribution'),
      fn = function(column) {
        map2(unique_combinations$condit, unique_combinations$vb, ~plot_violin_condit(.x, .y, pvar)) |>
          ggplot_image(height = px(50), aspect_ratio = 3)
      }
    )
})
#print(tables[[1]])
#print(tables[[2]])

library(gtExtras)
gt_two_column_layout(tables)


```

::: 


```{r fig.width=12, fig.height=11}


#https://modelsummary.com/articles/datasummary.html

datasummary(vx*vb ~ Mean + SD + Histogram, data = testAvg)
datasummary(vx*vb*condit ~ Mean + SD + Histogram, data = testAvg)

datasummary(vx*vb*condit ~ Mean + SD + Histogram, data = test)

datasummary_crosstab(vb ~ condit * tOrder, data = test)

datasummary_crosstab(result ~ condit,
                     statistic = 1 ~ Percent("col"),
                     data = test)

datasummary_crosstab(result ~ condit*vb,
                     statistic = 1 ~ Percent("col"),
                     data = test)


datasummary(vx+dist+vy ~ mean *condit*vb,
            data = e1)

datasummary(Heading("X Velocity")*vx+Heading("Abs. Deviation")*dist+vy ~ mean *condit*vb,
            data = e1)


datasummary(vx + dist ~ Factor(condit) * (mean + sd),
            data = test)

datasummary(vx + dist ~ Factor(condit)*vb * (mean + sd),
            data = test)

datasummary(vb*Factor(condit) * (mean + sd) ~ vx + dist,
            data = test)


tmp <- mtcars[, c("mpg", "hp")]

# create a list with individual variables
# remove missing and rescale
tmp_list <- lapply(tmp, na.omit)
tmp_list <- lapply(tmp_list, scale)

# create a table with `datasummary`
# add a histogram with column_spec and spec_hist
# add a boxplot with colun_spec and spec_box
emptycol = function(x) " "
datasummary(mpg + hp ~ Mean + SD + Heading("Boxplot") * emptycol + Heading("Histogram") * emptycol,
    output = "kableExtra",
    data = tmp) %>%
    kableExtra::column_spec(column = 4, image = spec_boxplot(tmp_list)) %>%
    kableExtra::column_spec(column = 5, image = spec_hist(tmp_list))

tmp_list <- lapply(test |> ungroup() |> select(vx,vb,condit), na.omit)
# scale only numeric columns - not all are numeric so can't just lapply

tmp_list <- map(tmp_list, ~if(is.numeric(.x)) scale(.x) else .x)

ts=test |> ungroup() |> select(vx,vb,condit) |> mutate(vxScale=scale(vx)[,1])

 datasummary(vxScale*vb ~  Mean + Histogram,
            data=ts, output = "kableExtra")  %>%
    kableExtra::column_spec(column = 4, image = spec_boxplot(ts))


datasummary(vx*vb ~  Mean + Median+ SD + Histogram,
            data=test, output = "markdown")  

datasummary(dist*vb ~  Mean + Median+ SD + Histogram,
            data=test, output = "markdown")  

datasummary(vx*vb*condit ~  Mean + Histogram,
            data=test, output = "kableExtra")  %>%
    kableExtra::column_spec(column = 5, image = spec_boxplot(tmp_list)) 

cap <- "Testing - No Feedback"
f <- (`Condit` = condit) ~ (` ` = vx) * (`Distribution` = Histogram) + vx * vb * ((`Avg.` = Mean)*Arguments(fmt='%.0f') + (`SD` = SD)*Arguments(fmt='%.0f'))

datasummary(f,
            data = test,
            output = 'gt',
            title = cap,
            notes = 'Artwork by @Thomas',
            sparse_header = TRUE) 




f <- (`Band` = vb) ~ (` ` = vx) * (`Distribution` = Histogram) + vx * condit * ((`Avg.` = Mean)*Arguments(fmt='%.0f') + (`SD` = SD)*Arguments(fmt='%.0f'))

datasummary(f,
            data = test,
            output = 'gt',
            title = cap,
            notes = 'Artwork by @Thomas',
            sparse_header = TRUE) 



f <- (`Band` = vb) ~ (` ` = vx) * (`Distribution` = Histogram) + vx * condit * expMode2 * ((`Avg.` = Mean)*Arguments(fmt='%.0f') + (`SD` = SD)*Arguments(fmt='%.0f'))

datasummary(f,
            data = e1,
            output = 'gt',
            sparse_header = TRUE) 


f <- (`Band` = vb)* expMode2 ~ vx * condit  * ((`Avg.` = Mean)*Arguments(fmt='%.0f') + (`SD` = SD)*Arguments(fmt='%.0f'))

datasummary(f,
            data = e1,
            output = 'gt',
            sparse_header = TRUE) 


f <- (`Band` = vb)* expMode2 ~ dist * condit  * ((`Avg.` = Mean)*Arguments(fmt='%.0f') )

datasummary(f,
            data = e1,
            output = 'gt',
            sparse_header = TRUE) 
```




```{r}
#| label: test vx table
#| tbl-cap: "Testing - No Feedback"
#| tbl-subcap: ["Constant Testing - X Velocity", "Varied Testing - X Velocity"]
#| layout-ncol: 2
#| echo: fenced

result <- test_summary_table(test, "vx", mfun = list(mean = mean, median = median, sd = sd))
result$constant #|> kable_styling(full_width = F)
result$varied #|> kable_styling(full_width = F)

```

<!-- ::: {layout="[ [1,4],[1,5],[1,10]]"} -->

```{r}
#| tbl-cap: "Mean Vx over blocks. Mean (Standard Error)"
#| tbl-cap-location: top
nb=5
vt1=e1 |> filter(expMode=="train") |>
  learn_curve_table(gt.train,vx,gw=Trial_Bin,groupVec=c(id,vb,condit),nbins=nb,prefix="Block_") |>
  rename("Band"=vb,"Group"=condit)
  
vt1 %>% kable() %>% add_header_above(c(" "=2, "Training Block "=1," " = ncol(vt1)-3))

# vt1 %>% gt() %>% tab_options(column_labels.background.color = "#176940",
#                 table.font.size = px(14))

```


# Training Performance

```{r}
#| fig-cap: "Mean Vx over training blocks"
#| fig-width: 10
#| fig-height: 9
#| tbl-cap: "test table"
#| eval: false
# vpt1=plotWithTable(vp1,vt1,arrange="V")
nb=5
vp1=e1 |> filter(expMode=="train") |> learn_curve_plot(gt.train,vx,condit,facet_var=vb,groupVec=c(gt.train,condit,tOrder,id,vb),nbins=nb)
vp1

#vp1 / gridExtra::tableGrob(vt1)
#vp1 / gt_temp(vt1)


```

```{r}
#| tbl-cap: "Deviation over blocks. Mean (Standard Error)"
#| tbl-cap-location: bottom
#| results: asis

# vt2=e1 |> filter(expMode=="train") |> 
#   learn_curve_table(gt.train,dist,gw=condit,groupVec=c(id,condit,vb),nbins=nb) %>%
#   rename("Block"=Trial_Bin)

vt2=e1 |> filter(expMode=="train") |> 
  learn_curve_table(gt.train,dist,gw=Trial_Bin,groupVec=c(id,condit,vb),nbins=nb,prefix="Block_") %>%
  rename("Band"=vb,"Group"=condit) 

```

```{r}
#| fig-cap: "Absolute Deviation from Band over training"
#| fig-width: 8
#| eval: false
#| 
vp2=e1 |> filter(expMode=="train") |> 
  learn_curve_plot(gt.train,dist,condit,facet_var=vb,groupVec=c(gt.train,condit,tOrder,id,vb),nbins=nb) 
vp2



```



```{r}
#| tbl-cap: "Signed Deviation over blocks. Mean (Standard Error)"
#| tbl-cap-location: bottom

vt3=e1 |> filter(expMode=="train") |> 
  learn_curve_table(gt.train,sdist,gw=Trial_Bin,groupVec=c(id,vb,condit),nbins=nb,prefix="Block_") |>
  rename("Band"=vb,"Group"=condit) 
vt3 %>% kable(format = "html",escape = FALSE)
```





```{r}
#| fig-cap: "Signed Deviation from Band over training"
#| fig-width: 8
#| eval: false
vp3=e1 |> filter(expMode=="train") |> learn_curve_plot(gt.train,sdist,condit,facet_var=vb,groupVec=c(gt.train,condit,tOrder,id,vb),nbins=nb) 
vp3

```


<!-- ::: -->

```{r}
vp4 <- e1 |> filter(expMode=="train") |> group_by(id) |>
  mutate(Trial_Bin = cut(gt.train,breaks = nb,include.lowest=TRUE,labels=FALSE)) |>
  group_by(id,Trial_Bin,condit, vb) |> summarise(nHits=sum(dist==0),n=n(),Percent_Hit=nHits/n) %>%
  learn_curve_plot2(Trial_Bin,Percent_Hit,color_var=condit,facet_var=vb,groupVec=c(id,vb,condit))



```


# 22
::: {.column-page-right}
```{r}
#| layout: [[65, 65],[-5], [1,550, 550]]
#| label: train tables2
#| echo: fenced
#| tbl-subcap: 
#|   - "Vx"
#|   - "Deviation from target"
#| eval: false
#| 
vt1 %>% kable(format = "html",escape = FALSE)
vt2 %>% kable(format = "html",escape = FALSE)

vp3+big_text()
vp4+big_text()

```
::: 


# 33
```{r}
#| layout: [[65, 65],[-5], [1,350, 350,10]]
#| label: train tables
#| echo: fenced
#| tbl-subcap: 
#|   - "Vx"
#|   - "Deviation from target"
#| column: screen-inset-right
#| eval: false
vt1 %>% kable(format = "html",escape = FALSE)
vt2 %>% kable(format = "html",escape = FALSE)

vp3
vp4
```


```{r}
#| label: tbl-example4
#| tbl-cap: "Example"
#| tbl-subcap: 
#|   - "Signed Distance"
#|   - "Percent Hit"
#| layout: [[80, 90],[2], [-60,650]]
#| echo: fenced
#| column: screen-inset-right
#| fig-width: 14
#| eval: false


vt2 %>% kable(format = "html",escape = FALSE) %>% kable_styling(font_size = 7)
vt3 %>% kable(format = "html",escape = FALSE)

vp4+vp3 + plot_layout(guides = "collect") & theme(legend.position = 'right')


```

```{r}
#| fig-width: 12
#| fig-height: 11
#| column: page-inset-right
#| eval: false
((vp1+big_text()+vp2+big_text())/(vp3+big_text()+vp4+big_text())) + plot_layout(guides = "collect") 



```



<!-- ```{r} -->
<!-- #| layout-ncol: 2 -->
<!-- #| fig-cap:  -->
<!-- #|   - "Speed and Stopping Distances of Cars" -->
<!-- #|   - "Vapor Pressure of Mercury as a Function of Temperature" -->

<!-- vp1 -->
<!-- vp2 -->
<!-- ``` -->

<!-- ```{r} -->
<!-- #| tbl-cap: "Training" -->
<!-- #| tbl-subcap:  -->
<!-- #|   - "Speed and Stopping Distances of Cars" -->
<!-- #|   - "Vapor Pressure of Mercury as a Function of Temperature" -->
<!-- #| layout-ncol: 2 -->
<!-- #| echo: fenced -->

<!-- vt1 %>% kable() -->
<!-- vt2 %>% kable() -->
<!-- ``` -->





## Testing


::: panel-tabset

### Varied vs. Constant
```{r}
#| fig-width: 8
e1 |> filter(expMode %in% c("test-Nf","test-train-nf")) |>  
  ggplot(aes(x = vb, y = dist,fill=condit)) +
    stat_summary(geom = "bar", position=position_dodge(), fun = mean) +
    stat_summary(geom = "errorbar", position=position_dodge(.9), fun.data = mean_se, width = .4, alpha = .7) 
```
### Dist
```{r}
e1 |> filter(expMode %in% c("test-Nf","test-train-nf")) |>  ggplot(aes(x = vb, y = dist,fill=condit)) +
    stat_summary(geom = "bar", position=position_dodge(), fun = mean) +
    stat_summary(geom = "errorbar", position=position_dodge(.9), fun.data = mean_se, width = .4, alpha = .7) 

# create a kable table to mirror plot of distance effects for vb and condit 


e1 |> filter(expMode %in% c("test-Nf","test-train-nf")) |> group_by(vb,condit) %>% 
  summarise(distMean=mean(dist),distSd=sd(dist)) %>% 
  mutate(meanLab=paste0("Mean=",round(distMean,0)),sdLab=paste0("Sd=",round(distSd,0))) %>% 
  mutate(sumStatLab=paste0(meanLab,"\n",sdLab)) %>% 
  select(vb,condit,sumStatLab) %>% 
  spread(condit,sumStatLab) %>% 
  kable(format = "html",escape = FALSE) %>% 
  kable_styling(font_size = 10)

```     




### Vx
```{r}

e1 |> filter(expMode %in% c("test-Nf","test-train-nf")) |>  ggplot(aes(x = vb, y = vx,fill=condit)) +
    stat_summary(geom = "bar", position=position_dodge(), fun = mean) +
    stat_summary(geom = "errorbar", position=position_dodge(.9), fun.data = mean_se, width = .4, alpha = .7) 

testAvg |>  ggplot(aes(x = vb, y = vx,fill=condit)) +
    stat_summary(geom = "bar", position=position_dodge(), fun = mean) +
    stat_summary(geom = "errorbar", position=position_dodge(.9), fun.data = mean_se, width = .4, alpha = .7) 


e1 |> filter(expMode %in% c("test-Nf","test-train-nf")) |> group_by(vb,condit) %>% 
  summarise(vxMean=mean(vx),vxSd=sd(vx)) %>% 
  mutate(meanLab=paste0("Mean=",round(vxMean,0)),sdLab=paste0("Sd=",round(vxSd,0))) %>% 
  mutate(sumStatLab=paste0(meanLab,"\n",sdLab)) %>% 
  select(vb,condit,sumStatLab) %>% 
  spread(condit,sumStatLab) %>% 
  kable(format = "html",escape = FALSE) %>% 
  kable_styling(font_size = 10)


```

### Signed Distance
```{r}

e1 |> filter(expMode %in% c("test-Nf","test-train-nf")) |>  ggplot(aes(x = vb, y = sdist,fill=condit)) +
    stat_summary(geom = "bar", position=position_dodge(), fun = mean) +
    stat_summary(geom = "errorbar", position=position_dodge(.9), fun.data = mean_se, width = .4, alpha = .7) 


testAvg |>  ggplot(aes(x = vb, y = sdist,fill=condit)) +
    stat_summary(geom = "bar", position=position_dodge(), fun = mean) +
    stat_summary(geom = "errorbar", position=position_dodge(.9), fun.data = mean_se, width = .4, alpha = .7) 

e1 |> filter(expMode %in% c("test-Nf","test-train-nf")) |> group_by(vb,condit) %>% 
  summarise(sdistMean=mean(sdist),sdistSd=sd(sdist)) %>% 
  mutate(meanLab=paste0("Mean=",round(sdistMean,0)),sdLab=paste0("Sd=",round(sdistSd,0))) %>% 
  mutate(sumStatLab=paste0(meanLab,"\n",sdLab)) %>% 
  select(vb,condit,sumStatLab) %>% 
  spread(condit,sumStatLab) %>% 
  kable(format = "html",escape = FALSE) %>% 
  kable_styling(font_size = 10)



```


### Percent Hit
```{r}

testAvg |>  ggplot(aes(x = vb, y = Percent_Hit,fill=condit)) +
    stat_summary(geom = "bar", position=position_dodge(), fun = mean) +
    stat_summary(geom = "errorbar", position=position_dodge(.9), fun.data = mean_se, width = .4, alpha = .7) 

testAvg |> group_by(vb,condit) %>% 
  summarise(Percent_HitMean=mean(Percent_Hit),Percent_HitSd=sd(Percent_Hit)) %>% 
  mutate(meanLab=paste0("Mean=",round(Percent_HitMean,3)),sdLab=paste0("Sd=",round(Percent_HitSd,2))) %>% 
  mutate(sumStatLab=paste0(meanLab,"\n",sdLab)) %>% 
  select(vb,condit,sumStatLab) %>% 
  spread(condit,sumStatLab) %>% 
  kable(format = "html",escape = FALSE) %>% 
  kable_styling(font_size = 10)


```



### Tables


#### BandType


```{r}

create_table(test, "vx")
create_table(test, "Percent_Hit")

```




####  Aggregation
```{r}

e1 %>%
  filter(expMode %in% c("test-Nf","test-train-nf")) %>%
  group_by(vb, condit, bandType) %>%
  summarise(distMean = mean(dist), distSd = sd(dist)) %>%
  mutate(
    meanLab = paste0("Mean=", round(distMean, 0)),
    sdLab = paste0("Sd=", round(distSd, 0))
  ) %>%
  mutate(sumStatLab = paste0(meanLab, "\n", sdLab)) %>%
  pivot_wider(
    id_cols = c(vb, bandType),
    names_from = condit,
    values_from = sumStatLab
  ) %>%
  arrange(vb, bandType) %>%
  kable(format = "html", escape = FALSE)

e1 |> filter(expMode %in% c("test-Nf","test-train-nf")) |> group_by(vb,condit) %>% 
  summarise(vxMean=mean(vx),vxSd=sd(vx)) %>% 
  mutate(meanLab=paste0("Mean=",round(vxMean,0)),sdLab=paste0("Sd=",round(vxSd,0))) %>% 
  mutate(sumStatLab=paste0(meanLab,"\n",sdLab)) %>% 
  select(vb,condit,sumStatLab) %>% 
  spread(condit,sumStatLab) %>% 
  kable(format = "html",escape = FALSE,caption = "Vx Mean") %>% 
  kable_styling(font_size = 10)

e1 |> filter(expMode %in% c("test-Nf","test-train-nf")) |> group_by(vb,condit) %>% 
  summarise(sdistMean=mean(sdist),sdistSd=sd(sdist)) %>% 
  mutate(meanLab=paste0("Mean=",round(sdistMean,0)),sdLab=paste0("Sd=",round(sdistSd,0))) %>% 
  mutate(sumStatLab=paste0(meanLab,"\n",sdLab)) %>% 
  select(vb,condit,sumStatLab) %>% 
  spread(condit,sumStatLab) %>% 
  kable(format = "html",escape = FALSE,caption = "Signed Deviation Mean") %>% 
  kable_styling(font_size = 10)

testAvg |> group_by(vb,condit) %>% 
  summarise(Percent_HitMean=mean(Percent_Hit),Percent_HitSd=sd(Percent_Hit)) %>% 
  mutate(meanLab=paste0("Mean=",round(Percent_HitMean,3)),sdLab=paste0("Sd=",round(Percent_HitSd,2))) %>% 
  mutate(sumStatLab=paste0(meanLab,"\n",sdLab)) %>% 
  select(vb,condit,sumStatLab) %>% 
  spread(condit,sumStatLab) %>% 
  kable(format = "html",escape = FALSE,caption = "Mean % Hit") %>% 
  kable_styling(font_size = 10)









```

#### Tables Aggregating by Id first
```{r}

e1 %>% group_by(id, vb, condit) %>%
  summarise(dist = mean(dist), .groups = 'drop') %>%
  group_by(vb, condit) %>%
  summarise(distMean = mean(dist), distSd = sd(dist), .groups = 'drop') %>%
  mutate(
    meanLab = paste0("Mean=", round(distMean, 0)),
    sdLab = paste0("Sd=", round(distSd, 0))
  ) %>%
  mutate(sumStatLab = paste0(meanLab, "\n", sdLab)) %>%
  select(vb, condit, sumStatLab) %>%
  spread(condit, sumStatLab) %>%
  kable(format = "html", escape = FALSE, caption = "Deviation Mean - Aggregate by Id") %>%
  kable_styling(font_size = 11)

e1 %>%
  filter(expMode %in% c("test-Nf", "test-train-nf")) %>%
  group_by(id, vb, condit) %>%
  summarise(vx = mean(vx), .groups = 'drop') %>%
  group_by(vb, condit) %>%
  summarise(vxMean = mean(vx), vxSd = sd(vx), .groups = 'drop') %>%
  mutate(
    meanLab = paste0("Mean=", round(vxMean, 0)),
    sdLab = paste0("Sd=", round(vxSd, 0))
  ) %>%
  mutate(sumStatLab = paste0(meanLab, "\n", sdLab)) %>%
  select(vb, condit, sumStatLab) %>%
  spread(condit, sumStatLab) %>%
  kable(format = "html", escape = FALSE, caption = "Vx Mean - Aggregate by Id") %>%
  kable_styling(font_size = 11)

e1 %>%
  filter(expMode %in% c("test-Nf", "test-train-nf")) %>%
  group_by(id, vb, condit) %>%
  summarise(sdist = mean(sdist), .groups = 'drop') %>%
  group_by(vb, condit) %>%
  summarise(sdistMean = mean(sdist), sdistSd = sd(sdist), .groups = 'drop') %>%
  mutate(
    meanLab = paste0("Mean=", round(sdistMean, 0)),
    sdLab = paste0("Sd=", round(sdistSd, 0))
  ) %>%
  mutate(sumStatLab = paste0(meanLab, "\n", sdLab)) %>%
  select(vb, condit, sumStatLab) %>%
  spread(condit, sumStatLab) %>%
  kable(format = "html", escape = FALSE, caption = "Signed Distance Mean - Aggregate by Id") %>%
  kable_styling(font_size = 11)

testAvg %>%
  group_by(id, vb, condit) %>%
  summarise(Percent_Hit = mean(Percent_Hit), .groups = 'drop') %>%
  group_by(vb, condit) %>%
  summarise(
    Percent_HitMean = mean(Percent_Hit),
    Percent_HitSd = sd(Percent_Hit),
    .groups = 'drop'
  ) %>%
  mutate(
    meanLab = paste0("Mean=", round(Percent_HitMean, 3)),
    sdLab = paste0("Sd=", round(Percent_HitSd, 2))
  ) %>%
  mutate(sumStatLab = paste0(meanLab, "\n", sdLab)) %>%
  select(vb, condit, sumStatLab) %>%
  spread(condit, sumStatLab) %>%
  kable(format = "html", escape = FALSE, caption = "Mean % Hit - Aggregate by Id") %>%
  kable_styling(font_size = 10)


```



#### Tables that also indicate bandType
```{r}

# Create the Constant table
constant_table <- e1 %>%
  filter(expMode %in% c("test-Nf", "test-train-nf")) %>%
  group_by(vb, bandType, condit) %>%
  summarise(distMean = mean(dist), distSd = sd(dist)) %>%
  filter(condit == "Constant") %>%
  mutate(
    meanLab = paste0("Mean=", round(distMean, 0)),
    sdLab = paste0("Sd=", round(distSd, 0)),
    sumStatLab = paste0(meanLab, "\n", sdLab)
  ) %>%
  select(vb, bandType, sumStatLab) %>%
  rename(Constant = sumStatLab)

# Create the Varied table
varied_table <- e1 %>%
  filter(expMode %in% c("test-Nf", "test-train-nf")) %>%
  group_by(vb, bandType, condit) %>%
  summarise(distMean = mean(dist), distSd = sd(dist)) %>%
  filter(condit == "Varied") %>%
  mutate(
    meanLab = paste0("Mean=", round(distMean, 0)),
    sdLab = paste0("Sd=", round(distSd, 0)),
    sumStatLab = paste0(meanLab, "\n", sdLab)
  ) %>%
  select(vb, bandType, sumStatLab) %>%
  rename(Varied = sumStatLab)

# Merge tables
final_table <- full_join(constant_table, varied_table, by = "vb")

# Create the table
final_table %>%
  kbl(digits = c(0, 0, 0, 0, 0),
      caption = "Data summary") %>%
  kable_minimal(full_width = FALSE,
      position = "left") %>%
  add_header_above(c("vb" = 1, "Constant" = 2, "Varied" = 2))


```



::: 





```{r}





```


```{r}



```














```{r}
#| fig-height: 11
#| fig-width: 12
#| eval: false
rectWidth=.4
vbRect<- e1 %>% group_by(vb) %>% 
  summarise(lowBound=first(bandInt),highBound=first(highBound)) %>% mutate(vbn=as.numeric(vb),
                  vbLag=vbn-rectWidth,vbLead=vbn+rectWidth)

e1 |> filter(expMode %in% c("test-Nf","test-train-nf")) |>  ggplot(aes(x = condit, y = vx,fill=vb)) + 
  ggdist::stat_halfeye()


vbRect<- e1 |> group_by(vb) |>
  summarise(lowBound=first(bandInt),highBound=first(highBound)) |> 
  mutate(vbn=as.numeric(vb), rectWidth=.2,
                  vbLag=vbn-rectWidth,vbLead=vbn+rectWidth)

e1 |> filter(expMode %in% c("test-Nf","test-train-nf")) |>  
  ggplot(aes(x = vb, y = vx,fill=vb)) + 
  ggdist::stat_halfeye(alpha=.5) + 
  geom_rect(data=vbRect,aes(xmin=vbLag,xmax=vbLead,ymin=lowBound,ymax=highBound,fill=vb),alpha=.3,inherit.aes = FALSE)+
  scale_y_continuous(expand=expansion(add=100),breaks=round(seq(0,2400,by=200),2)) +
  geom_segment(data=vbRect, aes(x=vbLag,xend=vbLead,y=highBound,yend=highBound),alpha=1,linetype="dashed")+
  geom_segment(data=vbRect, aes(x=vbLag,xend=vbLead,y=lowBound,yend=lowBound),alpha=1,linetype="dashed")+
  geom_text(data=vbRect,aes(x=vbLag-.03,y=lowBound+100,label=vb),angle=90,size=3.5,fontface="bold")


testAvg |>  
  ggplot(aes(x = vb, y = vx,fill=vb)) + 
  ggdist::stat_halfeye(alpha=.5,width=.7) + 
  geom_rect(data=vbRect,aes(xmin=vbLag,xmax=vbLead,ymin=lowBound,ymax=highBound,fill=vb),alpha=.3,inherit.aes = FALSE)+
  facet_wrap(~condit,ncol=1)+
  scale_y_continuous(expand=expansion(add=100),breaks=round(seq(0,2000,by=200),2)) +
  geom_segment(data=vbRect, aes(x=vbLag,xend=vbLead,y=highBound,yend=highBound),alpha=1,linetype="dashed",inherit.aes = FALSE)+
  geom_segment(data=vbRect, aes(x=vbLag,xend=vbLead,y=lowBound,yend=lowBound),alpha=1,linetype="dashed",inherit.aes = FALSE)+
  geom_text(data=vbRect,aes(x=vbLag-.03,y=lowBound+100,label=vb),angle=90,size=5.5,fontface="bold",inherit.aes = FALSE)  
  


# testAvg |>  
#   ggplot(aes(x = vb, y = sdist,fill=vb)) + 
#   ggdist::stat_halfeye(alpha=.5,width=.7) + 
#   facet_wrap(~condit,ncol=1)
# 
# 
# testAvg |>  
#   ggplot(aes(x = vb, y = dist,fill=vb)) + 
#   ggdist::stat_halfeye(alpha=.5,width=.7) + 
#   facet_wrap(~condit,ncol=1)
# 
# 


```



```{r}
#| fig-height: 11
#| fig-width: 12
#| eval: false
sumStats2 = test %>% group_by(id,condit,vb) %>%
  summarise(distMean=mean(dist),distMedian=median(dist),distSd=sd(dist)) %>%group_by(condit,vb) %>%
  summarise(groupMean=round(mean(distMean),0),groupMedian=round(mean(distMedian),0),groupSd=round(mean(distSd),0)) %>%
  mutate(meanLab=paste0("Mean=",groupMean),medianLab=paste0("Median=",groupMedian),sdLab=paste0("Sd=",groupSd)) %>%
  mutate(sumStatLab=paste0(meanLab,"\n",medianLab,"\n",sdLab))

sumStats = test %>% group_by(id,condit,vb) %>%
  summarise(distMean=mean(vx),distMedian=median(vx),distSd=sd(vx)) %>%group_by(condit,vb) %>%
  summarise(groupMean=round(mean(distMean),0),groupMedian=round(mean(distMedian),0),groupSd=round(mean(distSd),0)) %>%
  mutate(meanLab=paste0("Mean=",groupMean),medianLab=paste0("Median=",groupMedian),sdLab=paste0("Sd=",groupSd)) %>%
  mutate(sumStatLab=paste0(meanLab,"\n",medianLab,"\n",sdLab))

bandLines4 <- list(geom_segment(data=vbRect,aes(x=vbLag,xend=vbLead,y=highBound,yend=highBound),alpha=1,linetype="dashed"),
                   geom_segment(data=vbRect,aes(x=vbLag,xend=vbLead,y=lowBound,yend=lowBound),alpha=1,linetype="dashed"),
                   geom_text(data=vbRect,aes(x=vbLag-.03,y=lowBound+100,label=vb),angle=90,size=2.5,fontface="bold") )    


test %>% group_by(id,vb,condit) %>% 
  summarise(vxMean=mean(vx)) %>%
  ggplot(aes(x=vb,y=vxMean,fill=vb))+
  gghalves::geom_half_violin(color=NA)+ # remove border color
  gghalves::geom_half_boxplot(position=position_nudge(x=-0.05),side="r",outlier.shape = NA,center=TRUE, 
                    errorbar.draw = FALSE,width=.25)+
  gghalves::geom_half_point(transformation = position_jitter(width = 0.05, height = 0.05),size=.3,aes(color=vb))+
  facet_wrap(~condit,scale="free_x")+
  geom_rect(data=vbRect,aes(xmin=vbLag,xmax=vbLead,ymin=lowBound,ymax=highBound,fill=vb),alpha=.3,inherit.aes = FALSE)+
  bandLines4+
  #geom_text(data=sumStats,aes(x=vb,y=2100,label = groupMean),size=2, vjust = -0.5)+
  scale_y_continuous(expand=expansion(add=100),breaks=round(seq(0,2000,by=200),2))+
  theme(legend.position='none',
        plot.title=element_text(face="bold"),
        axis.title.x=element_text(face="bold"),
        axis.title.y=element_text(face="bold"),
        axis.text.x = element_text(size = 7.5))+
  ylab("Mean X Velocity")+xlab("Target Velocity Band") + 
  ggtitle("Testing Performance (no-feedback) - X-Velocity Per Band")+ 
  geom_text(data=sumStats2,aes(y=2090,label = sumStatLab),size=1.9)


#test %>% group_by(id,vb,condit) |> plot_distByCondit()


test %>% group_by(id,vb,condit) %>% 
  summarise(distMean=mean(dist)) %>% 
  ggplot(aes(x=vb,y=distMean,fill=vb))+
  geom_half_violin(color=NA)+ # remove border color
  geom_half_boxplot(position=position_nudge(x=-0.05),side="r",outlier.shape = NA,center=TRUE, 
                    errorbar.draw = FALSE,width=.25)+
  geom_half_point(transformation = position_jitter(width = 0.05, height = 0.05),size=.3,aes(color=vb))+
  facet_wrap(~condit,scale="free_x")+
  scale_y_continuous(expand=expansion(add=100),breaks=round(seq(0,2000,by=200),2))+
  theme(legend.position='none',
        plot.title=element_text(face="bold"),
        axis.title.x=element_text(face="bold"),
        axis.title.y=element_text(face="bold"),
        axis.text.x = element_text(size = 9.0))+
  ylab("Mean Distance From Boundary")+xlab("Target Velocity Band") + 
  ggtitle("Testing Performance (no-feedback) - Absolute Distance from Boundary")+ 
  geom_text(data=sumStats2,aes(y=1200,label = sumStatLab),size=3,fontface="bold")



```

```{r}
#| fig-height: 11
#| fig-width: 12
#| eval: false
test %>% group_by(sbjCode,condit,vb) %>% 
  summarise(distMean=mean(dist)) %>% 
  ggplot(aes(x=condit,y=distMean,fill=condit))+
  stat_summary(geom="bar",fun=mean,position=position_nudge(x=.21),alpha=.7,width=.2)+
   stat_summary(geom="errorbar",fun.data=mean_se,position=position_nudge(x=.21),alpha=.7,width=.1,)+
  geom_half_violin(position=position_dodge(.5),alpha=.55,color=NA)+ # remove border color
  geom_half_boxplot(position=position_dodge(.5),side="r",outlier.shape = NA,center=TRUE, 
                    errorbar.draw = FALSE,width=.25,alpha=.55)+
  geom_half_point(transformation = position_jitter(width = 0.05, height = 0.05),alpha=.2,size=.3)+
  facet_wrap(~vb,scale="free_x")+
  scale_y_continuous(expand=expansion(add=200),breaks=round(seq(0,2000,by=200),2))+
  theme(legend.position='none',
        plot.title=element_text(face="bold"),
        axis.title.x=element_text(face="bold"),
        axis.title.y=element_text(face="bold"),
        axis.text.x = element_text(size = 9.0))+
  ylab("Mean Distance From Boundary")+xlab("Training Condition") + 
  ggtitle("Testing Performance (no-feedback) - Absolute Distance from Boundary")+ 
  geom_text(data=sumStats2,aes(y=1390,label = sumStatLab),position=position_dodge(.5),size=3,fontface="bold")







```


```{r}





```






```{r}
#| fig-height: 13
#| fig-width: 12
library(ggforce)

testAvg %>% ggplot(aes(x=vx,y=condit,col=condit))+ ggdist::stat_halfeye()+facet_col(facets="vb")
testAvg %>% ggplot(aes(x=sdist,y=condit,col=condit))+ geom_boxplot()+facet_col(facets="vb")+geom_vline(xintercept=0)+theme_classic()

library(ggh4x)
library(ggdist)

testAvg %>% ggplot(aes(x=vx,y=vb,col=bandType)) +  ggdist::stat_halfeye(normalize="groups")+ facet_nested_wrap(vars(condit,tOrder),nrow=2)+theme_classic()

testAvg %>% ggplot(aes(x=vx,y=vb,col=bandType)) +  ggdist::stat_histinterval(normalize="groups")+ facet_nested_wrap(vars(condit,tOrder),nrow=2)+theme_classic()

testAvg %>% ggplot(aes(x=vx,y=vb,col=bandType)) +  ggdist::stat_histinterval(normalize="groups")+ facet_nested_wrap(vars(condit,tOrder),nrow=2)+theme_classic()


e1 |> group_by(id, condit, vb, bandInt,bandType,tOrder,expMode2) %>%
  summarise(vx=mean(vx),dist=mean(dist),sdist=mean(sdist)) %>%
  ggplot(aes(x=vx,y=condit,col=condit)) +  ggdist::stat_histinterval(normalize="groups")+ facet_nested_wrap(vars(vb,expMode2),nrow=6)+theme_classic()

e1 |> group_by(id, condit, vb, bandInt,bandType,tOrder,expMode2) %>%
  summarise(vx=mean(vx),dist=mean(dist),sdist=mean(sdist)) %>% ggplot(aes(x=sdist,y=vb,col=condit))+ stat_histinterval(normalize="groups",position=position_dodge())+facet_col(facets="expMode2")+geom_vline(xintercept=0)+theme_classic()

```


## Dictionary

| Variable Name | Variable Levels                 | Description                                                                               |
|---------------|---------------------------------|-------------------------------------------------------------------------------------------|
| condit        | Constant, Varied                | Condition of the experiment: constant or varied                                             |
| tOrder        | Test First, Train First         | Order of testing and training stages: test first or train first                            |
| expMode       | train, train-Nf, test-Nf, etc.  | Mode of the experiment: train, train-Nf, test-Nf, etc.                                    |
| trainStage    | Beginning, Middle, End, Test    | Stage of the training: beginning, middle, end, or test                                    |
| expStage      | TrainStart, intTest1, etc.      | Stage of the experiment: TrainStart, intTest1, TrainMid1, etc.                            |
| band          | 1, 2, 3, 4, 5, 6               | Band number                                                                                 |
| vb            | 100-300, 350-550, etc.         | Velocity band range                                                                         |
| lowBound      | 100, 350, 600, etc.            | Lower bound of the velocity band range                                                      |
| feedback      | 0, 1                           | Feedback type: 0 (no feedback), 1 (feedback)                                               |
| stage         | 1, 2, 3, etc.                  | Stage number of the experiment                                                              |

---






<!-- ::: panel-tabset -->
<!-- ## Figure -->
<!-- ```{r} -->
<!-- #| column: page-inset-right -->
<!-- #| fig-cap: "Mean Vx over training blocks" -->

<!-- # vpt1=plotWithTable(vp1,vt1,arrange="V") -->
<!-- nb=5 -->
<!-- vp1=e1 |> filter(expMode=="train") |> learn_curve_plot(gt.train,vx,condit,facet_var=vb,groupVec=c(gt.train,condit,tOrder,id,vb),nbins=nb) -->
<!-- vp1 -->

<!-- ``` -->


<!-- ## Table -->
<!-- ```{r} -->
<!-- #| tbl-cap: "Mean Vx over blocks. Mean (Standard Error)" -->
<!-- #| tbl-cap-location: top -->

<!-- vt1=e1 |> filter(expMode=="train") |> -->
<!--   learn_curve_table(gt.train,vx,gw=Trial_Bin,groupVec=c(vb,condit),nbins=nb,prefix="Block_") |> -->
<!--   rename("Band"=vb,"Group"=condit) -->
<!-- vt1 %>% kable() -->


<!-- ``` -->
<!-- ::: -->