mgaughan/mw-lifecycle-analysis
1
0
Fork 0
Code

preliminary re-aggregstion of DSL df, preliminary drafting of DSL model

Browse Source
This commit is contained in:
Matthew Gaughan 2025-11-17 09:15:37 -08:00
parent 7555259a3e
commit fb490e37f5
4 changed files with 3222 additions and 77 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3130
111725_DSL_frame.csv Normal file
View File

File diff suppressed because it is too large Load Diff

12
dsl/dsl.R
View File

@ -1,7 +1,7 @@
library(tidyverse)
library(dsl)
dsl_csv <-"~/dsl/102725_DSL_df_adac.csv"
dsl_csv <-"111725_DSL_frame.csv"
dsl_df <- read.csv(dsl_csv, header = TRUE)
@ -22,19 +22,17 @@ power_model <- power_dsl(
summary(power_model)
plot(power_model, coef_name = "human_SOL_prop_adac")
dsl_df <- dsl_df |>
filter(source=="c1")
trial_model <- dsl(
model = "logit",
formula = dsl_score ~ human_BI_prop_adac +
formula = dsl_score ~ human_TSOL_prop_adac +
median_gerrit_loc_delta + median_gerrit_reviewers +
as.factor(isAuthorWMF) +
as.factor(author_closer) +
median_PC4_adac +
week_index,
predicted_var = "human_BI_prop_adac",
prediction = "olmo_BI_prop_adac",
week_index + n_comments_before,
predicted_var = "human_TSOL_prop_adac",
prediction = "olmo_TSOL_prop_adac",
sample_prob = "sampling_prob",
data=dsl_df
)

140
dsl/dsl_aggregation.R
View File

@ -13,7 +13,9 @@ valid_categories <- c('EXPECTED BEHAVIOR', 'MOTIVATION','OBSERVED BUG BEHAVIOR',
'POTENTIAL NEW ISSUES AND REQUESTS', 'SOLUTION USAGE',
'WORKAROUNDS', 'ISSUE CONTENT MANAGEMENT', 'ACTION ON ISSUE',
'SOCIAL CONVERSATION')
library(dplyr)
library(purrr)
library(stringr)
human_list_unified_df <- unified_df %>%
filter(!is.na(human_labels)) |>
mutate(human_labels = tidyr::replace_na(human_labels, "")) |>
@ -26,85 +28,79 @@ human_list_unified_df <- unified_df %>%
.x
}
})) %>%
unnest(list_human_labels, keep_empty = TRUE) |>
tidyr::unnest(list_human_labels, keep_empty = TRUE) |>
filter(list_human_labels != "NA") |>
group_by(TaskPHID) |>
summarise(
# Overall proportions (all comments)
n_tags = sum(!is.na(list_human_labels)),
human_BE_prop = if_else(
human_EP_prop = if_else(
n_tags == 0L,
NA_real_,
mean(list_human_labels %in% c("OBSERVED BUG BEHAVIOR", "EXPECTED BEHAVIOR"), na.rm = TRUE)
mean(list_human_labels %in% c("OBSERVED BUG BEHAVIOR", "BUG REPRODUCTION", "INVESTIGATION AND EXPLORATION"), na.rm = TRUE)
),
human_SOL_prop = if_else(
human_TSOL_prop = if_else(
n_tags == 0L,
NA_real_,
mean(list_human_labels %in% c("SOLUTION DISCUSSION", "SOLUTION USAGE"), na.rm = TRUE)
mean(list_human_labels %in% c("SOLUTION DISCUSSION", "SOLUTION USAGE", "TESTING", "WORKAROUNDS", "TASK PROGRESS"), na.rm = TRUE)
),
human_VR_prop = if_else(
human_DIO_prop = if_else(
n_tags == 0L,
NA_real_,
mean(list_human_labels %in% c("OBSERVED BUG BEHAVIOR", "EXPECTED BEHAVIOR",
"SOLUTION DISCUSSION", "SOLUTION USAGE",
"INVESTIGATION AND EXPLORATION"), na.rm = TRUE)
mean(list_human_labels %in% c("MOTIVATION", "EXPECTED BEHAVIOR"), na.rm = TRUE)
),
human_BI_prop = if_else(
human_RK_prop = if_else(
n_tags == 0L,
NA_real_,
mean(list_human_labels %in% c("BUG REPRODUCTION",
"INVESTIGATION AND EXPLORATION"), na.rm = TRUE)
mean(list_human_labels %in% c("ACTION ON ISSUE",
"ISSUE CONTENT MANAGEMENT"), na.rm = TRUE)
),
# ADAC==1 proportions
n_tags_adac = sum(!is.na(list_human_labels) & ADAC == 1),
human_BE_prop_adac = if_else(
human_EP_prop_adac = if_else(
n_tags_adac == 0L,
NA_real_,
mean(list_human_labels[ADAC == 1] %in% c("OBSERVED BUG BEHAVIOR", "EXPECTED BEHAVIOR"), na.rm = TRUE)
mean(list_human_labels[ADAC == 1] %in% c("OBSERVED BUG BEHAVIOR", "BUG REPRODUCTION", "INVESTIGATION AND EXPLORATION"), na.rm = TRUE)
),
human_SOL_prop_adac = if_else(
human_TSOL_prop_adac = if_else(
n_tags_adac == 0L,
NA_real_,
mean(list_human_labels[ADAC == 1] %in% c("SOLUTION DISCUSSION", "SOLUTION USAGE"), na.rm = TRUE)
mean(list_human_labels[ADAC == 1] %in% c("SOLUTION DISCUSSION", "SOLUTION USAGE", "TESTING", "WORKAROUNDS", "TASK PROGRESS"), na.rm = TRUE)
),
human_VR_prop_adac = if_else(
human_DIO_prop_adac = if_else(
n_tags_adac == 0L,
NA_real_,
mean(list_human_labels[ADAC == 1] %in% c("OBSERVED BUG BEHAVIOR", "EXPECTED BEHAVIOR",
"SOLUTION DISCUSSION", "SOLUTION USAGE",
"INVESTIGATION AND EXPLORATION"), na.rm = TRUE)
mean(list_human_labels[ADAC == 1] %in% c("MOTIVATION", "EXPECTED BEHAVIOR"), na.rm = TRUE)
),
human_BI_prop_adac = if_else(
human_RK_prop_adac = if_else(
n_tags_adac == 0L,
NA_real_,
mean(list_human_labels[ADAC == 1] %in% c("BUG REPRODUCTION",
"INVESTIGATION AND EXPLORATION"), na.rm = TRUE)
mean(list_human_labels[ADAC == 1] %in% c("ACTION ON ISSUE",
"ISSUE CONTENT MANAGEMENT"), na.rm = TRUE)
),
# ADAC==0 proportions
n_tags_no_adac = sum(!is.na(list_human_labels) & ADAC == 0),
human_BE_prop_no_adac = if_else(
human_EP_prop_no_adac = if_else(
n_tags_no_adac == 0L,
NA_real_,
mean(list_human_labels[ADAC == 0] %in% c("OBSERVED BUG BEHAVIOR", "EXPECTED BEHAVIOR"), na.rm = TRUE)
mean(list_human_labels[ADAC == 0] %in% c("OBSERVED BUG BEHAVIOR", "BUG REPRODUCTION", "INVESTIGATION AND EXPLORATION"), na.rm = TRUE)
),
human_SOL_prop_no_adac = if_else(
human_TSOL_prop_no_adac = if_else(
n_tags_no_adac == 0L,
NA_real_,
mean(list_human_labels[ADAC == 0] %in% c("SOLUTION DISCUSSION", "SOLUTION USAGE"), na.rm = TRUE)
mean(list_human_labels[ADAC == 0] %in% c("SOLUTION DISCUSSION", "SOLUTION USAGE", "TESTING", "WORKAROUNDS", "TASK PROGRESS"), na.rm = TRUE)
),
human_VR_prop_no_adac = if_else(
human_DIO_prop_no_adac = if_else(
n_tags_no_adac == 0L,
NA_real_,
mean(list_human_labels[ADAC == 0] %in% c("OBSERVED BUG BEHAVIOR", "EXPECTED BEHAVIOR",
"SOLUTION DISCUSSION", "SOLUTION USAGE",
"INVESTIGATION AND EXPLORATION"), na.rm = TRUE)
mean(list_human_labels[ADAC == 0] %in% c("MOTIVATION", "EXPECTED BEHAVIOR"), na.rm = TRUE)
),
human_BI_prop_no_adac = if_else(
human_RK_prop_no_adac = if_else(
n_tags_no_adac == 0L,
NA_real_,
mean(list_human_labels[ADAC == 0] %in% c("BUG REPRODUCTION",
"INVESTIGATION AND EXPLORATION"), na.rm = TRUE)
mean(list_human_labels[ADAC == 0] %in% c("ACTION ON ISSUE",
"ISSUE CONTENT MANAGEMENT"), na.rm = TRUE)
),
.groups = "drop"
) |>
@ -117,7 +113,7 @@ olmo_list_unified_df <- unified_df %>%
olmo_sentence_labels,
"(?<=')[^']+(?=')|(?<=\")[^\"]+(?=\")"
)) %>%
unnest(list_olmo_labels, keep_empty = TRUE) |>
tidyr::unnest(list_olmo_labels, keep_empty = TRUE) |>
filter(!is.na(list_olmo_labels), list_olmo_labels != "", list_olmo_labels != "NA") %>%
mutate(list_olmo_labels = str_squish(str_replace_all(list_olmo_labels, "[^\\p{L} ]+", ""))) %>%
filter(list_olmo_labels != "") %>%
@ -134,76 +130,70 @@ olmo_list_unified_df <- unified_df %>%
summarise(
# Overall proportions (all comments)
n_tags = sum(!is.na(olmo_label)),
olmo_BE_prop = if_else(
olmo_EP_prop = if_else(
n_tags == 0L,
NA_real_,
mean(olmo_label %in% c("OBSERVED BUG BEHAVIOR", "EXPECTED BEHAVIOR"), na.rm = TRUE)
mean(olmo_label %in% c("OBSERVED BUG BEHAVIOR", "BUG REPRODUCTION", "INVESTIGATION AND EXPLORATION"), na.rm = TRUE)
),
olmo_SOL_prop = if_else(
olmo_TSOL_prop = if_else(
n_tags == 0L,
NA_real_,
mean(olmo_label %in% c("SOLUTION DISCUSSION", "SOLUTION USAGE"), na.rm = TRUE)
mean(olmo_label %in% c("SOLUTION DISCUSSION", "SOLUTION USAGE", "TESTING", "WORKAROUNDS", "TASK PROGRESS"), na.rm = TRUE)
),
olmo_VR_prop = if_else(
olmo_DIO_prop = if_else(
n_tags == 0L,
NA_real_,
mean(olmo_label %in% c("OBSERVED BUG BEHAVIOR", "EXPECTED BEHAVIOR",
"SOLUTION DISCUSSION", "SOLUTION USAGE",
"INVESTIGATION AND EXPLORATION"), na.rm = TRUE)
mean(olmo_label %in% c("MOTIVATION", "EXPECTED BEHAVIOR"), na.rm = TRUE)
),
olmo_BI_prop = if_else(
olmo_RK_prop = if_else(
n_tags == 0L,
NA_real_,
mean(olmo_label %in% c("BUG REPRODUCTION",
"INVESTIGATION AND EXPLORATION"), na.rm = TRUE)
mean(olmo_label %in% c("ACTION ON ISSUE",
"ISSUE CONTENT MANAGEMENT"), na.rm = TRUE)
),
n_tags_adac = sum(!is.na(olmo_label) & ADAC == 1),
olmo_BE_prop_adac = if_else(
olmo_EP_prop_adac = if_else(
n_tags_adac == 0L,
NA_real_,
mean(olmo_label[ADAC == 1] %in% c("OBSERVED BUG BEHAVIOR", "EXPECTED BEHAVIOR"), na.rm = TRUE)
mean(olmo_label[ADAC == 1] %in% c("OBSERVED BUG BEHAVIOR", "BUG REPRODUCTION", "INVESTIGATION AND EXPLORATION"), na.rm = TRUE)
),
olmo_SOL_prop_adac = if_else(
olmo_TSOL_prop_adac = if_else(
n_tags_adac == 0L,
NA_real_,
mean(olmo_label[ADAC == 1] %in% c("SOLUTION DISCUSSION", "SOLUTION USAGE"), na.rm = TRUE)
mean(olmo_label[ADAC == 1] %in% c("SOLUTION DISCUSSION", "SOLUTION USAGE", "TESTING", "WORKAROUNDS", "TASK PROGRESS"), na.rm = TRUE)
),
olmo_VR_prop_adac = if_else(
olmo_DIO_prop_adac = if_else(
n_tags_adac == 0L,
NA_real_,
mean(olmo_label[ADAC == 1] %in% c("OBSERVED BUG BEHAVIOR", "EXPECTED BEHAVIOR",
"SOLUTION DISCUSSION", "SOLUTION USAGE",
"INVESTIGATION AND EXPLORATION"), na.rm = TRUE)
mean(olmo_label[ADAC == 1] %in% c("MOTIVATION", "EXPECTED BEHAVIOR"), na.rm = TRUE)
),
olmo_BI_prop_adac = if_else(
olmo_RK_prop_adac = if_else(
n_tags_adac == 0L,
NA_real_,
mean(olmo_label[ADAC == 1] %in% c("BUG REPRODUCTION",
"INVESTIGATION AND EXPLORATION"), na.rm = TRUE)
mean(olmo_label[ADAC == 1] %in% c("ACTION ON ISSUE",
"ISSUE CONTENT MANAGEMENT"), na.rm = TRUE)
),
n_tags_no_adac = sum(!is.na(olmo_label) & ADAC == 0),
olmo_BE_prop_no_adac = if_else(
olmo_EP_prop_no_adac = if_else(
n_tags_no_adac == 0L,
NA_real_,
mean(olmo_label[ADAC == 0] %in% c("OBSERVED BUG BEHAVIOR", "EXPECTED BEHAVIOR"), na.rm = TRUE)
mean(olmo_label[ADAC == 0] %in% c("OBSERVED BUG BEHAVIOR", "BUG REPRODUCTION", "INVESTIGATION AND EXPLORATION"), na.rm = TRUE)
),
olmo_SOL_prop_no_adac = if_else(
olmo_TSOL_prop_no_adac = if_else(
n_tags_no_adac == 0L,
NA_real_,
mean(olmo_label[ADAC == 0] %in% c("SOLUTION DISCUSSION", "SOLUTION USAGE"), na.rm = TRUE)
mean(olmo_label[ADAC == 0] %in% c("SOLUTION DISCUSSION", "SOLUTION USAGE", "TESTING", "WORKAROUNDS", "TASK PROGRESS"), na.rm = TRUE)
),
olmo_VR_prop_no_adac = if_else(
olmo_DIO_prop_no_adac = if_else(
n_tags_no_adac == 0L,
NA_real_,
mean(olmo_label[ADAC == 0] %in% c("OBSERVED BUG BEHAVIOR", "EXPECTED BEHAVIOR",
"SOLUTION DISCUSSION", "SOLUTION USAGE",
"INVESTIGATION AND EXPLORATION"), na.rm = TRUE)
mean(olmo_label[ADAC == 0] %in% c("MOTIVATION", "EXPECTED BEHAVIOR"), na.rm = TRUE)
),
olmo_BI_prop_no_adac = if_else(
olmo_RK_prop_no_adac = if_else(
n_tags_no_adac == 0L,
NA_real_,
mean(olmo_label[ADAC == 0] %in% c("BUG REPRODUCTION",
"INVESTIGATION AND EXPLORATION"), na.rm = TRUE)
mean(olmo_label[ADAC == 0] %in% c("ACTION ON ISSUE",
"ISSUE CONTENT MANAGEMENT"), na.rm = TRUE)
),
.groups = "drop"
) |>
@ -289,5 +279,15 @@ ggplot(task_level_variables, aes(
)
ggplot(task_level_variables,
aes(
x=as.factor(source),
y=olmo_RK_prop,
fill=as.factor(source)
)) +
geom_boxplot(alpha = 0.7, position = position_dodge(width = 0.9)) +
scale_fill_viridis_d() +
theme_minimal()
# 4. save
write.csv(task_level_variables, "110925_DSL_df_adac.csv", row.names = FALSE)
write.csv(task_level_variables, "111725_DSL_frame.csv", row.names = FALSE)

17
mgaughan-rstudio-server_31035935.out Normal file
View File

@ -0,0 +1,17 @@
1. SSH tunnel from your workstation using the following command:
ssh -N -L 8787:n3439:35765 mjilg@klone.hyak.uw.edu
and point your web browser to http://localhost:8787
2. log in to RStudio Server using the following credentials:
user: mjilg
password: QKOjN5O9o8KE4QlK+t4M
When done using RStudio Server, terminate the job by:
1. Exit the RStudio Session ("power" button in the top right corner of the RStudio window)
2. Issue the following command on the login node:
scancel -f 31035935