nash-0/34-features-load-clean-all.Rmd at main · VandyDataScience/nash-0 · GitHub

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

---
title: "34-feature-engineering"
output: html_notebook
---

This notebook cleans and merges all city dataframes into one comprehensive dataframe.

```{r load}
source(knitr::purl("10-load-data.Rmd"))
fs::file_delete("10-load-data.R")

mdl_list <- list('Austin, TX', 'Nashville', 'San Fran, CA')
dfs <- mdl_list %>%
   map(~read_feather(expand_boxpath(str_c(., '/ready_for_first_model.feather'))))

austin_permits <- as.data.frame(dfs[1])
nash_permits <- as.data.frame(dfs[2])
sanfran_permits <- as.data.frame(dfs[3])

#appends city id to prevent duplicate permit numbers during merge of datasets

set_permit_prefix <- function(df, prefix) {

  df <- df %>%
    mutate(permit_number <- str_c(prefix, '-', permit_number))

  return(df)
}

nash_permits <- set_permit_prefix(nash_permits, "NSH")
austin_permits <- set_permit_prefix(nash_permits, "AUS")
sanfran_permits <- set_permit_prefix(nash_permits, "SAN")
```


```{r merge_df}

#merge nashville and austin

nash_austin_df <- bind_rows(nash_permits, austin_permits)
nash_austin_df

tail(nash_austin_df)

#merge nashville/austin with san fran

all_permits_df <- bind_rows(nash_austin_df, sanfran_permits)
all_permits_df

tail(all_permits_df)
nrow(all_permits_df)

#final dataframe cleaning

all_permits_clean <- all_permits_df %>%
  filter(const_cost > 1) #based on Allie's recommendation to not count projects of 1 or 0 dollars

all_permits_clean
nrow(all_permits_clean)

#filtering out these low cost permits only reduces dataset by 3%


```

```{r write_new_df}

write_feather(all_permits_clean, expand_boxpath("all_permits.feather"))

```