nguyenletai16122001
/
predict_caLamviec_nhansu
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

final222

This commit is contained in:
letai2001 2026-01-18 02:16:48 +07:00
parent ab717695d5
commit 1e07746aa9
13 changed files with 0 additions and 312 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

BIN
models/best_model_gradient_boosting.pkl
View File

Binary file not shown.

49
models/columns_to_scale.txt
View File

@ -1,49 +0,0 @@
so_ca_cua_toa
num_tasks
num_cleaning_tasks
num_trash_collection_tasks
num_monitoring_tasks
num_room_cleaning_tasks
num_deep_cleaning_tasks
num_maintenance_tasks
num_support_tasks
num_other_tasks
num_wc_tasks
num_hallway_tasks
num_lobby_tasks
num_patient_room_tasks
num_clinic_room_tasks
num_surgery_room_tasks
num_outdoor_tasks
num_elevator_tasks
num_office_tasks
num_technical_room_tasks
cleaning_ratio
trash_collection_ratio
monitoring_ratio
room_cleaning_ratio
area_diversity
task_complexity_score
so_tang
so_cua_thang_may
dien_tich_ngoai_canh
dien_tich_sanh
dien_tich_hanh_lang
dien_tich_wc
dien_tich_phong
dien_tich_tham
doc_ham
vien_phan_quang
op_tuong
op_chan_tuong
ranh_thoat_nuoc
dien_tich_kinh
hour_start
hour_end
work_hours_numeric
tasks_per_hour
tasks_per_floor
wc_per_floor
cleaning_workload
total_area
area_per_floor

22
models/feature_selection_results.csv
View File

@ -1,22 +0,0 @@
num_features,feature_set,model,val_mae,val_r2
49,All Features,Gradient Boosting,3.634581261520732,-0.012930925436683394
20,Top 20,Random Forest,3.666060757466078,0.03739430732677618
20,Top 20,Gradient Boosting,3.7771892175715056,-0.06012882358872096
20,Top 20,XGBoost,3.641647574774335,0.015787720680236816
20,Top 20,LightGBM,3.748396263264872,0.049515999401856026
25,Top 25,Random Forest,3.614868221684315,0.06809266983693418
25,Top 25,Gradient Boosting,3.780284523284983,-0.05195262322328653
25,Top 25,XGBoost,3.760102046004842,-0.01795041561126709
25,Top 25,LightGBM,3.7433780829414327,0.04227330886423497
30,Top 30,Random Forest,3.6134273351492454,0.0712211389609152
30,Top 30,Gradient Boosting,3.8167183021556417,-0.08445111108289871
30,Top 30,XGBoost,3.8263845290528256,-0.036774635314941406
30,Top 30,LightGBM,3.6913964071303145,0.0628657702857146
35,Top 35,Random Forest,3.6403561822073054,0.07301240475906845
35,Top 35,Gradient Boosting,3.91511098479336,-0.11251667765570872
35,Top 35,XGBoost,3.884480058477166,-0.052086710929870605
35,Top 35,LightGBM,3.8490180856510774,0.0009015985532334625
40,Top 40,Random Forest,3.6265115302924293,0.07854970083459945
40,Top 40,Gradient Boosting,3.8841932753369783,-0.08916849029447449
40,Top 40,XGBoost,3.9312105964911117,-0.07247793674468994
40,Top 40,LightGBM,3.747326913269243,0.029446228984498002
1 num_features feature_set model val_mae val_r2
2 49 All Features Gradient Boosting 3.634581261520732 -0.012930925436683394
3 20 Top 20 Random Forest 3.666060757466078 0.03739430732677618
4 20 Top 20 Gradient Boosting 3.7771892175715056 -0.06012882358872096
5 20 Top 20 XGBoost 3.641647574774335 0.015787720680236816
6 20 Top 20 LightGBM 3.748396263264872 0.049515999401856026
7 25 Top 25 Random Forest 3.614868221684315 0.06809266983693418
8 25 Top 25 Gradient Boosting 3.780284523284983 -0.05195262322328653
9 25 Top 25 XGBoost 3.760102046004842 -0.01795041561126709
10 25 Top 25 LightGBM 3.7433780829414327 0.04227330886423497
11 30 Top 30 Random Forest 3.6134273351492454 0.0712211389609152
12 30 Top 30 Gradient Boosting 3.8167183021556417 -0.08445111108289871
13 30 Top 30 XGBoost 3.8263845290528256 -0.036774635314941406
14 30 Top 30 LightGBM 3.6913964071303145 0.0628657702857146
15 35 Top 35 Random Forest 3.6403561822073054 0.07301240475906845
16 35 Top 35 Gradient Boosting 3.91511098479336 -0.11251667765570872
17 35 Top 35 XGBoost 3.884480058477166 -0.052086710929870605
18 35 Top 35 LightGBM 3.8490180856510774 0.0009015985532334625
19 40 Top 40 Random Forest 3.6265115302924293 0.07854970083459945
20 40 Top 40 Gradient Boosting 3.8841932753369783 -0.08916849029447449
21 40 Top 40 XGBoost 3.9312105964911117 -0.07247793674468994
22 40 Top 40 LightGBM 3.747326913269243 0.029446228984498002

BIN
models/final_model_random_forest_top15.pkl
View File

Binary file not shown.

BIN
models/final_model_random_forest_top30.pkl
View File

Binary file not shown.

7
models/kfold_cv_results.csv
View File

@ -1,7 +0,0 @@
Model,Mean_MAE,Std_MAE,Mean_R2,Std_R2
Gradient Boosting,3.3027367118164057,0.8887355706131049,-0.10233682335888679,0.5154478192871355
XGBoost,3.323695227191383,0.7236534911368803,-0.14202730655670165,0.5722418063476958
Random Forest,3.352265185800016,0.8553895357341633,-0.06523395173555772,0.35068316384382126
LightGBM,3.401600572401869,0.8548539009927945,-0.006448200247211444,0.24861511037048697
Linear Regression,3.765361427841345,0.8860772035556437,-0.17797005506788818,0.5737173906817382
Decision Tree,3.8242160456852985,0.8261306554174034,-0.7565366143260073,1.4465792608602213
1 Model Mean_MAE Std_MAE Mean_R2 Std_R2
2 Gradient Boosting 3.3027367118164057 0.8887355706131049 -0.10233682335888679 0.5154478192871355
3 XGBoost 3.323695227191383 0.7236534911368803 -0.14202730655670165 0.5722418063476958
4 Random Forest 3.352265185800016 0.8553895357341633 -0.06523395173555772 0.35068316384382126
5 LightGBM 3.401600572401869 0.8548539009927945 -0.006448200247211444 0.24861511037048697
6 Linear Regression 3.765361427841345 0.8860772035556437 -0.17797005506788818 0.5737173906817382
7 Decision Tree 3.8242160456852985 0.8261306554174034 -0.7565366143260073 1.4465792608602213

8
models/model_comparison.csv
View File

@ -1,8 +0,0 @@
Model,Train_MAE,Train_RMSE,Train_R2,Val_MAE,Val_RMSE,Val_R2
Mean Baseline,3.956762233641091,6.008718072893705,0.0,5.108172961558765,9.743328619259394,0.0
Linear Regression,2.8487431324413386,4.306219455232728,0.48639568660771626,4.93124960105569,10.131361272284664,-0.0820337223062475
Decision Tree,1.4165098388571058,2.927374540160555,0.7626479824388336,3.384117292326248,7.458279107629457,0.41361553592413347
Random Forest,1.5188249096595474,2.651411767397892,0.8052889029203236,3.557118610267275,8.094827409344068,0.3092508183789415
Gradient Boosting,0.5002223075020469,1.163143548193567,0.9625283361855776,3.0356483741790528,7.556467123302565,0.39807443890459426
XGBoost,0.5888747572898865,1.2343001041144357,0.9578033685684204,3.2599263191223145,8.016751870107681,0.3225112557411194
LightGBM,1.527799546634308,2.6791994654613362,0.8011862391495337,3.602681095009421,8.17893904810045,0.29482138206791697
1 Model Train_MAE Train_RMSE Train_R2 Val_MAE Val_RMSE Val_R2
2 Mean Baseline 3.956762233641091 6.008718072893705 0.0 5.108172961558765 9.743328619259394 0.0
3 Linear Regression 2.8487431324413386 4.306219455232728 0.48639568660771626 4.93124960105569 10.131361272284664 -0.0820337223062475
4 Decision Tree 1.4165098388571058 2.927374540160555 0.7626479824388336 3.384117292326248 7.458279107629457 0.41361553592413347
5 Random Forest 1.5188249096595474 2.651411767397892 0.8052889029203236 3.557118610267275 8.094827409344068 0.3092508183789415
6 Gradient Boosting 0.5002223075020469 1.163143548193567 0.9625283361855776 3.0356483741790528 7.556467123302565 0.39807443890459426
7 XGBoost 0.5888747572898865 1.2343001041144357 0.9578033685684204 3.2599263191223145 8.016751870107681 0.3225112557411194
8 LightGBM 1.527799546634308 2.6791994654613362 0.8011862391495337 3.602681095009421 8.17893904810045 0.29482138206791697

BIN
models/scaler.pkl
View File

Binary file not shown.

BIN
models/scaler_final.pkl
View File

Binary file not shown.

15
models/selected_features_top15.txt
View File

@ -1,15 +0,0 @@
op_chan_tuong
loai_hinh
so_cua_thang_may
num_hallway_tasks
cleaning_ratio
op_tuong
area_diversity
num_office_tasks
tong_gio_lam
num_tasks
dien_tich_ngoai_canh
num_clinic_room_tasks
doc_ham
num_elevator_tasks
num_maintenance_tasks

30
models/selected_features_top30.txt
View File

@ -1,30 +0,0 @@
dien_tich_phong
tasks_per_hour
num_trash_collection_tasks
num_surgery_room_tasks
dien_tich_hanh_lang
area_diversity
num_lobby_tasks
hour_end
dien_tich_ngoai_canh
monitoring_ratio
num_patient_room_tasks
num_monitoring_tasks
tasks_per_floor
work_hours_numeric
trash_collection_ratio
dien_tich_tham
so_tang
so_ca_cua_toa
dien_tich_wc
num_wc_tasks
dien_tich_kinh
num_elevator_tasks
num_office_tasks
hour_start
cleaning_ratio
num_technical_room_tasks
ranh_thoat_nuoc
num_room_cleaning_tasks
num_deep_cleaning_tasks
num_tasks

77
models/training_report.json
View File

@ -1,77 +0,0 @@
{
"best_model": "Gradient Boosting",
"test_mae": 2.715735914594871,
"test_rmse": 4.8725908385837755,
"test_r2": 0.47268268124245805,
"val_mae": 2.9036818323527176,
"val_r2": 0.14702091178419308,
"train_size": 304,
"val_size": 66,
"test_size": 66,
"n_features": 63,
"feature_names": [
"so_ca_cua_toa",
"num_tasks",
"num_cleaning_tasks",
"num_trash_collection_tasks",
"num_monitoring_tasks",
"num_room_cleaning_tasks",
"num_deep_cleaning_tasks",
"num_maintenance_tasks",
"num_support_tasks",
"num_other_tasks",
"num_wc_tasks",
"num_hallway_tasks",
"num_lobby_tasks",
"num_patient_room_tasks",
"num_clinic_room_tasks",
"num_surgery_room_tasks",
"num_outdoor_tasks",
"num_elevator_tasks",
"num_office_tasks",
"num_technical_room_tasks",
"cleaning_ratio",
"trash_collection_ratio",
"monitoring_ratio",
"room_cleaning_ratio",
"area_diversity",
"task_complexity_score",
"loai_hinh",
"muc_do_luu_luong",
"so_tang",
"so_cua_thang_may",
"dien_tich_ngoai_canh",
"dien_tich_sanh",
"dien_tich_hanh_lang",
"dien_tich_wc",
"dien_tich_phong",
"dien_tich_tham",
"doc_ham",
"vien_phan_quang",
"op_tuong",
"op_chan_tuong",
"ranh_thoat_nuoc",
"dien_tich_kinh",
"hour_start",
"hour_end",
"work_hours_numeric",
"is_morning_shift",
"is_afternoon_shift",
"is_evening_shift",
"is_night_shift",
"tasks_per_hour",
"tasks_per_floor",
"wc_per_floor",
"cleaning_workload",
"total_area",
"area_per_floor",
"has_special_areas",
"loai_ca_24/24",
"loai_ca_Ca chiều",
"loai_ca_Ca gãy",
"loai_ca_Ca sáng",
"loai_ca_Ca đêm",
"loai_ca_Hành chính",
"loai_ca_Part time"
]
}

104
models/training_report_final.json
View File

@ -1,104 +0,0 @@
{
"timestamp": "2026-01-08 17:59:10",
"model_name": "Random Forest",
"num_features": 30,
"total_features_available": 49,
"training_samples": 356,
"test_samples": 89,
"test_mae": 2.959679790870802,
"test_mse": 22.52809628510788,
"test_rmse": 4.74637717476265,
"test_r2": 0.4146924609040473,
"baseline_val_mae": 3.634581261520732,
"improvement_vs_baseline": 0.6749014706499299,
"improvement_pct": 18.568892042533307,
"k_fold_cv": "Completed (5-fold)",
"feature_selection_methods": [
"Random Forest",
"Mutual Information",
"Correlation"
],
"selected_features": [
"dien_tich_phong",
"tasks_per_hour",
"num_trash_collection_tasks",
"num_surgery_room_tasks",
"dien_tich_hanh_lang",
"area_diversity",
"num_lobby_tasks",
"hour_end",
"dien_tich_ngoai_canh",
"monitoring_ratio",
"num_patient_room_tasks",
"num_monitoring_tasks",
"tasks_per_floor",
"work_hours_numeric",
"trash_collection_ratio",
"dien_tich_tham",
"so_tang",
"so_ca_cua_toa",
"dien_tich_wc",
"num_wc_tasks",
"dien_tich_kinh",
"num_elevator_tasks",
"num_office_tasks",
"hour_start",
"cleaning_ratio",
"num_technical_room_tasks",
"ranh_thoat_nuoc",
"num_room_cleaning_tasks",
"num_deep_cleaning_tasks",
"num_tasks"
],
"cols_to_scale": [
"so_ca_cua_toa",
"num_tasks",
"num_cleaning_tasks",
"num_trash_collection_tasks",
"num_monitoring_tasks",
"num_room_cleaning_tasks",
"num_deep_cleaning_tasks",
"num_maintenance_tasks",
"num_support_tasks",
"num_other_tasks",
"num_wc_tasks",
"num_hallway_tasks",
"num_lobby_tasks",
"num_patient_room_tasks",
"num_clinic_room_tasks",
"num_surgery_room_tasks",
"num_outdoor_tasks",
"num_elevator_tasks",
"num_office_tasks",
"num_technical_room_tasks",
"cleaning_ratio",
"trash_collection_ratio",
"monitoring_ratio",
"room_cleaning_ratio",
"area_diversity",
"task_complexity_score",
"so_tang",
"so_cua_thang_may",
"dien_tich_ngoai_canh",
"dien_tich_sanh",
"dien_tich_hanh_lang",
"dien_tich_wc",
"dien_tich_phong",
"dien_tich_tham",
"doc_ham",
"vien_phan_quang",
"op_tuong",
"op_chan_tuong",
"ranh_thoat_nuoc",
"dien_tich_kinh",
"hour_start",
"hour_end",
"work_hours_numeric",
"tasks_per_hour",
"tasks_per_floor",
"wc_per_floor",
"cleaning_workload",
"total_area",
"area_per_floor"
]
}