-
Notifications
You must be signed in to change notification settings - Fork 0
/
nested-cv-ranger-kj.R
254 lines (175 loc) · 6.64 KB
/
nested-cv-ranger-kj.R
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
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
# Nested cross-validation for tuning and algorithm comparison
# Kuhn-Johnson method
# ranger
# Notes
# 1. *** Make sure the target column is last in dataframe ***
# 2. Using the Ranger Random Forest model function from the ranger package
# Sections
# 1. Set-Up
# 2. Error
# 3. Model Functions
# 4. Hyperparameter Grids
# 5. Functions used in the loops
# 6. Compare algorithms
####################################################
# Set-Up
####################################################
# texts me if an error occurs
options(error = function() {
library(RPushbullet)
pbPost("note", "Error", geterrmessage())
if(!interactive()) stop(geterrmessage())
})
# start MLflow server
sys::exec_background("mlflow server")
Sys.sleep(10)
library(tictoc)
tic()
set.seed(2019)
# simulated data; generates 10 multi-patterned, numeric predictors plus outcome variable
sim_data <- function(n) {
tmp <- mlbench::mlbench.friedman1(n, sd=1)
tmp <- cbind(tmp$x, tmp$y)
tmp <- as.data.frame(tmp)
names(tmp)[ncol(tmp)] <- "y"
tmp
}
# Use small data to tune and compare models
small_dat <- sim_data(100)
pacman::p_load(RPushbullet, glue, ranger, tidymodels, data.table, dtplyr, dplyr, furrr, mlflow)
# make explicit the name of the exeriement to record to
mlflow_set_experiment("ncv_duration")
plan(multiprocess)
ncv_dat_10 <- nested_cv(small_dat,
outside = vfold_cv(v = 10, repeats = 2),
inside = bootstraps(times = 25))
##################################
# Error Functions
##################################
error_FUN <- function(y_obs, y_hat){
y_obs <- unlist(y_obs)
y_hat <- unlist(y_hat)
Metrics::mae(y_obs, y_hat)
}
#####################################
# Model Functions
#####################################
# Random Forest
ranger_FUN <- function(params, analysis_set) {
mtry <- params$mtry[[1]]
trees <- params$trees[[1]]
model <- ranger::ranger(y ~ ., data = analysis_set, mtry = mtry, num.trees = trees)
model
}
# Elastic Net Regression
glm_FUN <- function(params, analysis_set) {
alpha <- params$mixture[[1]]
lambda <- params$penalty[[1]]
model <- parsnip::linear_reg(mixture = alpha, penalty = lambda) %>%
parsnip::set_engine("glmnet") %>%
generics::fit(y ~ ., data = analysis_set)
model
}
mod_FUN_list_ranger <- list(glmnet = glm_FUN, ranger = ranger_FUN)
###################################
# Hyperparameter Grids
###################################
# size = number of rows
# Default ranges look good for mixture and penalty
glm_params <- grid_latin_hypercube(
mixture(),
penalty(),
size = 100
)
rf_params <- grid_latin_hypercube(
mtry(range = c(3, 4)),
trees(range = c(200, 300)),
size = 100
)
params_list <- list(glmnet = glm_params, ranger = rf_params)
#####################################################
# Functions used in the loops
#####################################################
# inputs params, model, and resample, calls model and error functions, outputs error
mod_error <- function(params, mod_FUN, dat) {
y_col <- ncol(dat$data)
y_obs <- assessment(dat)[y_col]
mod <- mod_FUN(params, analysis(dat))
pred <- predict(mod, assessment(dat))
if (!is.data.frame(pred)) {
pred <- pred$predictions
}
error <- error_FUN(y_obs, pred)
error
}
# inputs resample, loops hyperparam grid values to model/error function, collects error value for hyperparam combo
tune_over_params <- function(dat, mod_FUN, params) {
params$error <- map_dbl(1:nrow(params), function(row) {
params <- params[row,]
mod_error(params, mod_FUN, dat)
})
params
}
# inputs and sends fold's resamples to tuning function, collects and averages fold's error for each hyperparameter combo
summarize_tune_results <- function(dat, mod_FUN, params) {
# Return row-bound tibble that has the 25 bootstrap results
param_names <- names(params)
future_map_dfr(dat$splits, tune_over_params, mod_FUN, params, .progress = TRUE) %>%
lazy_dt(., key_by = param_names) %>%
# For each value of the tuning parameter, compute the
# average <error> which is the inner bootstrap estimate.
group_by_at(vars(param_names)) %>%
summarize(mean_error = mean(error, na.rm = TRUE),
n = length(error)) %>%
as_tibble()
}
######################################################
# Compare algorithms
######################################################
compare_algs <- function(mod_FUN, params, ncv_dat){
# tune models by grid searching on resamples in the inner loop (e.g. 5 repeats 10 folds = list of 50 tibbles with param and mean_error cols)
tuning_results <- map(ncv_dat$inner_resamples, summarize_tune_results, mod_FUN, params)
# Choose best hyperparameter combos across all the resamples for each fold (e.g. 5 repeats 10 folds = 50 best hyperparam combos)
best_hyper_vals <- tuning_results %>%
map_df(function(dat) {
dat[which.min(dat$mean_error),]
}) %>%
select(names(params))
# fit models on the outer-loop folds using best hyperparams (e.g. 5 repeats, 10 folds = 50 models)
outer_fold_error <- future_map2_dbl(ncv_dat$splits, 1:nrow(best_hyper_vals), function(dat, row) {
params <- best_hyper_vals[row,]
mod_error(params, mod_FUN, dat)
}, .progress = TRUE)
# hyperparam values for final model will be the ones most selected to use on the outer-loop folds
chosen_params <- best_hyper_vals %>%
group_by_all() %>%
tally() %>%
ungroup() %>%
filter(n == max(n))
# output error stats and chosen hyperparams
tibble(
chosen_params = list(chosen_params),
mean_error = mean(outer_fold_error),
median_error = median(outer_fold_error),
sd_error = sd(outer_fold_error)
)
}
# Start the nested-cv
algorithm_comparison_ten_rang <- map2_dfr(mod_FUN_list_ranger, params_list, compare_algs, ncv_dat_10) %>%
mutate(model = names(mod_FUN_list_ranger)) %>%
select(model, everything())
toc(log = TRUE)
# log duration metric to MLflow
ncv_times <- tic.log(format = FALSE)
duration <- as.numeric(ncv_times[[1]]$toc - ncv_times[[1]]$tic)
mlflow_log_metric("duration", duration)
mlflow_set_tag("implementation", "ranger")
mlflow_set_tag("method", "kj")
mlflow_end_run()
# text me results
log.txt <- tic.log(format = TRUE)
msg <- glue("Using ranger-kj script: \n After running 10 fold rang, {log.txt[[1]]}")
pbPost("note", title="ranger-kj script finished", body=msg)
tic.clearlog()
# MLflow uses waitress for Windows. Killing it also kills mlflow.exe, python.exe, console window host processes
installr::kill_process(process = c("waitress-serve.exe"))