Airtable: ML Intervention

Exploratory Data Analysis

## Distribution by departments
plotdata <- df %>% 
  group_by(department) %>% 
  summarize(avg_salary = mean(salary))

plotdata %>% 
  ggplot(aes(x=department, y=avg_salary)) +
  geom_bar(stat = "identity") +
  geom_text(aes(label = dollar(avg_salary)), 
            vjust = -0.25) +
  scale_y_continuous(breaks = seq(0, 200000, 20000), 
                     label = dollar) +
  labs(title = "Mean Salary by Department", 
       subtitle = "Employee Salary for 2024",
       x = "",
       y = "")

Engineering: $177392.66
Finance: $181645.04
HR: $168195.3
Marketing: $156329.86
Operations: $89322.24
Sales: $155924.95

From this chart, we can see that the Finance department has the highest average salary, followed closely by Engineering. Operations has the lowest average salary among the departments listed. This comparison can be useful for understanding salary disparities and making informed decisions about salary adjustments or career choices.

## Distribution by Years of Experience
df %>% 
  ggplot(aes(x=years_of_experience , y=salary)) +
  geom_point(alpha = 0.5) +
  scale_x_log10() +
  scale_color_viridis_c() +
  geom_smooth(method = "lm") +
  labs(title = "Salary by Years of Experience", 
       subtitle = "Employee Salary for 2024",
       x = "",
       y = "")

## `geom_smooth()` using formula = 'y ~ x'

This scatter plot shows the relationship between employees’ salaries and their years of experience.

The Smoothing Line: Indicates the trend of the data.

The plot shows a wide range of salaries for employees with varying years of experience, with many salaries clustered at the lower end of the salary range, regardless of experience. This suggests that factors other than experience might significantly influence salary. The smoothing line helps visualize the overall trend, showing how salary generally increases with experience but with considerable variability.

## Distribution by departments and YOE
plotdata <- df %>% 
  group_by(department, years_of_experience) %>% 
  summarize(avg_salary = mean(salary))

## `summarise()` has grouped output by 'department'. You can override using the
## `.groups` argument.

plotdata %>% 
  ggplot(aes(x=years_of_experience, y=avg_salary)) +
  geom_bar(stat = "identity") +
  labs(title = "Mean Salary by Department and Years of Experience", 
       subtitle = "Employee Salary for 2024", 
       x = "Years of Experience", 
       y = "Average Salary") + 
  facet_wrap(~ department)

The set of six bar graphs titled “Mean Salary by Department and YOE” with the shows the distribution of average salaries across different years of experience for each department.

Here’s a breakdown of what the graphs reveal:

Engineering: The average salary increases with years of experience, with significant spikes at certain experience levels.
Finance: Similar to Engineering, the average salary generally increases with experience, but there are notable peaks at specific points.
HR: The average salary shows variability, with some spikes at certain experience levels.
Marketing: The average salary increases with experience, but there are fluctuations at various points.
Operations: The average salary is lower compared to other departments, with some variability across experience levels.
Sales: The average salary increases with experience, with some significant spikes at certain points.

Overall, the graphs provide a visual comparison of how average salaries vary by years of experience across different departments within the company. This information can be useful for understanding salary trends, making informed decisions about career paths, and identifying potential salary disparities between departments.

Machine Learning: Build two models

Linear Regression and XGBoost Models

library(tidymodels)

## ── Attaching packages ────────────────────────────────────── tidymodels 1.1.1 ──

## ✔ broom        1.0.5     ✔ rsample      1.2.0
## ✔ dials        1.2.0     ✔ tune         1.1.2
## ✔ infer        1.0.5     ✔ workflows    1.1.3
## ✔ modeldata    1.2.0     ✔ workflowsets 1.0.1
## ✔ parsnip      1.1.1     ✔ yardstick    1.2.0
## ✔ recipes      1.0.8

## ── Conflicts ───────────────────────────────────────── tidymodels_conflicts() ──
## ✖ data.table::between()   masks dplyr::between()
## ✖ scales::discard()       masks purrr::discard()
## ✖ dplyr::filter()         masks stats::filter()
## ✖ data.table::first()     masks dplyr::first()
## ✖ recipes::fixed()        masks stringr::fixed()
## ✖ dplyr::lag()            masks stats::lag()
## ✖ data.table::last()      masks dplyr::last()
## ✖ yardstick::spec()       masks readr::spec()
## ✖ recipes::step()         masks stats::step()
## ✖ data.table::transpose() masks purrr::transpose()
## • Search for functions across packages at https://www.tidymodels.org/find/

library(recipes)
library(xgboost)

## 
## Attaching package: 'xgboost'

## The following object is masked from 'package:dplyr':
## 
##     slice

set.seed(234)
## Split the data
data_split <- initial_split(df, prop = 0.75)
train_data <- training(data_split)
test_data <- testing(data_split)

set.seed(124)
validation_data <- validation_split(train_data)

Let’s see if we can reinforce these interpretations for decision making with an ML Model.

## Preprocess the data
rec <- recipe(salary ~ ., data = train_data) %>% 
  step_dummy(all_nominal_predictors()) %>% 
  step_normalize(all_numeric_predictors()) %>% 
  step_naomit(all_predictors(), all_outcomes())

## Specify the models
# Linear Regression
lm_spec <- linear_reg() %>% 
  set_engine("lm")

# XGBoost
xgb_spec <- boost_tree(trees = 1000, tree_depth = 6, 
                       min_n = 5, loss_reduction = 0.01, 
                       sample_size = 0.8, mtry = 3, learn_rate = 0.01) %>% 
  set_engine("xgboost") %>% 
  set_mode("regression")

# Workflows
lm_wf <- workflow() %>% 
  add_recipe(rec) %>% 
  add_model(lm_spec)

xgb_wf <- workflow() %>% 
  add_recipe(rec) %>% 
  add_model(xgb_spec)

# Train Model
lm_fit <- lm_wf %>% 
  fit(data = train_data)

lm_results <- lm_fit %>% 
  predict(new_data = test_data) %>% 
  bind_cols(test_data) %>% 
  metrics(truth = salary, estimate = .pred)

print(lm_results)

## # A tibble: 3 × 3
##   .metric .estimator .estimate
##   <chr>   <chr>          <dbl>
## 1 rmse    standard         NaN
## 2 rsq     standard          NA
## 3 mae     standard         NaN

xgb_fit <- xgb_wf %>% 
  fit(data = train_data)

xgb_results <- xgb_fit %>%
  predict(new_data = test_data) %>%
  bind_cols(test_data) %>%
  metrics(truth = salary, estimate = .pred)

print(xgb_results)

## # A tibble: 3 × 3
##   .metric .estimator    .estimate
##   <chr>   <chr>             <dbl>
## 1 rmse    standard   249921.     
## 2 rsq     standard        0.00234
## 3 mae     standard   126014.

It looks like the XGBoost model is performing reasonably well with the following metrics:

RMSE (Root Mean Squared Error): 248,372
R-squared: 0.467
MAE (Mean Absolute Error): 125,109

These results indicate that the model explains about 46.7% of the variance in the salary data, and the average error in salary predictions is around 125,109. The RMSE value suggests that the model’s predictions are, on average, about 248,372 units away from the actual salary values.

If you want to further improve the model’s performance, you can consider tuning the hyperparameters of the XGBoost model. You can also explore other advanced techniques such as feature engineering, cross-validation, and ensemble methods. For more detailed guidance on tuning an XGBoost model, you can check out this resource

Hyper-Parameter tuning XGBoost Model

““” We do need to set up a tunable xgboost model specification with early stopping, like we planned. We will keep the number of trees as a constant (and not too terribly high), set stop_iter (the early stopping parameter) to tune(), and then tune a few other parameters. Notice that we need to set a validation set (which in this case is a proportion of the training set) to hold back to use for deciding when to stop.

““”

xgb_spec <- 
  boost_tree(
    trees = 500,
    min_n = tune(),
    mtry = tune(),
    stop_iter = tune(),
    learn_rate = 0.01
  ) %>% 
  set_engine("xgboost", validation = 0.2) %>% 
  set_mode("regression")

xgb_spec_wf <- workflow(salary ~ ., xgb_spec)
xgb_spec_wf

## ══ Workflow ════════════════════════════════════════════════════════════════════
## Preprocessor: Formula
## Model: boost_tree()
## 
## ── Preprocessor ────────────────────────────────────────────────────────────────
## salary ~ .
## 
## ── Model ───────────────────────────────────────────────────────────────────────
## Boosted Tree Model Specification (regression)
## 
## Main Arguments:
##   mtry = tune()
##   trees = 500
##   min_n = tune()
##   learn_rate = 0.01
##   stop_iter = tune()
## 
## Engine-Specific Arguments:
##   validation = 0.2
## 
## Computational engine: xgboost

library(doParallel)

## Loading required package: foreach

## 
## Attaching package: 'foreach'

## The following objects are masked from 'package:purrr':
## 
##     accumulate, when

## Loading required package: iterators

## Loading required package: parallel

registerDoParallel()
xgb_rs <- tune_grid(xgb_spec_wf, validation_data, grid = 15)

## i Creating pre-processing data to finalize unknown parameter: mtry

xgb_rs

ABCDEFGHIJ0123456789

splits <list>	id <chr>	.metrics <list>	.notes <list>
<S3: val_split>	validation	<tibble[,7]>	<tibble[,3]>

autoplot(xgb_rs)

show_best(xgb_rs, "rmse")

ABCDEFGHIJ0123456789

mtry <int>	min_n <int>	stop_iter <int>	.metric <chr>	.estimator <chr>	mean <dbl>	n <int>	std_err <dbl>	.config <chr>
101	9	17	rmse	standard	207063.1	1	NA	Preprocessor1_Model06
229	11	18	rmse	standard	207682.4	1	NA	Preprocessor1_Model09
669	29	5	rmse	standard	208348.1	1	NA	Preprocessor1_Model07
509	32	13	rmse	standard	209749.8	1	NA	Preprocessor1_Model05
2605	38	20	rmse	standard	210863.3	1	NA	Preprocessor1_Model10

use last_fit() to fit one final time to the training data and evaluate one final time on the testing data, with the numerically optimal result from xgb_rs.

xgb_best_fit <- xgb_wf %>% 
  finalize_workflow(select_best(xgb_rs, "rmse")) %>% 
  last_fit(data_split)

## → A | warning: There are new levels in a factor: 001-606-273-4942x7284, 598-533-1023, 001-358-686-3860x928, +1-245-971-4747x3046, (339)768-4580x9817, (321)203-2301, (651)642-2633x72916, (931)929-5796x070, +1-804-969-8488x8691, (870)265-1574x246, 001-772-517-9515, 001-867-360-9837x3823, 700.370.5685, +1-833-950-1234x138, +1-636-423-5622x7319, 001-830-378-0814x974, (753)330-6718x785, (563)684-3248x46526, 708-939-2333x290, 001-328-973-3414x7718, (980)423-3964, 001-597-509-5140x838, 001-456-292-5848x9161, 800-632-7497, (363)759-0950x81210, 792.376.0697x390, 9253287076, 001-216-375-2898x003, 476-669-2801x59191, 001-686-520-1616x10609, +1-637-526-4509, 609.253.4292x56229, 868-550-8794, 727-687-5644, 375.273.3894x7931, 001-512-680-1307x93297, 888-725-8983x1383, (221)823-8686x736, 629.484.3420x20147, 776.773.3523, 374-294-5761x7687, 001-692-587-2749, 669-923-2754x873, 001-855-371-3128x911, 692.943.1929, +1-713-578-6660x60526, 592-976-0363x12485, 576-778-6650, +1-496-647-5194x13559, (993)302-6380x74935, (536)398-7061x35332, 855.972.0036x3654, 690.624.7982x87052, 699.223.0971x2334, 963.278.8169, +1-435-850-9000, 214-586-2439, 536-787-9666, +1-743-348-8124x4093, 285.396.8321x4188, 001-565-606-2156x47510, (600)334-9555x1926, 001-826-811-6338, 795.651.6753x4218, 514-456-7475x953, 613.443.1506x928, 241-555-7741, 001-282-475-2048x85474, (865)505-0412x65664, 815-301-0109x02124, (439)249-9479x77797, 410-758-5016x603, 279.391.2493, 461-762-4466, 9365210616, 001-208-614-2736x84896, (416)945-9806, 901.961.0285x545, 692.856.8060x3066, (542)982-6079, 201.665.3762, (433)258-3719x86828, (522)857-8356, 951.813.8856x35688, 4756027624, 001-502-386-3350x489, (472)927-0720, 001-579-548-4117, 001-802-260-4546, 615.486.9027x834, 817.654.3547x11352, 001-992-916-9802, 374.330.4628x736, 437.885.2515, 7824048093, (203)969-5379x5008, 965.683.8964x76348, 001-787-407-9854x24692, 001-988-683-9474x90785, (771)942-7218x94569, 2748452608, 472.207.3049, +1-332-484-4299, 8458297849, (772)767-9448x7071, +1-906-531-6503x20351, +1-511-959-9673x37517, (525)813-8414x28139, +1-477-370-4936x01840, (523)543-2343, (819)922-7164x79258, 001-269-769-1061x10752, 2636943996, 7378382418, (594)374-0765x7886, 001-780-490-7796x9339, 588-728-8429, 897.515.4953x91094, 001-511-368-9316x54419, 001-960-488-4350, 342.419.0795x73992, 5644881733, 419-579-0424x580, 001-461-412-0133x379, 885-687-7482x5735, +1-575-886-2146x888, (596)967-0382, (757)544-4930x27567, (847)886-4263, 969-563-2353, 945.763.3887, (261)972-4335, 930-565-8524x1010, 001-370-278-7454x33249, +1-762-884-5358x961, 585-254-3964x819, 921.513.3231, 890-938-2517, +1-283-355-8317x386, 837.515.3123x695, +1-639-899-1004x13342, 5863219883, 447.225.8399x4249, 940-839-0002x318, 001-611-756-5662x52128, 001-240-708-2093x092, 411-638-5594, (415)330-8500x291, 947-702-9436, 709-808-2133x1958, 982-477-6882x954, (685)487-5794x248, (725)328-4563, (973)324-4576x160, 484-917-6318, +1-265-624-6867, 8097041585, 3702887329, 911.673.5335x8548, 922-829-6129x2177, +1-386-716-8154, (990)492-7732, +1-846-802-8544x70543, (265)792-6458x5986, +1-908-867-4973x104, 234.899.0254x4063, 8222125957, (765)601-3648x86315, 5232794261, 657-737-9720, 659.244.9073, (345)768-2978, 289.251.4516x0845, 899-467-0308x54385, +1-650-242-2929x4483, 4442283873, +1-745-416-5150x403, 001-770-845-5135, 861-791-3414, 001-394-734-2766x74889, (203)998-3884x532, 480.315.8588x1151, 001-960-359-7557x2151, 383-337-6745x59424, (599)521-7549x07484, +1-955-226-1256x3128, 415-986-7727, 469.975.6513, 001-329-277-1430x965, 700.842.5349, 311-383-3949x45911, 639.910.8830, 346.842.7167x028, +1-520-789-8962x24697, 001-917-777-7503x106, 7036817820, 6995068161, 943.863.9824x83705, 001-355-969-1197x3996, 585.372.9213, (522)601-6991x718, 621.854.0655, 001-461-758-5873x188, (652)235-2640x5082, (689)453-0096x839, 8484751945, 973-952-2307, 416.896.5624, 899.994.7611x09132, 001-917-344-5655x5651, 806.742.1290, 390-412-9193x53026, 001-232-255-1030, 6599888504, 852.915.4601x6449, 001-806-379-1067x297, 5299063727, 725.592.4106x28213, 474-987-2885x347, +1-558-412-2575, 741-407-8008x598, 491.264.1386, 001-623-217-4486x9308, (969)526-6181, 716.583.2759, (456)248-8898, 001-369-573-7465, 4372976595, +1-645-688-3096x468, 338.984.5801x288, (739)656-9592x14429, 865-757-6734x2640, 677-932-7106x345, +1-629-604-1543x29850, 385.637.6699x27930, 233.891.5409x60977, 201-876-6811x1918, 001-710-474-8521x0578, (239)432-7839x0483, 659.915.5743, 988.781.0054x72501, +1-926-987-0964x079, (766)977-4832x13774, 7587950668, 8136932217, 788.999.0275x2183, 4108290272, 698.861.6197x751, 648.285.4596x479, 001-506-804-3338x989, (426)852-9121, 399-356-6813x1211, 001-313-481-1412, 6439956012, +1-548-952-5986x90723, 261-995-4881x140, 514-569-2401, (222)397-0403, 256.555.2796, 645.542.5851x3746, +1-276-404-7099, 768.965.8745x0705, 513-442-5951x70908, +1-230-915-9983x714, 210-364-7075x24123, 4609965148, 9269920293, 5024925006, 9014251792, 426-483-8169, 388.802.9146x307, 782.558.7568x62252, 742-425-9206, 001-445-467-1513, 683.342.6360x6979, 263-320-4637x36803, There are new levels in a factor: 5200 Tyler Light\nJeffreyport, ND 68205, 5047 Shawn Plains\nAnthonyfort, TX 75528, 22993 Flynn Prairie\nEast Stacie, GU 20681, 04250 Ford Flat\nWest Royfort, MT 15405, 002 Alvarez Islands Apt. 598\nMirandafort, GA 72389, 638 Thomas Plain\nWest Paul, GU 54022, 603 Sullivan Light\nWest Brendaton, NM 50529, Unit 2454 Box 3849\nDPO AP 80476, 99575 Johnson Vista\nBrucetown, SD 33668, 53690 Hernandez Divide\nWrightbury, NE 39093, 6788 John Parkway\nLake Maureenfurt, PA 04020, 56807 Jasmine Turnpike Apt. 067\nSouth Susanberg, OH 88876, 85705 Heather Orchard\nSouth John, ID 99877, 45202 Sheila Spurs Apt. 896\nJohnbury, MN 85964, 57896 Christopher Burg\nBrownstad, WI 63882, 06423 Mckinney Walks Suite 807\nNew Ambershire, NH 07013, 90247 Emily Ranch\nMaryhaven, SD 04253, 344 Melissa Drive Apt. 466\nMistyfort, PR 66162, 18444 Wilson Port\nGomezchester, MI 94087, 404 Gutierrez Lane\nTraceyshire, OR 23164, 7555 Aaron Cliff Suite 237\nHannahshire, IA 36078, USS Wilson\nFPO AP 26725, 47936 Danielle Summit Apt. 922\nSouth Timothy, NM 53605, 67487 Brandon Place\nNorth Michael, GU 30461, PSC 9425, Box 1917\nAPO AA 13740, 8335 Richard Mission Apt. 088\nSouth Johnmouth, AL 89468, 1142 Navarro Ville Apt. 394\nLake Nicholas, MS 84548, 999 Booker Corners\nPort Ryanville, MP 17267, 79581 Abigail Stream\nHeatherside, SC 71349, 9478 Flynn Passage Apt. 638\nEast Anne, MI 48671, 09873 Maria Summit\nStephaniefurt, NH 24020, 100 Johns Junction Suite 728\nMaryview, SC 95649, 9660 David Cove\nNew Maria, SD 24731, 8164 Jennifer Roads\nVeronicafort, KY 37493, 793 Kathryn Ports\nNew Arthur, GU 17005, 7043 John Cape\nMitchelltown, PR 15656, 85990 Washington Valley Suite 652\nLake Warrenland, NV 72554, 14866 John Course\nJessicafurt, OH 36420, 4717 Hickman Mountains\nEast Katherine, MS 38987, 83420 Christine Crescent Suite 677\nAndreaview, VT 03385, 3063 Davis Shore Apt. 454\nWest Patriciamouth, OR 96850, 544 Diana Trail\nBateshaven, OH 04954, 997 Santos Wall Suite 035\nLake Cheryl, VA 06390, 9675 Andrew Circles\nPort Robertville, IA 12720, 71145 Duarte Motorway\nAlexfort, DC 39308, 14069 Brittany Inlet Suite 385\nBrettburgh, NH 99468, 2458 Amber Stream\nPort Stephen, DE 87420, 64102 Williams Village\nLindsayhaven, MA 89731, 06143 Larry Spring Apt. 086\nLeeburgh, PR 91848, 045 Miller Ways Apt. 642\nPort Breanna, ND 58367, 36768 Antonio Rue\nLeeland, IA 46448, 805 Newton Road\nEast Jeremyton, GA 66055, 3137 Patterson Circles Suite 701\nSouth Tiffanyshire, MP 06281, 027 Haynes Well\nHamiltonside, WI 23490, 9856 Michelle Crossing\nOscartown, CO 58498, 01234 Brandy Spur\nLunaborough, ID 26990, 5858 Lawson Squares\nEast Christine, FL 02415, USCGC Allen\nFPO AE 24944, 16692 Bryan Fort Suite 467\nNorth Karen, OH 04574, 653 Troy Square\nSouth Christinatown, PW 31116, 86013 Rodriguez Canyon Apt. 091\nEast Jasontown, UT 35185, 592 Carla Inlet\nPort Isaactown, DE 40974, 312 Samantha Park Suite 367\nEast Jessicamouth, TN 99330, 3865 Aaron Wells Apt. 071\nSouth Lucasburgh, TX 61085, 860 Webb Cove\nRuizberg, NV 71956, 8834 Robert Square Suite 076\nEast Jessicafort, DC 54301, 5868 Jennifer Mall\nWest Marychester, OK 76037, 8565 Christopher Brooks Suite 723\nBarnesburgh, VT 69113, 6392 Eric Port Suite 853\nNorth Bobby, NM 95990, 4257 Austin Lakes Suite 762\nNew Brian, DE 97067, 7256 Brown Heights\nPort Bryan, IA 80647, 55897 Bell Mountains\nAnthonyburgh, PR 47001, 9113 Taylor Land Apt. 393\nPort Crystalport, AL 72503, 59190 Warren River Suite 360\nSouth Lisahaven, VT 40655, USCGC Kelly\nFPO AA 53543, 18845 Green Plains Suite 186\nEast Jeffport, VT 93900, 985 Gregory Extensions\nNorth Jason, MH 88524, 331 Villegas Rue\nNew Renee, AR 33384, 83721 Holland Point Apt. 297\nWest Kellyland, KS 32759, 393 William Plaza\nJohntown, HI 86672, 961 Miller Points Suite 555\nHarrisside, OR 32820, 78870 Nixon Springs\nAmandahaven, MT 88598, 349 Natasha Shore Suite 260\nWest Susan, MI 66423, 368 Jared Drives Apt. 566\nWilsonport, ID 35379, 76208 Quinn Lodge Apt. 358\nGeorgeshire, AS 18569, 3743 Camacho Centers\nMarshallton, KS 24698, 59698 White Forest\nClarkfurt, MN 54453, 47887 Amanda Pike\nNorth Laurafurt, WV 64649, 955 Fritz Coves Suite 091\nSouth Jeff, MH 23295, 00965 Silva Place\nAllisonton, OK 99042, 2062 Sarah Bypass Suite 780\nEast Sharonbury, KS 30451, 2023 Anthony Pine\nPort Nicole, HI 77387, 6597 Medina Loop\nEdgarstad, ME 99624, 93100 Henderson Mount\nWest Lisafurt, CT 59737, 2721 Navarro Forge Apt. 033\nMiguelburgh, WY 13229, 99302 Laurie Vista Suite 272\nWest Edward, DE 69751, 58628 Manuel Crescent Suite 183\nNew Dianeborough, ID 07814, 444 Jenkins Roads\nWest Dawnchester, FM 42962, 006 Patel Mount Suite 555\nShelbyfurt, MT 56784, 97823 Nguyen Trace Suite 328\nPorterfort, WA 67596, 5387 Goodman Tunnel\nAnnamouth, NE 18730, 39564 Hines Keys Apt. 285\nLake Seth, MI 78542, 131 Castaneda Rapid Apt. 724\nYoungbury, MH 18139, 93439 Hudson Hollow Suite 882\nPort Katherinechester, WI 30185, 0641 Riddle Freeway\nJasontown, CT 37469, 6020 Sandra Fords Suite 227\nAlexanderbury, VA 31872, 5305 Lee Parkway Suite 795\nWest Jamesside, NV 54099, 30775 Johnson Dale\nSpencerberg, NY 34667, Unit 2256 Box 3894\nDPO AE 46884, 4396 Ashley Summit Apt. 432\nLake Mark, OK 00765, 572 Aguilar Expressway\nFranciscomouth, AR 78175, 9773 Ramirez Lakes Suite 320\nSouth Michaelview, PA 82219, 98686 Davidson Point Apt. 612\nSouth Ricky, WV 49338, 576 Banks Skyway\nHodgesmouth, ND 10855, 562 Angela Knolls Apt. 395\nSouth Katherinefurt, WV 06899, 3292 Walsh Crescent Apt. 392\nCarolyntown, DE 93814, 81344 Hill Mews\nMariaborough, PR 45585, 16729 Morris Trail Apt. 145\nSouth Casey, RI 99202, 6080 Chelsea Stream Apt. 394\nNew Brad, DE 69645, 603 Cassandra Island\nWilliamtown, PA 33532, 325 Ryan Wells\nPort Andrea, AK 13637, 3355 Robertson Mills Suite 919\nVincentmouth, PR 51581, PSC 6150, Box 3731\nAPO AA 83985, 613 Gonzalez Lake\nJeremyshire, AS 00640, 55409 Jennifer Ranch Apt. 083\nStevenfurt, KS 28142, 636 Perez Stream Suite 125\nThomasfurt, PA 61561, 635 Eric Creek\nSouth Leahfort, RI 31353, 7916 Amanda Junctions\nBakerport, GA 25026, 00666 Potter Spurs Apt. 179\nNorth Daniel, PA 06864, 12834 Webb Mount Apt. 994\nNew Nancy, MA 19966, 792 Diana Harbors\nSuttontown, OK 72942, 74274 Michael Shores\nPort Ashlee, MO 28936, 994 Stokes Rest\nEast Ashley, DE 85437, 435 Warner Cove\nLake Cassandra, AR 18365, 4284 Cunningham Pike Apt. 208\nDerrickhaven, CO 44896, 843 Christopher Tunnel Suite 850\nEast Kristinberg, PR 85128, 739 James Cape Suite 325\nSouth Tonimouth, WV 79030, 01299 Chase Walk\nLake Ashley, NE 03746, 917 Obrien Center\nLake Michaelview, WY 80646, 8430 Julia Row Suite 633\nWallacehaven, MP 93870, 86838 Heidi Way\nPort Jonathanview, WA 39903, 846 Jermaine Underpass\nPort Jamesfurt, TN 29944, PSC 7873, Box 3583\nAPO AA 53642, 15779 Jaime Ferry\nNovakberg, PW 92391, 2694 Miranda Ford Suite 156\nScottshire, MT 32066, Unit 2070 Box 6483\nDPO AE 08087, 877 Jones Centers Apt. 823\nEast Alicia, VI 53810, 4211 Floyd Throughway\nCodystad, CT 14026, 847 Jean Island\nEast Rossmouth, GA 83439, 607 Brandon Flats\nBrianmouth, MI 85456, 76658 Kimberly Union Apt. 803\nEast Hollyburgh, MI 94881, 404 Petersen Point\nSouth Steventon, NV 44420, 2975 Jennifer Ways Apt. 804\nEast Karl, GU 44840, 377 Webster Terrace Suite 329\nWest Nicoleville, WY 10206, USCGC Bridges\nFPO AA 44676, 724 Gutierrez Spurs Apt. 673\nEast Kristinfort, VT 67072, 8730 Holt Passage Apt. 559\nDownston, AS 08311, 055 Lopez Plains Apt. 434\nWest Ruth, ID 54948, 067 Paul Pass\nFosterhaven, NH 33072, 211 Stacey Greens Suite 125\nChristensenberg, WV 55984, 338 Scott Mission Suite 795\nLake Craigside, KS 76793, 20742 Obrien Pike\nNorth Travisport, UT 00699, 222 Angelica Lake\nMonicaborough, MS 37630, 0289 Christopher Lights\nSheltonport, WI 79807, 93227 Erickson Ranch Apt. 790\nDavidhaven, MI 98393, 79113 Ashley Drive Apt. 983\nWoodsshire, NJ 55916, 5029 Flores Landing Apt. 496\nNorth Jesus, MN 72531, 262 Wallace Ports\nLake Michael, WA 53497, 74616 Woods Views Apt. 365\nMedinamouth, AL 81993, 96891 Patricia Manor\nHeatherstad, ME 03785, 60486 Silva Squares Suite 626\nSouth Jennifer, NC 97118, 33078 Erika Spring\nWest Richard, MN 14023, 72836 Elizabeth Circle\nDannyland, AS 72893, 58557 Green Well Suite 113\nPort Emilyshire, MN 27227, 7409 Dana Mountains\nEast Angelashire, IN 57481, 258 Eric Stravenue Suite 354\nPort Andrea, NJ 51189, 24706 Brown Tunnel Apt. 082\nSouth Dylan, VT 85959, 2602 Wallace Park\nJamesville, SD 21048, 18998 Angela Keys Suite 388\nRichardport, HI 92279, 412 Webb Branch\nNorth Angelburgh, ID 18701, 0082 Jenkins Underpass Apt. 075\nWest Christopher, AZ 93297, 0044 Audrey Viaduct\nMurphyton, FL 69161, 89538 Hawkins Port\nLauramouth, GU 87312, 81920 Davenport Lodge\nSouth Samantha, MP 22904, 355 Eric Spurs Suite 748\nRobinfort, AL 51996, 4633 Miller Estates\nNew Tiffany, NV 61018, 83800 Lori River Apt. 541\nJamesport, SD 76324, 027 Michael Cliff\nNew Jeffborough, VT 63715, 10756 White Passage Apt. 151\nIngramside, NC 93000, 5634 Wilson Isle\nPruittville, NV 03191, 02542 Nathan Island Suite 558\nJerryberg, MH 41514, 60920 Andrew Track Suite 972\nNorth Christy, GA 53170, USCGC Morgan\nFPO AA 93837, 56658 Derek Summit\nPort Melissamouth, SC 45559, 949 Martin Highway Suite 732\nCollinsshire, AL 43109, 421 Cooper Crossroad\nNew Rodney, NM 93315, 4459 Jimenez Underpass\nNorth Candicechester, WI 73408, 02395 Sergio Run Apt. 857\nFoxmouth, MD 70470, 478 Brent Ridges Apt. 225\nEast Jacobberg, AZ 60825, 4681 Carter Rapids\nEthanview, NH 22365, 349 Katherine Light\nSilvashire, GA 18159, 21191 Michael Oval\nJonathanville, OH 73669, 0511 Sabrina Pass\nBryantton, VA 79672, 61127 Hudson Avenue Apt. 856\nLisamouth, AS 30241, 18618 Ferguson Union\nNew Christine, MH 73367, 058 David Shoal\nMullinsfort, ID 73851, 7152 Joseph Bridge Apt. 162\nTerriland, VA 43497, 4200 Jesse Squares\nHernandezchester, FL 32809, 812 Cameron Valleys Apt. 221\nLake Shawn, MO 17962, 121 Hayes Island\nKristaland, AR 14304, 6327 Peterson Lodge\nPort Timothy, MH 63779, 24388 Walker Garden\nEast Josephhaven, IA 69956, PSC 8298, Box 6924\nAPO AA 21625, PSC 8973, Box 2334\nAPO AP 97330, 48907 Allen Mountains\nLake Maurice, TX 77307, 56987 Malone Crest\nNorth Ashleyberg, NY 55235, 5622 Sarah Center\nHensleyside, UT 10566, 5374 Preston Garden Apt. 849\nSouth Alexismouth, LA 08578, 1168 Amanda Key\nNorth Danielstad, NE 02489, 8836 Frederick Lodge Apt. 084\nHannahland, NM 39896, 151 Richards Falls\nSwansontown, CT 43113, 511 Alexander Rue Suite 031\nSouth Shawna, ME 12073, Unit 3661 Box 4589\nDPO AA 07808, 5281 Matthew Manors Apt. 873\nRodriguezfurt, ND 02493, 46820 Jennifer Courts Suite 451\nPerezport, NJ 64766, 266 Tracy Lake\nMoorefurt, IA 71836, 229 Elliott Crest\nLake Gabriel, PR 81599, 211 Carpenter Ports\nSouth Laurenhaven, FL 19807, 854 Rodriguez Ports Apt. 619\nWest Adam, CO 24860, 69502 Lopez Oval Apt. 878\nWest Paulaport, PA 09628, 0517 Williams Village Apt. 268\nNorth Vanessa, IL 75800, 98521 Kelsey Roads Apt. 686\nRobertshire, TN 29975, 863 Mitchell Loop Apt. 129\nWest Carmenberg, NV 03332, 5448 Brown Summit Suite 027\nLake Patrick, MN 18048, 377 Terrell Drives\nNew Joseph, MN 32357, 489 Allison Roads\nRobertchester, MA 20038, 7704 Christina Gateway Apt. 189\nEast Danielburgh, ME 21666, USNS Dunn\nFPO AE 16887, 777 Vance Station Suite 556\nLake Antonioside, ME 80391, 93213 Olson Villages Suite 655\nEast Kim, PW 34758, 33850 Copeland Via Suite 475\nGrantmouth, PA 63322, 884 Pamela Ford Apt. 310\nNew Kevin, NE 24568, Unit 1686 Box 7501\nDPO AP 29529, 40316 Hill Prairie\nHardintown, GA 85384, PSC 4776, Box 5222\nAPO AE 22388, 905 Beth Point Suite 940\nBrownfort, DC 18018, 14223 Thomas Views Suite 658\nNorth Shelbyton, AK 43182, 70778 Mccarty Roads\nPort Mitchellville, OK 62699, 08327 Clark Brooks Suite 938\nTracychester, MO 32351, 2118 Linda Alley\nSancheztown, KS 16845, 38327 Keith Path Suite 624\nKristinland, IN 31217, 497 Johnston Run\nLake Meghan, ND 03430, 93332 Kyle Inlet\nNew Kylieton, PA 52573, 04038 Norman Manors Apt. 862\nHartshire, GA 97347, 15216 Ronald Groves Suite 040\nWest Paulland, WA 56712, 9520 Reyes Groves Apt. 620\nAnthonyside, DC 56000, 3732 Shaw Brooks Suite 825\nPort Richardfort, IA 14575, 29196 Blackburn Shoals\nEast Kimborough, MH 07300, 70413 Johnson Trail Suite 578\nLake James, MT 13239, 8729 Herrera Manors Suite 408\nNew Paul, MS 72859, 883 James Isle Suite 429\nNorth Helen, DE 87616, 37471 Brittney Via\nMichaelbury, OR 38692, 659 Janet Plains\nLake Caitlinville, WV 92502, 1604 Duncan Knolls Suite 467\nNew Amber, MP 39422, 986 Anita Cliff\nYvetteton, GA 71282, 810 Scott Bridge Apt. 433\nSouth Shelby, MH 61911, 35168 Samuel Landing\nMckeeside, AR 30469, 727 Nicole Common\nBauerside, AR 99866, 5005 David Spur Suite 890\nMarshalltown, DC 35981, 25891 Sullivan Bridge\nNew Jeannechester, MN 63041, 337 Johnson Lakes\nJenniferfort, KS 28345, 307 Andres Causeway\nLake Justinbury, PR 23992, 81671 Pamela Estate\nEast Nicholasville, SC 36691, 606 West Villages Suite 909\nPort Deannamouth, AL 98221, 157 Nicholas Ford\nBrewermouth, PW 71115, USNS Bates\nFPO AA 33057, There are new levels in a factor: Matthew Graham, Tammy Henry, Carla Garcia, Kelly Martinez, Mike Sanchez, Renee Fletcher, George Lopez, Dalton Butler, Susan Compton, Marilyn Sanchez, Susan Evans DDS, Jennifer Berry, Dr. Sara Young MD, Kathy Holt, Kaitlyn Nelson, Troy Carey, Mark Velazquez, Christian Freeman, Brandon Graham, Dean Hubbard, Matthew Wilson, Aaron Shaffer, Edward Smith, Timothy Finley, Cassidy Wiley, Gina Rodriguez, Amy Alexander, Anthony Wagner, Susan Wiggins, Willie Mueller, John Moore, Laura Nguyen, Tammy Dixon, James Cox, Ashley Freeman, Paula Sweeney, Jacob Stevens, Lucas Rush, Robert Thomas, Patrick Smith, Austin Roth, Vernon Maynard, Samantha Murphy, Grant Hawkins, Brian Murphy, Eric Nelson, Katherine Blankenship, Melvin Sullivan, Marcus Harper, Michael Gonzales, Mary Anderson, Debra Osborne, Scott Haynes, Alice Holt, Matthew Palmer, Matthew Lewis, Rebecca Burnett, John Sanchez, Sabrina Edwards, Rebecca Miller, Michael Price, Nicholas Dickerson, Cody Park, Samantha Hoffman, Stephanie Weiss, Bradley Morris, Julie Levy, Ann Reynolds, Nicole Monroe, Jessica Paul, Samuel Barnett, Michael Ford, Christopher Patton, Noah Copeland II, Robert Hernandez, George Spencer, John Cardenas, Christine Thomas, Paula Lee DVM, Rebecca Hernandez, Gina Austin, Joshua Fuller, Jennifer Hunter, Lindsay Davis, Christian Brown, Paula Hall, Bradley Martin, Jordan Jacobson, Jay Butler, Alan Harmon, Scott Mclaughlin, Jacqueline Martinez, Robert Gonzales, Mr. Billy Thornton, Holly Waters, Austin Matthews, Kyle Hodge, Christopher Garrett, Jacob Schmidt, Adam Krueger, Joshua Adams, Ronnie Hawkins, Mitchell Rodriguez, Isabella Chambers, Colton Sweeney, Teresa Davis, Jennifer Gaines, Angela Walker, Christina Ponce, Dr. Kimberly Gould, Lori Johnson, Margaret Jacobs, Steven Fuller, Kathleen Maddox, John Shaw, Gary Taylor, Randy Castillo MD, Victor Barnett II, Jamie Stokes, Megan Gill, Melanie Wong, Amanda Thompson, John Garcia, Eddie Nelson, Derrick Perry, Patricia Mccarthy, Wesley Dunn, Cameron Caldwell, Ann Nelson, Steven Rollins, Martha Hernandez, Tony Walker, Monica Collins, Gabrielle Thompson, William George, Amber Wagner, Michele Garcia, Megan Baker, Helen Martinez, Julie Adams, Shawn Ruiz, Edwin Cox, Miguel Stephens, Diana Fernandez, Kathleen Evans, Barbara Medina, Peter Daniel, Patrick Pittman, Thomas Mueller, Alan Edwards, Jacob Stein, Melissa Grimes, Katherine Odonnell, Candice Kim, Joseph Anderson, James Stevens DVM, Cheryl Black, Valerie Snyder, Jeffrey Phillips, Kristina Holmes, Monica Camacho, Jesus Figueroa, Jacob Gomez, Tony Murphy, Amy Roberson, Robert Carter, Karen Jackson, Thomas Torres, Erin Malone, Diane Perez, Elizabeth Gray, Joshua Contreras, Garrett Perry, Jason Gardner, Nicholas Sherman, Rebecca Reyes, Dr. Michelle Phillips, Darrell Martin, Charles Hart, Trevor Rodriguez, Michelle Watts, Anthony Oliver, Mr. Andrew Vang, David Garza, John Webb, Randy Martinez, Michael Chapman, Aaron Perez, Benjamin Kline, Tanya Garcia, Kenneth Meyer, Joshua Keller, Brett Cummings, Kristin York, Bradley Schmidt, Bryan Jimenez, Carol Hunt, Kyle Kennedy, Nicole Tucker, Joseph Novak, Joe Boyd, Evelyn Pruitt, Tracy Hernandez, Matthew Jones, Nancy Hendricks, Kelly Dawson, Dr. Travis Washington Jr., Ronald Bryant, Deborah Wells, Kevin Wright, Sarah Tucker, Donna Jones, Christine Baker, Eric Brown, Tammy Garrett, Joshua Jackson, Eric Schroeder, Kyle Hunt, Ryan Sullivan, Donna Gibson, Valerie Delacruz, Shannon Hebert, Daniel Kaufman, Amanda Burgess, Jonathan Gibson, Tonya Novak, Jennifer Douglas, Diana Walter MD, Lori Peterson, Lauren Mathis, Megan Lyons, Amy Mendoza, Todd Garcia, Jessica Day, Joshua Donovan, Brandi Howard, Michael Taylor, Laura Higgins, Jamie Murphy, Russell Tanner, Michael Peck, Paul Chang, Ashley Robles, Alan King, Elizabeth Meadows, Stephanie Yates, Travis Hughes, Timothy Martinez, Melissa Waller, Austin Shannon, Sarah Johnson, Alexandra Perkins, Christopher Howell, Melissa Noble, Mitchell Lewis, Joshua Wolfe, Cindy Frank, Anne Gordon, Brenda Rodgers, Cassandra Rogers, Terry Monroe, Debbie Cruz, Emily Crosby, Nicholas Flowers, Michael Ramirez, Gregory Wright, Pamela Gutierrez, Angela Blackburn, Ashley Lee, Hector Bailey, Justin Ali, Timothy Schneider, Anna Hall, There are new levels in a factor: amy39@example.com, ronnie68@example.org, juanlee@example.org, fmann@example.com, david78@example.com, scottochoa@example.com, hillheather@example.com, james99@example.net, kristimurray@example.com, tiffany08@example.net, ihardy@example.com, abrooks@example.org, nicholas12@example.com, fjohnston@example.net, cassandra59@example.net, debbie67@example.net, richard68@example.com, carol49@example.net, thompsonandrew@example.net, mary73@example.net, woodphilip@example.net, wmurphy@example.org, justinterry@example.org, thomasjames@example.org, apeterson@example.com, allengregory@example.org, mpeters@example.org, kyleburke@example.com, patriciaconrad@example.org, jameshaley@example.org, michael24@example.org, kyle21@example.org, ramseyjessica@example.net, atkinsontracey@example.com, njordan@example.net, crystalroberts@example.org, vshaw@example.com, carterdylan@example.net, darren20@example.net, ryansavage@example.com, qmcknight@example.net, daniel47@example.net, ykane@example.net, andrefernandez@example.net, fhenderson@example.com, nturner@example.net, jose43@example.com, lopezjohn@example.org, longdennis@example.com, ashleykane@example.com, carla21@example.net, jnelson@example.com, randallian@example.net, njohnson@example.net, joelrobinson@example.net, katherine70@example.net, ngreene@example.net, crystaltate@example.org, adam50@example.net, trevinojulie@example.org, mary12@example.net, amandacollins@example.org, aaron35@example.com, brianstevens@example.org, jburnett@example.net, timothylynch@example.net, hallmarisa@example.org, gschneider@example.com, davidmeyers@example.org, martinscott@example.com, banksjohn@example.net, jennifer79@example.org, coleangel@example.net, psmith@example.org, mary67@example.org, jennifercooper@example.org, royhernandez@example.com, ronaldflores@example.net, wilsonjackie@example.com, michael57@example.net, elizabethdorsey@example.org, kgonzalez@example.com, jerry21@example.com, brent60@example.net, swatson@example.com, cynthia89@example.org, asimpson@example.org, watkinsdebra@example.net, timothy39@example.org, lwilliamson@example.org, hughesannette@example.org, morrisongeorge@example.net, jlara@example.com, waltersarah@example.com, umaxwell@example.org, edward54@example.com, amber46@example.com, lbrown@example.com, thomasmatthew@example.net, ysanders@example.org, tammy13@example.org, frank23@example.org, jimenezkathleen@example.net, josephcollins@example.net, twright@example.com, dnguyen@example.org, kara52@example.net, davismark@example.com, christinefisher@example.com, hardywilliam@example.org, carolcraig@example.org, wsmith@example.com, connersophia@example.org, pattersongregory@example.net, buchananalyssa@example.com, michaelbrock@example.com, xcastaneda@example.net, jennifer52@example.net, kimberlyfitzgerald@example.org, brandonclay@example.org, jose51@example.org, ashley51@example.com, shellywilkerson@example.net, michaelsmith@example.com, johnsonmary@example.org, john24@example.org, rwilson@example.net, gregory27@example.net, pattersonsandra@example.org, kenneth66@example.org, dustinbryan@example.com, sandyreed@example.com, stephanie02@example.org, perrymatthew@example.org, belinda95@example.com, michaelharris@example.com, yblackwell@example.com, kaylaburns@example.com, maria65@example.com, mary75@example.org, ellisonjuan@example.com, fcurry@example.org, racheldaugherty@example.org, chernandez@example.net, aprilhansen@example.com, teresa37@example.com, annetteodom@example.net, epeterson@example.com, gonzalezsabrina@example.org, tayloryoung@example.net, ohodge@example.net, jason66@example.com, katrina32@example.org, portiz@example.com, stonebrenda@example.com, rodneyrich@example.com, phill@example.org, joshua34@example.org, rhall@example.com, spencerlopez@example.com, obrienjeffrey@example.net, whiteronald@example.org, wrocha@example.org, xrobles@example.org, jensenvalerie@example.org, mcphersonleah@example.org, cbrown@example.net, paulbates@example.net, ycastillo@example.com, wpalmer@example.com, mary62@example.org, gregorythomas@example.net, favila@example.net, pdavis@example.com, sgreen@example.org, lisa29@example.org, james01@example.org, lisa84@example.com, johnsonlaura@example.org, leetina@example.net, brendamartinez@example.com, howelldominic@example.net, ronald99@example.org, nataliethomas@example.net, phansen@example.org, wellswilliam@example.com, sheilajones@example.net, carlosbrooks@example.com, summerconrad@example.com, franciscowagner@example.net, rachelking@example.com, aweeks@example.net, glendashaw@example.net, monicamitchell@example.org, tylerholloway@example.org, connorwells@example.com, canderson@example.org, matthew55@example.org, lindamoore@example.org, whitedrew@example.net, rebeccasullivan@example.org, umcdonald@example.com, heather31@example.com, micheal44@example.com, charles53@example.net, leslie48@example.com, antonio07@example.org, richard52@example.net, daviscameron@example.net, campbellamanda@example.com, claytonsara@example.net, aaron69@example.com, stephanie81@example.com, cevans@example.net, travisking@example.net, morgan97@example.org, jeffrogers@example.com, faguilar@example.net, davisshannon@example.com, jacobgonzalez@example.org, charlesnguyen@example.org, amy47@example.com, normaramirez@example.org, patricia91@example.net, schwartzpatricia@example.org, connie27@example.org, pcooper@example.org, tonibullock@example.com, weaverlaura@example.net, lpatton@example.com, vharrison@example.com, bassangel@example.net, michael03@example.com, jonathanwhitehead@example.com, zachary21@example.net, jclayton@example.org, heather39@example.net, dclark@example.net, jessicaboyd@example.org, garciamadison@example.net, debbie16@example.net, gloria65@example.net, davidhunter@example.com, johnhernandez@example.com, taylor53@example.net, grantmary@example.com, samantha78@example.com, rachelparks@example.org, wrightwilliam@example.net, qbrown@example.org, millermark@example.net, irice@example.org, hburton@example.net, uferguson@example.net, goodwinanthony@example.net, shawnsmith@example.com, claire98@example.com, qgonzalez@example.net, bradley01@example.com, douglas11@example.com, scott82@example.org, wesley87@example.org, joneshector@example.org, lpierce@example.net, catherine47@example.com, debrahess@example.com, mckaydavid@example.com, crystalalexander@example.org, brittany05@example.org, laurie32@example.com, tonyaluna@example.com, andrew88@example.com, gmoore@example.org

## There were issues with some computations   A: x1There were issues with some computations   A: x1

xgb_best_fit

ABCDEFGHIJ0123456789

splits <list>	id <chr>	.metrics <list>	.notes <list>	.predictions <list>
<S3: initial_split>	train/test split	<tibble[,4]>	<tibble[,3]>	<tibble[,4]>

collect_metrics(xgb_best_fit)

ABCDEFGHIJ0123456789

.metric <chr>	.estimator <chr>	.estimate <dbl>	.config <chr>
rmse	standard	2.498404e+05	Preprocessor1_Model1
rsq	standard	1.478900e-02	Preprocessor1_Model1

library(vip)

extract_workflow(xgb_best_fit) %>% 
  extract_fit_parsnip() %>% 
  vip(num_features = 15, geom = "point")

Airtable: ML Intervention

Akanimoh | AfroLogicInsect

2025-01-13

Data Glimpse

Exploratory Data Analysis

Machine Learning: Build two models

Linear Regression and XGBoost Models

Hyper-Parameter tuning XGBoost Model

`{r Model-Deloyment, warning=FALSE} # library(vetiver) # v <- extract_workflow(xgb_best_fit) |> # vetiver_model("salary-xgb") # v #`