-
Notifications
You must be signed in to change notification settings - Fork 0
/
README.Rmd
136 lines (107 loc) · 4.55 KB
/
README.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
---
title: "tornado"
output:
md_document:
variant: gfm
---
<!-- DO NOT EDIT: README.md is created from README.Rmd -->
<div>
<table>
<tr>
<td>
<img align="left" width="200" height="200" src="man/figures/logo.png"/>
</td>
</tr>
</table>
</div>
<footer>© Copyright 2024 Robert Carnell</footer>
```{r setup, echo=FALSE, include=FALSE}
require(knitr)
require(survival)
require(caret)
require(randomForest)
require(tornado)
knitr::opts_chunk$set(fig.path = "man/figures/")
```
```{r badges, echo=FALSE, results="asis"}
cat(paste0("|<sub>",
paste0(
c("Actions", "Code Coverage", "Website", "Downloads", "CRAN"),
collapse="</sub>|<sub>"),
"</sub>|\n"))
cat(paste0("|", paste0(rep(":---:", 5), collapse="|"), "|\n"))
cat(paste0("|",
paste0(c("[](https://github.com/bertcarnell/tornado/actions/workflows/r-cmd-check.yml)",
"[](https://codecov.io/github/bertcarnell/tornado?branch=master)",
"[](https://bertcarnell.github.io/tornado/)",
"[](https://cran.r-project.org/package=tornado)",
"[](https://cran.r-project.org/package=tornado)"),
collapse="|"),
"|"))
```
# tornado
`tornado` plots for model sensitivity analysis.
[website](https://bertcarnell.github.io/tornado/)
## Installation
You can also install the development version of `tornado` from github with:
```{r install2, echo=TRUE, eval=FALSE}
if (!require(devtools)) install.packages("devtools")
devtools::install_github("bertcarnell/tornado")
```
## Quick Start - Torndao Plots
### Linear Models
```{r random, echo=TRUE}
lm1 <- lm(mpg ~ cyl*wt*hp, data = mtcars)
torn1 <- tornado(lm1, type = "PercentChange", alpha = 0.10)
plot(torn1, xlabel = "MPG", geom_bar_control = list(width = 0.5))
```
### Generalized Linear Models
Plots can also be created for the range of the input data
```{r glm_section, echo=TRUE}
glm2 <- glm(vs ~ wt + disp + cyl, data = mtcars, family = binomial(link = "logit"))
torn2 <- tornado(glm2, type = "ranges", alpha = 0.10)
plot(torn2, xlabel = "Probability of Straight Engine", geom_bar_control = list(width = 0.5))
```
Plots can include factors and percentiles of the data as well.
```{r glm_factor_section, echo=TRUE}
mydata <- mtcars
mydata$am <- factor(mtcars$am)
mydata$vs <- factor(mtcars$vs)
glm2 <- glm(mpg ~ wt + cyl + am + vs, data = mydata, family = gaussian)
torn2 <- tornado(glm2, type = "percentiles", alpha = 0.10)
plot(torn2, xlabel = "MPG",
geom_bar_control = list(width = 0.5),
sensitivity_colors = c("#E41A1C", "#377EB8"),
geom_point_control = list(fill = "red", col = "red", size = 3))
```
### Censored Data
Accelerated Failure Model
```{r censored_section, echo=TRUE}
survreg3 <- survival::survreg(survival::Surv(futime, fustat) ~ ecog.ps + rx + age + resid.ds,
survival::ovarian, dist = 'weibull', scale = 1)
torn3 <- tornado(survreg3, modeldata = survival::ovarian, type = "PercentChange", alpha = 0.10)
plot(torn3, xlabel = "Survival Time", geom_bar_control = list(width = 0.5))
```
Cox Proportional Hazard Model with 2 Standard Deviation bars
```{r censored_coxph_setion, echo=TRUE}
coxph4 <- survival::coxph(survival::Surv(stop, event) ~ rx + size + number,
survival::bladder)
torn4 <- tornado(coxph4, modeldata = survival::bladder, type = "StdDev",
alpha = 2)
plot(torn4, xlabel = "Risk", geom_bar_control = list(width = 0.5))
```
### Machine Learning Models from caret::train
The `plot` method can also return a ggplot object
```{r train_section, echo = TRUE}
mydat <- mtcars
mydat$am <- factor(mydat$am, labels = c("A", "M"))
mydat$vs <- factor(mydat$vs, labels = c("V", "S"))
mydat$cyl <- factor(mydat$cyl)
mydat$gear <- factor(mydat$gear)
mydat$carb <- factor(mydat$carb)
gtest <- caret::train(x = subset(mydat, select = -vs), y = mydat$vs, method = "rf")
torn <- tornado(gtest, type = "percentiles", alpha = 0.10, class_number = 1)
g <- plot(torn, plot = FALSE, xlabel = "Probability of Class V-shaped Engine", geom_bar_control = list(width = 0.5))
g <- g + ggtitle("Classifier caret::train randomforest, 90-110% of the mean in each variable")
plot(g)
```