s452487/ium_452487
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
0dadba5cc1
ium_452487/dane.ipynb

348 KiB
Raw Blame History

1. Pobieranie zbioru danych

!pip install --user kaggle
Collecting kaggle
  Using cached kaggle-1.6.6.tar.gz (84 kB)
Requirement already satisfied: six>=1.10 in c:\users\adrian\appdata\roaming\python\python39\site-packages (from kaggle) (1.16.0)
Requirement already satisfied: certifi in c:\users\adrian\appdata\roaming\python\python39\site-packages (from kaggle) (2022.6.15)
Requirement already satisfied: python-dateutil in c:\users\adrian\appdata\roaming\python\python39\site-packages (from kaggle) (2.8.2)
Requirement already satisfied: requests in c:\users\adrian\appdata\roaming\python\python39\site-packages (from kaggle) (2.28.1)
Requirement already satisfied: tqdm in c:\users\adrian\appdata\roaming\python\python39\site-packages (from kaggle) (4.64.1)
Collecting python-slugify
  Using cached python_slugify-8.0.4-py2.py3-none-any.whl (10 kB)
Requirement already satisfied: urllib3 in c:\users\adrian\appdata\roaming\python\python39\site-packages (from kaggle) (1.26.11)
Requirement already satisfied: bleach in c:\users\adrian\miniconda3\lib\site-packages (from kaggle) (4.1.0)
Requirement already satisfied: webencodings in c:\users\adrian\miniconda3\lib\site-packages (from bleach->kaggle) (0.5.1)
Requirement already satisfied: packaging in c:\users\adrian\appdata\roaming\python\python39\site-packages (from bleach->kaggle) (22.0)
Collecting text-unidecode>=1.3
  Using cached text_unidecode-1.3-py2.py3-none-any.whl (78 kB)
Requirement already satisfied: idna<4,>=2.5 in c:\users\adrian\appdata\roaming\python\python39\site-packages (from requests->kaggle) (2.10)
Requirement already satisfied: charset-normalizer<3,>=2 in c:\users\adrian\appdata\roaming\python\python39\site-packages (from requests->kaggle) (2.1.0)
Requirement already satisfied: colorama in c:\users\adrian\appdata\roaming\python\python39\site-packages (from tqdm->kaggle) (0.4.5)
Building wheels for collected packages: kaggle
  Building wheel for kaggle (setup.py): started
  Building wheel for kaggle (setup.py): finished with status 'done'
  Created wheel for kaggle: filename=kaggle-1.6.6-py3-none-any.whl size=111961 sha256=3aa19c7655c19d77b65c2542567b6e34a57813227f1f2df6d0fd84accad6824f
  Stored in directory: c:\users\adrian\appdata\local\pip\cache\wheels\46\aa\c3\b3e421522fb5acdd7c366a05c5fc80787615bdeed207e7f79b
Successfully built kaggle
Installing collected packages: text-unidecode, python-slugify, kaggle
Successfully installed kaggle-1.6.6 python-slugify-8.0.4 text-unidecode-1.3

!kaggle datasets download -d kamilpytlak/personal-key-indicators-of-heart-disease/
Downloading personal-key-indicators-of-heart-disease.zip to C:\Users\Adrian\Desktop\Semestr 1 (II ST)\ML\zadania


  0%|          | 0.00/21.4M [00:00<?, ?B/s]
  5%|4         | 1.00M/21.4M [00:00<00:11, 1.80MB/s]
  9%|9         | 2.00M/21.4M [00:00<00:05, 3.51MB/s]
 19%|#8        | 4.00M/21.4M [00:00<00:02, 7.25MB/s]
 33%|###2      | 7.00M/21.4M [00:00<00:01, 12.0MB/s]
 51%|#####1    | 11.0M/21.4M [00:01<00:00, 17.9MB/s]
 65%|######5   | 14.0M/21.4M [00:01<00:00, 20.7MB/s]
 79%|#######9  | 17.0M/21.4M [00:01<00:00, 20.6MB/s]
 93%|#########3| 20.0M/21.4M [00:01<00:00, 22.7MB/s]
100%|##########| 21.4M/21.4M [00:01<00:00, 14.9MB/s]

#!unzip -o personal-key-indicators-of-heart-disease.zip #nie działa na Windowsie więc korzystam z modułu zipfile
'unzip' is not recognized as an internal or external command,
operable program or batch file.

import zipfile
with zipfile.ZipFile("personal-key-indicators-of-heart-disease.zip", 'r') as zip_ref:
    zip_ref.extractall("dataset_extracted")
import pandas as pd
# W pobranym zbiorze danych jest kilka podzbiorów więc celowo otwieram ten z NaN, żeby manualnie go oczyścić dla praktyki
df = pd.read_csv("dataset_extracted/2022/heart_2022_with_nans.csv")

Przeglądanie nieoczyszczonego datasetu

df.info()
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 445132 entries, 0 to 445131
Data columns (total 40 columns):
 #   Column                     Non-Null Count   Dtype  
---  ------                     --------------   -----  
 0   State                      445132 non-null  object 
 1   Sex                        445132 non-null  object 
 2   GeneralHealth              443934 non-null  object 
 3   PhysicalHealthDays         434205 non-null  float64
 4   MentalHealthDays           436065 non-null  float64
 5   LastCheckupTime            436824 non-null  object 
 6   PhysicalActivities         444039 non-null  object 
 7   SleepHours                 439679 non-null  float64
 8   RemovedTeeth               433772 non-null  object 
 9   HadHeartAttack             442067 non-null  object 
 10  HadAngina                  440727 non-null  object 
 11  HadStroke                  443575 non-null  object 
 12  HadAsthma                  443359 non-null  object 
 13  HadSkinCancer              441989 non-null  object 
 14  HadCOPD                    442913 non-null  object 
 15  HadDepressiveDisorder      442320 non-null  object 
 16  HadKidneyDisease           443206 non-null  object 
 17  HadArthritis               442499 non-null  object 
 18  HadDiabetes                444045 non-null  object 
 19  DeafOrHardOfHearing        424485 non-null  object 
 20  BlindOrVisionDifficulty    423568 non-null  object 
 21  DifficultyConcentrating    420892 non-null  object 
 22  DifficultyWalking          421120 non-null  object 
 23  DifficultyDressingBathing  421217 non-null  object 
 24  DifficultyErrands          419476 non-null  object 
 25  SmokerStatus               409670 non-null  object 
 26  ECigaretteUsage            409472 non-null  object 
 27  ChestScan                  389086 non-null  object 
 28  RaceEthnicityCategory      431075 non-null  object 
 29  AgeCategory                436053 non-null  object 
 30  HeightInMeters             416480 non-null  float64
 31  WeightInKilograms          403054 non-null  float64
 32  BMI                        396326 non-null  float64
 33  AlcoholDrinkers            398558 non-null  object 
 34  HIVTesting                 379005 non-null  object 
 35  FluVaxLast12               398011 non-null  object 
 36  PneumoVaxEver              368092 non-null  object 
 37  TetanusLast10Tdap          362616 non-null  object 
 38  HighRiskLastYear           394509 non-null  object 
 39  CovidPos                   394368 non-null  object 
dtypes: float64(6), object(34)
memory usage: 135.8+ MB

df.head()
State Sex GeneralHealth PhysicalHealthDays MentalHealthDays LastCheckupTime PhysicalActivities SleepHours RemovedTeeth HadHeartAttack ... HeightInMeters WeightInKilograms BMI AlcoholDrinkers HIVTesting FluVaxLast12 PneumoVaxEver TetanusLast10Tdap HighRiskLastYear CovidPos
0 Alabama Female Very good 0.0 0.0 Within past year (anytime less than 12 months ... No 8.0 NaN No ... NaN NaN NaN No No Yes No Yes, received tetanus shot but not sure what type No No
1 Alabama Female Excellent 0.0 0.0 NaN No 6.0 NaN No ... 1.60 68.04 26.57 No No No No No, did not receive any tetanus shot in the pa... No No
2 Alabama Female Very good 2.0 3.0 Within past year (anytime less than 12 months ... Yes 5.0 NaN No ... 1.57 63.50 25.61 No No No No NaN No Yes
3 Alabama Female Excellent 0.0 0.0 Within past year (anytime less than 12 months ... Yes 7.0 NaN No ... 1.65 63.50 23.30 No No Yes Yes No, did not receive any tetanus shot in the pa... No No
4 Alabama Female Fair 2.0 0.0 Within past year (anytime less than 12 months ... Yes 9.0 NaN No ... 1.57 53.98 21.77 Yes No No Yes No, did not receive any tetanus shot in the pa... No No

5 rows × 40 columns

df.describe()
PhysicalHealthDays MentalHealthDays SleepHours HeightInMeters WeightInKilograms BMI
count 434205.000000 436065.000000 439679.000000 416480.000000 403054.000000 396326.000000
mean 4.347919 4.382649 7.022983 1.702691 83.074470 28.529842
std 8.688912 8.387475 1.502425 0.107177 21.448173 6.554889
min 0.000000 0.000000 1.000000 0.910000 22.680000 12.020000
25% 0.000000 0.000000 6.000000 1.630000 68.040000 24.130000
50% 0.000000 0.000000 7.000000 1.700000 80.740000 27.440000
75% 3.000000 5.000000 8.000000 1.780000 95.250000 31.750000
max 30.000000 30.000000 24.000000 2.410000 292.570000 99.640000

Tylko 6 kolumn jest numeryczne na razie więc wiele statystyk nie zostaje wyświetlonych w tym podsumowaniu

Zbiór danych jest niezbalansowany, zmienna którą chcemy przewidzieć w znacznej większości przypadków wynosi 0:

df["HadHeartAttack"].value_counts().plot(kind="pie")
<AxesSubplot:ylabel='HadHeartAttack'>
df["HadHeartAttack"].value_counts()
No     416959
Yes     25108
Name: HadHeartAttack, dtype: int64

2. Podział na podzbiory (train / dev / test - 8:1:1)) i oversampling

from sklearn.model_selection import train_test_split
#Funkcji z sklearn musimy użyć dwukrotnie, bo dzieli tylko na dwa podzbiory
train, test_and_valid = train_test_split(df, test_size=0.2) #0.8 train, 0.2 test&valid

test, valid = train_test_split(test_and_valid, test_size=0.5) #0.1 test, 0.1 valid
train["HadHeartAttack"].value_counts()
No     333641
Yes     20042
Name: HadHeartAttack, dtype: int64

Zbiór treningowy jest nadal niezbalansowany więc zrobię prosty oversampling przez kopiowanie mniejszej klasy aż będą prawie równe

def oversample(dataset):
    num_true = len(dataset[dataset["HadHeartAttack"]=="Yes"])
    num_false = len(dataset[dataset["HadHeartAttack"]=="No"])
    num_oversampling_steps = num_false//num_true
    oversampled = dataset.copy()
    for x in range(num_oversampling_steps):
        oversampled = pd.concat([oversampled, dataset[dataset["HadHeartAttack"]=="Yes"]], ignore_index=True)
    return oversampled
train = oversample(train)
train["HadHeartAttack"].value_counts().plot(kind="pie")
<AxesSubplot:ylabel='HadHeartAttack'>
test["HadHeartAttack"].value_counts().plot(kind="pie")
<AxesSubplot:ylabel='HadHeartAttack'>
valid["HadHeartAttack"].value_counts().plot(kind="pie")
<AxesSubplot:ylabel='HadHeartAttack'>

Proporcje osób palących / niepalących w pierwotnym zbiorze danych:

df["SmokerStatus"].value_counts().plot(kind="pie")
<AxesSubplot:ylabel='SmokerStatus'>
df["ECigaretteUsage"].value_counts().plot(kind="pie")
<AxesSubplot:ylabel='ECigaretteUsage'>

Statystyki covidowe

df["CovidPos"].value_counts().plot(kind="pie")
<AxesSubplot:ylabel='CovidPos'>

Normalizacja część 1 - zamiana na kolumny liczbowe i kategoryczne

Kolumny zawierające stan zdrowia i podobne cechy opisane w sposób "poor/fair/good/excellent" etc. starałem się zamienić na liczbowe w sposób sensowny, rosnący względem pozytywnego aspektu tego czynnika zdrowotnego. Podobnie z tym jak często dana osoba paliła. Część kolumn zamieniłem na kategoryczne Kolumnę płci zamieniłem na numeryczną w celu późniejszego wykorzystania przez model, choć mialem wątpliwości co do robienia tego pod względem poprawności politycznej

df["Sex"].unique()
array(['Female', 'Male'], dtype=object)
df["GeneralHealth"].unique()
array(['Very good', 'Excellent', 'Fair', 'Poor', 'Good', nan],
      dtype=object)
health_map = {
    "Excellent": 5,
    "Very good": 4,
    "Good": 3,
    "Fair": 2,
    "Poor": 1
}
for col in df:
    print(f"{col}:")
    print(df[col].unique())
State:
['Alabama' 'Alaska' 'Arizona' 'Arkansas' 'California' 'Colorado'
 'Connecticut' 'Delaware' 'District of Columbia' 'Florida' 'Georgia'
 'Hawaii' 'Idaho' 'Illinois' 'Indiana' 'Iowa' 'Kansas' 'Kentucky'
 'Louisiana' 'Maine' 'Maryland' 'Massachusetts' 'Michigan' 'Minnesota'
 'Mississippi' 'Missouri' 'Montana' 'Nebraska' 'Nevada' 'New Hampshire'
 'New Jersey' 'New Mexico' 'New York' 'North Carolina' 'North Dakota'
 'Ohio' 'Oklahoma' 'Oregon' 'Pennsylvania' 'Rhode Island' 'South Carolina'
 'South Dakota' 'Tennessee' 'Texas' 'Utah' 'Vermont' 'Virginia'
 'Washington' 'West Virginia' 'Wisconsin' 'Wyoming' 'Guam' 'Puerto Rico'
 'Virgin Islands']
Sex:
['Female' 'Male']
GeneralHealth:
['Very good' 'Excellent' 'Fair' 'Poor' 'Good' nan]
PhysicalHealthDays:
[ 0.  2.  1.  8.  5. 30.  4. 23. 14. nan 15.  3. 10.  7. 25.  6. 21. 20.
 29. 16.  9. 27. 28. 12. 13. 11. 26. 17. 24. 19. 18. 22.]
MentalHealthDays:
[ 0.  3.  9.  5. 15. 20. 14. 10. 18.  1. nan  2. 30.  4.  6.  7. 25.  8.
 22. 29. 27. 21. 12. 28. 16. 13. 26. 17. 11. 23. 19. 24.]
LastCheckupTime:
['Within past year (anytime less than 12 months ago)' nan
 'Within past 2 years (1 year but less than 2 years ago)'
 'Within past 5 years (2 years but less than 5 years ago)'
 '5 or more years ago']
PhysicalActivities:
['No' 'Yes' nan]
SleepHours:
[ 8.  6.  5.  7.  9.  4. 10.  1. 12. nan 18.  3.  2. 11. 16. 15. 13. 14.
 20. 23. 17. 24. 22. 19. 21.]
RemovedTeeth:
[nan 'None of them' '1 to 5' '6 or more, but not all' 'All']
HadHeartAttack:
['No' 'Yes' nan]
HadAngina:
['No' 'Yes' nan]
HadStroke:
['No' 'Yes' nan]
HadAsthma:
['No' 'Yes' nan]
HadSkinCancer:
['No' 'Yes' nan]
HadCOPD:
['No' 'Yes' nan]
HadDepressiveDisorder:
['No' 'Yes' nan]
HadKidneyDisease:
['No' 'Yes' nan]
HadArthritis:
['No' 'Yes' nan]
HadDiabetes:
['Yes' 'No' 'No, pre-diabetes or borderline diabetes' nan
 'Yes, but only during pregnancy (female)']
DeafOrHardOfHearing:
['No' nan 'Yes']
BlindOrVisionDifficulty:
['No' 'Yes' nan]
DifficultyConcentrating:
['No' nan 'Yes']
DifficultyWalking:
['No' 'Yes' nan]
DifficultyDressingBathing:
['No' nan 'Yes']
DifficultyErrands:
['No' 'Yes' nan]
SmokerStatus:
['Never smoked' 'Current smoker - now smokes some days' 'Former smoker'
 nan 'Current smoker - now smokes every day']
ECigaretteUsage:
['Not at all (right now)' 'Never used e-cigarettes in my entire life' nan
 'Use them every day' 'Use them some days']
ChestScan:
['No' 'Yes' nan]
RaceEthnicityCategory:
['White only, Non-Hispanic' 'Black only, Non-Hispanic'
 'Other race only, Non-Hispanic' 'Multiracial, Non-Hispanic' nan
 'Hispanic']
AgeCategory:
['Age 80 or older' 'Age 55 to 59' nan 'Age 40 to 44' 'Age 75 to 79'
 'Age 70 to 74' 'Age 65 to 69' 'Age 60 to 64' 'Age 50 to 54'
 'Age 45 to 49' 'Age 35 to 39' 'Age 25 to 29' 'Age 30 to 34'
 'Age 18 to 24']
HeightInMeters:
[ nan 1.6  1.57 1.65 1.8  1.63 1.7  1.68 1.73 1.55 1.93 1.88 1.78 1.85
 1.75 1.52 1.83 1.91 1.96 1.5  1.45 1.42 1.24 1.47 1.22 1.98 2.03 2.01
 1.3  1.4  1.35 1.82 1.67 1.76 2.11 1.37 1.64 1.71 2.16 2.26 0.91 2.06
 1.14 1.74 1.51 1.53 1.69 1.56 1.84 1.9  1.54 1.72 1.87 1.61 1.49 1.59
 1.58 1.62 1.79 1.46 1.89 2.13 0.99 2.08 2.21 1.32 2.18 1.77 2.36 1.25
 1.66 1.86 1.95 1.19 1.05 1.48 1.03 1.18 1.81 1.38 1.44 1.07 1.27 1.2
 1.17 1.04 2.24 1.1  1.43 1.92 2.05 1.12 2.41 2.34 0.97 1.06 1.15 2.29
 1.16 1.09 0.92 2.07 1.   1.08 1.02 1.33 2.   2.02 1.94 0.95]
WeightInKilograms:
[   nan  68.04  63.5   53.98  84.82  62.6   73.48  81.65  74.84  59.42
  85.28 106.59  71.21  64.41  61.23  90.72  65.77  66.22  80.29  86.18
  47.63 107.05  57.15 105.23  77.11  56.7   79.38 113.4  102.06  59.87
 104.33  53.52  61.69 136.08  34.47  99.79 127.01  78.93  95.25  58.97
  92.08  72.57  83.91  49.9  117.93  71.67 102.97  62.14  83.46  54.43
  94.35  60.78 117.03  65.32  76.66  88.45  89.81  74.39  68.95  79.83
 108.41  90.26  55.79  91.63  47.17  78.02  50.8   91.17  84.37 145.15
  93.89 122.47  48.99  73.94  88.9   80.74  81.19 158.76  97.52  51.71
  82.55  76.2   68.49  75.3   70.31  63.05  60.33 115.67  86.64 108.86
  92.53 124.74  43.09  58.51  63.96  92.99  44.45 128.82  98.88  45.36
 110.68  46.72  58.06  73.03  95.71 131.09  78.47  69.4   85.73  67.59
 103.87 120.2   88.    54.88 111.58  52.16  77.56 126.55  94.8  123.83
  89.36  75.75  69.85 112.49  82.1  106.14  57.61  70.76 148.78  96.16
  67.13  48.08 163.29 109.77 100.7  142.88  64.86 111.13 121.11  55.34
 101.6   93.44 117.48 120.66  66.68  44.91 132.   107.5  107.95  36.29
 103.42  87.09  83.01  56.25  96.62 134.26  97.07  34.93  99.34  72.12
  49.44 122.02  98.43 129.73 181.44  52.62 121.56 110.22  48.53 140.61
 156.49 116.57  87.54  44.   114.31  31.75  97.98 101.15 112.04 100.24
 113.85 154.22 118.39 133.81 149.69  41.73 119.75 138.35 151.95 129.27
 131.54 104.78 132.45 102.51 116.12  40.37 105.69 136.98 195.04  53.07
 132.9  124.28 112.94 114.76  45.81 119.29 167.83  51.26 172.37 162.39
  46.27 127.91 123.38  38.56 130.63 143.34 115.21 166.92 135.17 109.32
 135.62 204.12 127.46 118.84 139.25 126.1  122.92 151.5  133.36  42.64
  50.35  80.   190.51  37.19 147.87  35.38 144.24 149.23  37.65  86.
 147.42 281.   165.56 162.84 155.58  70.   137.89 189.6  206.38 148.32
  42.18 153.77  38.1   90.   176.9  191.87 249.48  67.    95.    82.
 170.1   62.    40.82  53.   139.71 130.18 100.   165.11  64.    43.54
  24.   134.72 141.52 125.19  75.    60.    34.02 164.65  30.84 250.
  58.    76.    73.   112.    74.    55.   200.    54.    66.    72.
 152.41  39.46 220.    41.28 168.28 188.24  59.    46.   265.   238.14
 168.74 145.   190.    93.   159.66  78.    50.   185.07  91.   104.
 165.   183.7   33.57 161.93  68.   125.65 134.   130.    32.21 143.79
  69.   179.17  63.   105.   210.92  65.    32.   292.57 280.    85.
 174.63  56.   128.37  87.    39.92  83.   169.64 156.04 177.   121.
 151.05  89.   146.96 146.06  98.   166.47  36.74 171.46 227.25  29.48
 190.06 161.03  35.83 226.8  175.09 138.8  240.4  158.3  170.55  61.
 137.44 145.6  141.07 155.13  52.   120.    57.    77.    27.22  25.4
 240.    96.    47.   115.    41.    45.   170.   150.59 272.16  26.31
  48.    39.01 236.    92.   197.31 156.    84.    94.    29.03  49.
  79.   157.85 192.78 255.   108.   185.   222.26 229.97 180.    81.
  24.95  71.    26.   107.   101.   208.65 140.   175.   111.   110.
 141.97  22.68 284.86 136.53 210.   103.   185.97 140.16 146.51  24.49
  25.85 150.   102.   229.52  23.59 125.   163.    38.   135.   176.45
 185.52 152.86 232.69 124.   192.32 186.88 118.   160.12 160.   193.68
 201.85 144.7  184.16 142.43 169.   166.01  32.66 180.53 196.41  51.
  40.   171.91 195.95  33.11 153.31 159.21 164.2  219.99 215.46 182.34
  30.   160.57 173.27 158.   213.19 276.24 199.58 175.99 235.87 217.72
 200.03 230.88 146.    24.04 178.72 150.14 157.4  163.75 191.42 174.18
  28.58  97.   256.28 205.48 161.48 178.26 179.62 205.02 254.01 154.68
 209.56 201.4  234.96 177.81 200.49 231.79 227.7  273.52 189.15 173.73
 183.25 167.38 211.83 223.62 228.61  30.39 197.77 184.61 250.38 181.89
  31.3  290.3  285.   113.   242.67 231.33 180.08 202.76 176.   188.69
 206.84 164.   156.94 114.   122.   222.   137.   166.   180.98 272.
 172.82 274.42 234.51 199.13 244.94 203.21  23.13 265.35 198.22 263.08
 216.82 154.   169.19 239.04 177.35 210.47 224.98 117.    37.   126.
 273.06 203.66 252.2  238.59 194.59 187.33 221.35 162.   224.53  23.
 223.17 187.79 212.73 152.   233.6  193.23 205.   229.06 230.   247.21
  99.    28.12 230.42 175.54 205.93 171.    26.76 212.28 217.   280.32
 281.68 248.57 195.    42.   258.55 215.   116.    28.   123.   186.43
 228.16 119.   219.09 214.55 278.96 182.8  138.   217.27 246.3  189.  ]
BMI:
[  nan 26.57 25.61 ... 13.51 28.39 48.63]
AlcoholDrinkers:
['No' 'Yes' nan]
HIVTesting:
['No' 'Yes' nan]
FluVaxLast12:
['Yes' 'No' nan]
PneumoVaxEver:
['No' 'Yes' nan]
TetanusLast10Tdap:
['Yes, received tetanus shot but not sure what type'
 'No, did not receive any tetanus shot in the past 10 years' nan
 'Yes, received Tdap' 'Yes, received tetanus shot, but not Tdap']
HighRiskLastYear:
['No' nan 'Yes']
CovidPos:
['No' 'Yes' nan
 'Tested positive using home test without a health professional']

from collections import defaultdict
def normalize_dataset(dataset):
    dataset["GeneralHealth"] = dataset["GeneralHealth"].map(defaultdict(lambda: float('NaN'), health_map), na_action='ignore')
    dataset["Sex"] = dataset["Sex"].map({"Female":0,"Male":1}).astype(float) #Zamiana z kolumn tekstowych na numeryczne
    dataset.rename(columns ={"Sex":"Male"},inplace=True)
    dataset["State"] = dataset["State"].astype('category')
    dataset["PhysicalHealthDays"].astype(float)
    dataset["MentalHealthDays"].astype(float)
    dataset["LastCheckupTime"] = dataset["LastCheckupTime"].fillna("Unknown").astype('category') # Potem korzystam z fillna-->median ale nie działa to na kolumnach kategorycznych więc wykonuję to przed konwersją
    dataset["PhysicalActivities"]= dataset["PhysicalActivities"].map({"No":0,"Yes":1})
    dataset["SleepHours"].astype(float)
    dataset["RemovedTeeth"] = dataset["RemovedTeeth"].map(defaultdict(lambda: float('NaN'), {"None of them":0,"1 to 5":1, "6 or more, but not all":2, "All":3}), na_action='ignore')
    dataset["HadHeartAttack"]= dataset["HadHeartAttack"].map({"No":0,"Yes":1})
    dataset["HadAngina"]= dataset["HadAngina"].map({"No":0,"Yes":1})
    dataset["HadStroke"]= dataset["HadStroke"].map({"No":0,"Yes":1})
    dataset["HadAsthma"]= dataset["HadAsthma"].map({"No":0,"Yes":1})
    dataset["HadSkinCancer"]= dataset["HadSkinCancer"].map({"No":0,"Yes":1})
    dataset["HadCOPD"]= dataset["HadCOPD"].map({"No":0,"Yes":1})
    dataset["HadDepressiveDisorder"]= dataset["HadDepressiveDisorder"].map({"No":0,"Yes":1})
    dataset["HadKidneyDisease"]= dataset["HadKidneyDisease"].map({"No":0,"Yes":1})
    dataset["HadArthritis"]= dataset["HadArthritis"].map({"No":0,"Yes":1})
    dataset["HadDiabetes"]= dataset["HadDiabetes"].map({"No":0,"Yes, but only during pregnancy (female)":1,"No, pre-diabetes or borderline diabetes":2,"Yes":3})

    dataset["DeafOrHardOfHearing"]= dataset["DeafOrHardOfHearing"].map({"No":0,"Yes":1})
    dataset["BlindOrVisionDifficulty"]= dataset["BlindOrVisionDifficulty"].map({"No":0,"Yes":1})
    dataset["DifficultyConcentrating"]= dataset["DifficultyConcentrating"].map({"No":0,"Yes":1})
    dataset["DifficultyWalking"]= dataset["DifficultyWalking"].map({"No":0,"Yes":1})
    dataset["DifficultyDressingBathing"]= dataset["DifficultyDressingBathing"].map({"No":0,"Yes":1})
    dataset["DifficultyErrands"]= dataset["DifficultyErrands"].map({"No":0,"Yes":1})
    dataset["SmokerStatus"]= dataset["SmokerStatus"].map({"Never smoked":0,"Current smoker - now smokes some days":1,"Former smoker":2,"Current smoker - now smokes every day":3})
    dataset["ECigaretteUsage"]= dataset["ECigaretteUsage"].map({"Never used e-cigarettes in my entire life":0,"Not at all (right now)":1,"Use them some days":2,"Use them every day":3})
    dataset["ChestScan"]= dataset["ChestScan"].map({"No":0,"Yes":1})
    dataset["RaceEthnicityCategory"] = dataset["RaceEthnicityCategory"].fillna("Unknown").astype('category')
    dataset["AgeCategory"] = dataset["AgeCategory"].fillna("Unknown").astype('category')
    dataset["HeightInMeters"] = dataset["HeightInMeters"].astype(float)
    dataset["WeightInKilograms"] = dataset["WeightInKilograms"].astype(float)
    dataset["BMI"] = dataset["BMI"].astype(float)
    dataset["AlcoholDrinkers"]= dataset["AlcoholDrinkers"].map({"No":0,"Yes":1})
    dataset["HIVTesting"]= dataset["HIVTesting"].map({"No":0,"Yes":1})
    dataset["FluVaxLast12"]= dataset["FluVaxLast12"].map({"No":0,"Yes":1})
    dataset["PneumoVaxEver"]= dataset["PneumoVaxEver"].map({"No":0,"Yes":1})
    dataset["TetanusLast10Tdap"]= dataset["TetanusLast10Tdap"].apply(lambda x: float('NaN') if type(x)!=str else 1.0 if 'Yes,' in x else 1.0 if 'No,' in x else float('NaN'))
    dataset["HighRiskLastYear"]= dataset["HighRiskLastYear"].map({"No":0,"Yes":1})
    dataset["CovidPos"]= dataset["CovidPos"].map({"No":0,"Yes":1})

Zbiór test przed zmianą typu danych

test.head()
State Sex GeneralHealth PhysicalHealthDays MentalHealthDays LastCheckupTime PhysicalActivities SleepHours RemovedTeeth HadHeartAttack ... HeightInMeters WeightInKilograms BMI AlcoholDrinkers HIVTesting FluVaxLast12 PneumoVaxEver TetanusLast10Tdap HighRiskLastYear CovidPos
339824 South Dakota Female Good 3.0 21.0 Within past year (anytime less than 12 months ... Yes 8.0 6 or more, but not all No ... 1.60 52.16 20.37 No Yes Yes Yes Yes, received Tdap No No
127927 Kansas Female Good 30.0 0.0 Within past year (anytime less than 12 months ... Yes 10.0 1 to 5 No ... 1.68 97.52 34.70 No No Yes Yes NaN No No
362523 Utah Male Excellent 0.0 0.0 Within past year (anytime less than 12 months ... Yes 7.0 1 to 5 No ... 1.83 113.85 34.04 No No No No Yes, received Tdap No Yes
183687 Michigan Male Good 0.0 7.0 Within past year (anytime less than 12 months ... Yes 8.0 None of them No ... 1.78 83.91 26.54 Yes No Yes Yes Yes, received Tdap No No
191905 Michigan Female Very good 0.0 0.0 Within past year (anytime less than 12 months ... Yes 7.0 None of them No ... 1.57 68.04 27.44 Yes Yes No Yes Yes, received tetanus shot but not sure what type No No

5 rows × 40 columns

Zbiór test po zmianie typu danych

normalize_dataset(test)
test.head()
State Male GeneralHealth PhysicalHealthDays MentalHealthDays LastCheckupTime PhysicalActivities SleepHours RemovedTeeth HadHeartAttack ... HeightInMeters WeightInKilograms BMI AlcoholDrinkers HIVTesting FluVaxLast12 PneumoVaxEver TetanusLast10Tdap HighRiskLastYear CovidPos
339824 South Dakota 0.0 3.0 3.0 21.0 Within past year (anytime less than 12 months ... 1.0 8.0 2.0 0.0 ... 1.60 52.16 20.37 0.0 1.0 1.0 1.0 1.0 0.0 0.0
127927 Kansas 0.0 3.0 30.0 0.0 Within past year (anytime less than 12 months ... 1.0 10.0 1.0 0.0 ... 1.68 97.52 34.70 0.0 0.0 1.0 1.0 NaN 0.0 0.0
362523 Utah 1.0 5.0 0.0 0.0 Within past year (anytime less than 12 months ... 1.0 7.0 1.0 0.0 ... 1.83 113.85 34.04 0.0 0.0 0.0 0.0 1.0 0.0 1.0
183687 Michigan 1.0 3.0 0.0 7.0 Within past year (anytime less than 12 months ... 1.0 8.0 0.0 0.0 ... 1.78 83.91 26.54 1.0 0.0 1.0 1.0 1.0 0.0 0.0
191905 Michigan 0.0 4.0 0.0 0.0 Within past year (anytime less than 12 months ... 1.0 7.0 0.0 0.0 ... 1.57 68.04 27.44 1.0 1.0 0.0 1.0 1.0 0.0 0.0

5 rows × 40 columns

test.info()
<class 'pandas.core.frame.DataFrame'>
Int64Index: 44513 entries, 339824 to 52161
Data columns (total 40 columns):
 #   Column                     Non-Null Count  Dtype   
---  ------                     --------------  -----   
 0   State                      44513 non-null  category
 1   Male                       44513 non-null  float64 
 2   GeneralHealth              44393 non-null  float64 
 3   PhysicalHealthDays         43469 non-null  float64 
 4   MentalHealthDays           43622 non-null  float64 
 5   LastCheckupTime            44513 non-null  category
 6   PhysicalActivities         44408 non-null  float64 
 7   SleepHours                 44008 non-null  float64 
 8   RemovedTeeth               43413 non-null  float64 
 9   HadHeartAttack             44182 non-null  float64 
 10  HadAngina                  44074 non-null  float64 
 11  HadStroke                  44368 non-null  float64 
 12  HadAsthma                  44339 non-null  float64 
 13  HadSkinCancer              44184 non-null  float64 
 14  HadCOPD                    44299 non-null  float64 
 15  HadDepressiveDisorder      44218 non-null  float64 
 16  HadKidneyDisease           44320 non-null  float64 
 17  HadArthritis               44243 non-null  float64 
 18  HadDiabetes                44411 non-null  float64 
 19  DeafOrHardOfHearing        42485 non-null  float64 
 20  BlindOrVisionDifficulty    42387 non-null  float64 
 21  DifficultyConcentrating    42169 non-null  float64 
 22  DifficultyWalking          42172 non-null  float64 
 23  DifficultyDressingBathing  42182 non-null  float64 
 24  DifficultyErrands          41999 non-null  float64 
 25  SmokerStatus               41005 non-null  float64 
 26  ECigaretteUsage            41003 non-null  float64 
 27  ChestScan                  38958 non-null  float64 
 28  RaceEthnicityCategory      44513 non-null  category
 29  AgeCategory                44513 non-null  category
 30  HeightInMeters             41714 non-null  float64 
 31  WeightInKilograms          40397 non-null  float64 
 32  BMI                        39724 non-null  float64 
 33  AlcoholDrinkers            39956 non-null  float64 
 34  HIVTesting                 38018 non-null  float64 
 35  FluVaxLast12               39886 non-null  float64 
 36  PneumoVaxEver              36860 non-null  float64 
 37  TetanusLast10Tdap          36315 non-null  float64 
 38  HighRiskLastYear           39538 non-null  float64 
 39  CovidPos                   38114 non-null  float64 
dtypes: category(4), float64(36)
memory usage: 12.7 MB

normalize_dataset(train)
normalize_dataset(valid)

Statystyki dla zbiorów po zamianie na kolumny numeryczne

_50. centyl to mediana

train.describe()
Male GeneralHealth PhysicalHealthDays MentalHealthDays PhysicalActivities SleepHours RemovedTeeth HadHeartAttack HadAngina HadStroke ... HeightInMeters WeightInKilograms BMI AlcoholDrinkers HIVTesting FluVaxLast12 PneumoVaxEver TetanusLast10Tdap HighRiskLastYear CovidPos
count 676777.000000 674433.000000 655630.000000 660417.000000 674814.000000 665609.000000 654359.000000 674355.000000 657726.000000 672927.000000 ... 637313.000000 619546.000000 611024.000000 606636.000000 572562.000000 605920.000000 570114.000000 554467.0 600540.000000 585150.000000
mean 0.538139 3.055519 6.737547 4.863972 0.690146 7.032336 0.983081 0.505244 0.264549 0.117193 ... 1.707193 84.657015 28.917363 0.456819 0.325787 0.569879 0.527672 1.0 0.035087 0.273055
std 0.498544 1.137862 10.713287 9.115863 0.462434 1.726387 1.019679 0.499973 0.441093 0.321650 ... 0.108002 21.753692 6.607455 0.498132 0.468668 0.495093 0.499234 0.0 0.183999 0.445529
min 0.000000 1.000000 0.000000 0.000000 0.000000 1.000000 0.000000 0.000000 0.000000 0.000000 ... 0.910000 22.680000 12.050000 0.000000 0.000000 0.000000 0.000000 1.0 0.000000 0.000000
25% 0.000000 2.000000 0.000000 0.000000 0.000000 6.000000 0.000000 0.000000 0.000000 0.000000 ... 1.630000 69.400000 24.410000 0.000000 0.000000 0.000000 0.000000 1.0 0.000000 0.000000
50% 1.000000 3.000000 0.000000 0.000000 1.000000 7.000000 1.000000 1.000000 0.000000 0.000000 ... 1.700000 81.650000 27.890000 0.000000 0.000000 1.000000 1.000000 1.0 0.000000 0.000000
75% 1.000000 4.000000 10.000000 5.000000 1.000000 8.000000 2.000000 1.000000 1.000000 0.000000 ... 1.780000 96.160000 32.260000 1.000000 1.000000 1.000000 1.000000 1.0 0.000000 1.000000
max 1.000000 5.000000 30.000000 30.000000 1.000000 24.000000 3.000000 1.000000 1.000000 1.000000 ... 2.410000 292.570000 97.650000 1.000000 1.000000 1.000000 1.000000 1.0 1.000000 1.000000

8 rows × 36 columns

test.describe()
Male GeneralHealth PhysicalHealthDays MentalHealthDays PhysicalActivities SleepHours RemovedTeeth HadHeartAttack HadAngina HadStroke ... HeightInMeters WeightInKilograms BMI AlcoholDrinkers HIVTesting FluVaxLast12 PneumoVaxEver TetanusLast10Tdap HighRiskLastYear CovidPos
count 44513.000000 44393.000000 43469.000000 43622.000000 44408.000000 44008.000000 43413.000000 44182.000000 44074.000000 44368.000000 ... 41714.000000 40397.000000 39724.000000 39956.000000 38018.000000 39886.000000 36860.000000 36315.0 39538.000000 38114.000000
mean 0.471593 3.441511 4.275001 4.298221 0.760021 7.036584 0.685302 0.057241 0.061056 0.042869 ... 1.703194 83.021746 28.512326 0.529532 0.342233 0.527002 0.413592 1.0 0.043427 0.289500
std 0.499198 1.050924 8.588663 8.299250 0.427075 1.512667 0.884912 0.232304 0.239436 0.202563 ... 0.107438 21.551394 6.596149 0.499133 0.474463 0.499277 0.492484 0.0 0.203818 0.453536
min 0.000000 1.000000 0.000000 0.000000 0.000000 1.000000 0.000000 0.000000 0.000000 0.000000 ... 0.910000 22.680000 12.020000 0.000000 0.000000 0.000000 0.000000 1.0 0.000000 0.000000
25% 0.000000 3.000000 0.000000 0.000000 1.000000 6.000000 0.000000 0.000000 0.000000 0.000000 ... 1.630000 68.040000 24.030000 0.000000 0.000000 0.000000 0.000000 1.0 0.000000 0.000000
50% 0.000000 3.000000 0.000000 0.000000 1.000000 7.000000 0.000000 0.000000 0.000000 0.000000 ... 1.700000 80.740000 27.410000 1.000000 0.000000 1.000000 0.000000 1.0 0.000000 0.000000
75% 1.000000 4.000000 3.000000 4.000000 1.000000 8.000000 1.000000 0.000000 0.000000 0.000000 ... 1.780000 95.250000 31.650000 1.000000 1.000000 1.000000 1.000000 1.0 0.000000 1.000000
max 1.000000 5.000000 30.000000 30.000000 1.000000 24.000000 3.000000 1.000000 1.000000 1.000000 ... 2.340000 290.300000 97.650000 1.000000 1.000000 1.000000 1.000000 1.0 1.000000 1.000000

8 rows × 36 columns

valid.describe()
Male GeneralHealth PhysicalHealthDays MentalHealthDays PhysicalActivities SleepHours RemovedTeeth HadHeartAttack HadAngina HadStroke ... HeightInMeters WeightInKilograms BMI AlcoholDrinkers HIVTesting FluVaxLast12 PneumoVaxEver TetanusLast10Tdap HighRiskLastYear CovidPos
count 44514.000000 44388.000000 43458.000000 43578.000000 44401.000000 43966.000000 43360.000000 44202.000000 44031.000000 44344.000000 ... 41677.000000 40327.000000 39626.000000 39950.000000 38041.000000 39885.000000 36926.000000 36250.0 39535.000000 38212.000000
mean 0.469043 3.434554 4.355470 4.379022 0.758361 7.013010 0.684732 0.057396 0.058936 0.042463 ... 1.702146 82.981070 28.521370 0.527910 0.342525 0.525285 0.412771 1.0 0.044846 0.291872
std 0.499046 1.051996 8.718506 8.383576 0.428081 1.491967 0.882396 0.232600 0.235507 0.201646 ... 0.106978 21.512676 6.622255 0.499227 0.474560 0.499367 0.492339 0.0 0.206969 0.454630
min 0.000000 1.000000 0.000000 0.000000 0.000000 1.000000 0.000000 0.000000 0.000000 0.000000 ... 0.910000 22.680000 12.160000 0.000000 0.000000 0.000000 0.000000 1.0 0.000000 0.000000
25% 0.000000 3.000000 0.000000 0.000000 1.000000 6.000000 0.000000 0.000000 0.000000 0.000000 ... 1.630000 68.040000 24.030000 0.000000 0.000000 0.000000 0.000000 1.0 0.000000 0.000000
50% 0.000000 3.000000 0.000000 0.000000 1.000000 7.000000 0.000000 0.000000 0.000000 0.000000 ... 1.700000 79.830000 27.400000 1.000000 0.000000 1.000000 0.000000 1.0 0.000000 0.000000
75% 1.000000 4.000000 3.000000 5.000000 1.000000 8.000000 1.000000 0.000000 0.000000 0.000000 ... 1.780000 95.250000 31.750000 1.000000 1.000000 1.000000 1.000000 1.0 0.000000 1.000000
max 1.000000 5.000000 30.000000 30.000000 1.000000 24.000000 3.000000 1.000000 1.000000 1.000000 ... 2.360000 263.080000 99.640000 1.000000 1.000000 1.000000 1.000000 1.0 1.000000 1.000000

8 rows × 36 columns

Wydaje się być korelacja między masą ciała i zawałem:

import seaborn as sns
sns.set_theme()
g = sns.catplot(
    data=train, kind="bar",
    x="GeneralHealth", y="WeightInKilograms", hue="HadHeartAttack",
    errorbar="sd", palette="dark", alpha=.6, height=6
)
g.despine(left=True)
g.set_axis_labels("General health index", "Body mass (kg)")
g.legend.set_title("Had heart attack")

Osoby palące częsciej miały zawał:

valid.groupby('SmokerStatus', as_index=False)['HadHeartAttack'].mean()
SmokerStatus HadHeartAttack
0 0.0 0.037883
1 1.0 0.072598
2 2.0 0.088887
3 3.0 0.090192

Osoby z gorszym wskaźnikiem "GeneralHealth" w tym zbiorze danych częściej miały zawał:

valid.groupby('GeneralHealth', as_index=False)['HadHeartAttack'].mean()
GeneralHealth HadHeartAttack
0 1.0 0.228411
1 2.0 0.129270
2 3.0 0.056693
3 4.0 0.027336
4 5.0 0.014743 
valid.pivot_table('HadHeartAttack',index='GeneralHealth', columns='SmokerStatus')
SmokerStatus 0.0 1.0 2.0 3.0
GeneralHealth
1.0 0.194640 0.310680 0.257100 0.222222
2.0 0.090772 0.146429 0.184443 0.155059
3.0 0.039989 0.031068 0.091645 0.060469
4.0 0.021611 0.032070 0.035265 0.048292
5.0 0.011078 0.012579 0.026298 0.018315

Normalizacja część 2 - Skalowanie kolumn numerycznych do 0-1

from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler()
def scale_float_columns(dataset):
    numerical_columns = list(dataset.select_dtypes(include=['float64']).columns)
    dataset[numerical_columns] = scaler.fit_transform(dataset[numerical_columns])
test.head()
State Male GeneralHealth PhysicalHealthDays MentalHealthDays LastCheckupTime PhysicalActivities SleepHours RemovedTeeth HadHeartAttack ... HeightInMeters WeightInKilograms BMI AlcoholDrinkers HIVTesting FluVaxLast12 PneumoVaxEver TetanusLast10Tdap HighRiskLastYear CovidPos
339824 South Dakota 0.0 3.0 3.0 21.0 Within past year (anytime less than 12 months ... 1.0 8.0 2.0 0.0 ... 1.60 52.16 20.37 0.0 1.0 1.0 1.0 1.0 0.0 0.0
127927 Kansas 0.0 3.0 30.0 0.0 Within past year (anytime less than 12 months ... 1.0 10.0 1.0 0.0 ... 1.68 97.52 34.70 0.0 0.0 1.0 1.0 NaN 0.0 0.0
362523 Utah 1.0 5.0 0.0 0.0 Within past year (anytime less than 12 months ... 1.0 7.0 1.0 0.0 ... 1.83 113.85 34.04 0.0 0.0 0.0 0.0 1.0 0.0 1.0
183687 Michigan 1.0 3.0 0.0 7.0 Within past year (anytime less than 12 months ... 1.0 8.0 0.0 0.0 ... 1.78 83.91 26.54 1.0 0.0 1.0 1.0 1.0 0.0 0.0
191905 Michigan 0.0 4.0 0.0 0.0 Within past year (anytime less than 12 months ... 1.0 7.0 0.0 0.0 ... 1.57 68.04 27.44 1.0 1.0 0.0 1.0 1.0 0.0 0.0

5 rows × 40 columns

scale_float_columns(test)
scale_float_columns(train)
scale_float_columns(valid)
test.head()
State Male GeneralHealth PhysicalHealthDays MentalHealthDays LastCheckupTime PhysicalActivities SleepHours RemovedTeeth HadHeartAttack ... HeightInMeters WeightInKilograms BMI AlcoholDrinkers HIVTesting FluVaxLast12 PneumoVaxEver TetanusLast10Tdap HighRiskLastYear CovidPos
339824 South Dakota 0.0 0.50 0.1 0.700000 Within past year (anytime less than 12 months ... 1.0 0.304348 0.666667 0.0 ... 0.482517 0.110156 0.097513 0.0 1.0 1.0 1.0 0.0 0.0 0.0
127927 Kansas 0.0 0.50 1.0 0.000000 Within past year (anytime less than 12 months ... 1.0 0.391304 0.333333 0.0 ... 0.538462 0.279650 0.264860 0.0 0.0 1.0 1.0 NaN 0.0 0.0
362523 Utah 1.0 1.00 0.0 0.000000 Within past year (anytime less than 12 months ... 1.0 0.260870 0.333333 0.0 ... 0.643357 0.340670 0.257153 0.0 0.0 0.0 0.0 0.0 0.0 1.0
183687 Michigan 1.0 0.50 0.0 0.233333 Within past year (anytime less than 12 months ... 1.0 0.304348 0.000000 0.0 ... 0.608392 0.228795 0.169567 1.0 0.0 1.0 1.0 0.0 0.0 0.0
191905 Michigan 0.0 0.75 0.0 0.000000 Within past year (anytime less than 12 months ... 1.0 0.260870 0.000000 0.0 ... 0.461538 0.169494 0.180077 1.0 1.0 0.0 1.0 0.0 0.0 0.0

5 rows × 40 columns

5. Czyszczenie brakujących pól

Nie możemy użyć .dropna() gdyż większość wierszy ma brakujące wartości:

print(df.shape[0])
print(df.shape[0] - df.dropna().shape[0])
445132
199110

test.head()
State Male GeneralHealth PhysicalHealthDays MentalHealthDays LastCheckupTime PhysicalActivities SleepHours RemovedTeeth HadHeartAttack ... HeightInMeters WeightInKilograms BMI AlcoholDrinkers HIVTesting FluVaxLast12 PneumoVaxEver TetanusLast10Tdap HighRiskLastYear CovidPos
339824 South Dakota 0.0 0.50 0.1 0.700000 Within past year (anytime less than 12 months ... 1.0 0.304348 0.666667 0.0 ... 0.482517 0.110156 0.097513 0.0 1.0 1.0 1.0 0.0 0.0 0.0
127927 Kansas 0.0 0.50 1.0 0.000000 Within past year (anytime less than 12 months ... 1.0 0.391304 0.333333 0.0 ... 0.538462 0.279650 0.264860 0.0 0.0 1.0 1.0 NaN 0.0 0.0
362523 Utah 1.0 1.00 0.0 0.000000 Within past year (anytime less than 12 months ... 1.0 0.260870 0.333333 0.0 ... 0.643357 0.340670 0.257153 0.0 0.0 0.0 0.0 0.0 0.0 1.0
183687 Michigan 1.0 0.50 0.0 0.233333 Within past year (anytime less than 12 months ... 1.0 0.304348 0.000000 0.0 ... 0.608392 0.228795 0.169567 1.0 0.0 1.0 1.0 0.0 0.0 0.0
191905 Michigan 0.0 0.75 0.0 0.000000 Within past year (anytime less than 12 months ... 1.0 0.260870 0.000000 0.0 ... 0.461538 0.169494 0.180077 1.0 1.0 0.0 1.0 0.0 0.0 0.0

5 rows × 40 columns

Uzupełniam brakujące wartości medianą:

#test.dropna(inplace=True)
#train.dropna(inplace=True)
#valid.dropna(inplace=True)
test.fillna(test.median(),inplace=True)
train.fillna(train.median(),inplace=True)
valid.fillna(valid.median(),inplace=True)
C:\Users\Adrian\AppData\Local\Temp\ipykernel_18732\896322512.py:4: FutureWarning: Dropping of nuisance columns in DataFrame reductions (with 'numeric_only=None') is deprecated; in a future version this will raise TypeError.  Select only valid columns before calling the reduction.
  test.fillna(test.median(),inplace=True)
C:\Users\Adrian\AppData\Local\Temp\ipykernel_18732\896322512.py:5: FutureWarning: Dropping of nuisance columns in DataFrame reductions (with 'numeric_only=None') is deprecated; in a future version this will raise TypeError.  Select only valid columns before calling the reduction.
  train.fillna(train.median(),inplace=True)
C:\Users\Adrian\AppData\Local\Temp\ipykernel_18732\896322512.py:6: FutureWarning: Dropping of nuisance columns in DataFrame reductions (with 'numeric_only=None') is deprecated; in a future version this will raise TypeError.  Select only valid columns before calling the reduction.
  valid.fillna(valid.median(),inplace=True)

test.head()
State Male GeneralHealth PhysicalHealthDays MentalHealthDays LastCheckupTime PhysicalActivities SleepHours RemovedTeeth HadHeartAttack ... HeightInMeters WeightInKilograms BMI AlcoholDrinkers HIVTesting FluVaxLast12 PneumoVaxEver TetanusLast10Tdap HighRiskLastYear CovidPos
339824 South Dakota 0.0 0.50 0.1 0.700000 Within past year (anytime less than 12 months ... 1.0 0.304348 0.666667 0.0 ... 0.482517 0.110156 0.097513 0.0 1.0 1.0 1.0 0.0 0.0 0.0
127927 Kansas 0.0 0.50 1.0 0.000000 Within past year (anytime less than 12 months ... 1.0 0.391304 0.333333 0.0 ... 0.538462 0.279650 0.264860 0.0 0.0 1.0 1.0 0.0 0.0 0.0
362523 Utah 1.0 1.00 0.0 0.000000 Within past year (anytime less than 12 months ... 1.0 0.260870 0.333333 0.0 ... 0.643357 0.340670 0.257153 0.0 0.0 0.0 0.0 0.0 0.0 1.0
183687 Michigan 1.0 0.50 0.0 0.233333 Within past year (anytime less than 12 months ... 1.0 0.304348 0.000000 0.0 ... 0.608392 0.228795 0.169567 1.0 0.0 1.0 1.0 0.0 0.0 0.0
191905 Michigan 0.0 0.75 0.0 0.000000 Within past year (anytime less than 12 months ... 1.0 0.260870 0.000000 0.0 ... 0.461538 0.169494 0.180077 1.0 1.0 0.0 1.0 0.0 0.0 0.0

5 rows × 40 columns

Kolumny kategoryczne wypełniłem w czasie normalizacji wartościami "Unknown" ponieważ fillna-->median nie działa dla tego typu danych (https://stackoverflow.com/questions/49127897/python-pandas-fillna-median-not-working)

test["HighRiskLastYear"].value_counts()
0.0    42796
1.0     1717
Name: HighRiskLastYear, dtype: int64
test["HighRiskLastYear"].isna().sum()
0

Brak wartości non-null:

test.info()
<class 'pandas.core.frame.DataFrame'>
Int64Index: 44513 entries, 339824 to 52161
Data columns (total 40 columns):
 #   Column                     Non-Null Count  Dtype   
---  ------                     --------------  -----   
 0   State                      44513 non-null  category
 1   Male                       44513 non-null  float64 
 2   GeneralHealth              44513 non-null  float64 
 3   PhysicalHealthDays         44513 non-null  float64 
 4   MentalHealthDays           44513 non-null  float64 
 5   LastCheckupTime            44513 non-null  category
 6   PhysicalActivities         44513 non-null  float64 
 7   SleepHours                 44513 non-null  float64 
 8   RemovedTeeth               44513 non-null  float64 
 9   HadHeartAttack             44513 non-null  float64 
 10  HadAngina                  44513 non-null  float64 
 11  HadStroke                  44513 non-null  float64 
 12  HadAsthma                  44513 non-null  float64 
 13  HadSkinCancer              44513 non-null  float64 
 14  HadCOPD                    44513 non-null  float64 
 15  HadDepressiveDisorder      44513 non-null  float64 
 16  HadKidneyDisease           44513 non-null  float64 
 17  HadArthritis               44513 non-null  float64 
 18  HadDiabetes                44513 non-null  float64 
 19  DeafOrHardOfHearing        44513 non-null  float64 
 20  BlindOrVisionDifficulty    44513 non-null  float64 
 21  DifficultyConcentrating    44513 non-null  float64 
 22  DifficultyWalking          44513 non-null  float64 
 23  DifficultyDressingBathing  44513 non-null  float64 
 24  DifficultyErrands          44513 non-null  float64 
 25  SmokerStatus               44513 non-null  float64 
 26  ECigaretteUsage            44513 non-null  float64 
 27  ChestScan                  44513 non-null  float64 
 28  RaceEthnicityCategory      44513 non-null  category
 29  AgeCategory                44513 non-null  category
 30  HeightInMeters             44513 non-null  float64 
 31  WeightInKilograms          44513 non-null  float64 
 32  BMI                        44513 non-null  float64 
 33  AlcoholDrinkers            44513 non-null  float64 
 34  HIVTesting                 44513 non-null  float64 
 35  FluVaxLast12               44513 non-null  float64 
 36  PneumoVaxEver              44513 non-null  float64 
 37  TetanusLast10Tdap          44513 non-null  float64 
 38  HighRiskLastYear           44513 non-null  float64 
 39  CovidPos                   44513 non-null  float64 
dtypes: category(4), float64(36)
memory usage: 12.7 MB

train.info()
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 676777 entries, 0 to 676776
Data columns (total 40 columns):
 #   Column                     Non-Null Count   Dtype   
---  ------                     --------------   -----   
 0   State                      676777 non-null  category
 1   Male                       676777 non-null  float64 
 2   GeneralHealth              676777 non-null  float64 
 3   PhysicalHealthDays         676777 non-null  float64 
 4   MentalHealthDays           676777 non-null  float64 
 5   LastCheckupTime            676777 non-null  category
 6   PhysicalActivities         676777 non-null  float64 
 7   SleepHours                 676777 non-null  float64 
 8   RemovedTeeth               676777 non-null  float64 
 9   HadHeartAttack             676777 non-null  float64 
 10  HadAngina                  676777 non-null  float64 
 11  HadStroke                  676777 non-null  float64 
 12  HadAsthma                  676777 non-null  float64 
 13  HadSkinCancer              676777 non-null  float64 
 14  HadCOPD                    676777 non-null  float64 
 15  HadDepressiveDisorder      676777 non-null  float64 
 16  HadKidneyDisease           676777 non-null  float64 
 17  HadArthritis               676777 non-null  float64 
 18  HadDiabetes                676777 non-null  float64 
 19  DeafOrHardOfHearing        676777 non-null  float64 
 20  BlindOrVisionDifficulty    676777 non-null  float64 
 21  DifficultyConcentrating    676777 non-null  float64 
 22  DifficultyWalking          676777 non-null  float64 
 23  DifficultyDressingBathing  676777 non-null  float64 
 24  DifficultyErrands          676777 non-null  float64 
 25  SmokerStatus               676777 non-null  float64 
 26  ECigaretteUsage            676777 non-null  float64 
 27  ChestScan                  676777 non-null  float64 
 28  RaceEthnicityCategory      676777 non-null  category
 29  AgeCategory                676777 non-null  category
 30  HeightInMeters             676777 non-null  float64 
 31  WeightInKilograms          676777 non-null  float64 
 32  BMI                        676777 non-null  float64 
 33  AlcoholDrinkers            676777 non-null  float64 
 34  HIVTesting                 676777 non-null  float64 
 35  FluVaxLast12               676777 non-null  float64 
 36  PneumoVaxEver              676777 non-null  float64 
 37  TetanusLast10Tdap          676777 non-null  float64 
 38  HighRiskLastYear           676777 non-null  float64 
 39  CovidPos                   676777 non-null  float64 
dtypes: category(4), float64(36)
memory usage: 188.5 MB

valid.info()
<class 'pandas.core.frame.DataFrame'>
Int64Index: 44514 entries, 66965 to 224311
Data columns (total 40 columns):
 #   Column                     Non-Null Count  Dtype   
---  ------                     --------------  -----   
 0   State                      44514 non-null  category
 1   Male                       44514 non-null  float64 
 2   GeneralHealth              44514 non-null  float64 
 3   PhysicalHealthDays         44514 non-null  float64 
 4   MentalHealthDays           44514 non-null  float64 
 5   LastCheckupTime            44514 non-null  category
 6   PhysicalActivities         44514 non-null  float64 
 7   SleepHours                 44514 non-null  float64 
 8   RemovedTeeth               44514 non-null  float64 
 9   HadHeartAttack             44514 non-null  float64 
 10  HadAngina                  44514 non-null  float64 
 11  HadStroke                  44514 non-null  float64 
 12  HadAsthma                  44514 non-null  float64 
 13  HadSkinCancer              44514 non-null  float64 
 14  HadCOPD                    44514 non-null  float64 
 15  HadDepressiveDisorder      44514 non-null  float64 
 16  HadKidneyDisease           44514 non-null  float64 
 17  HadArthritis               44514 non-null  float64 
 18  HadDiabetes                44514 non-null  float64 
 19  DeafOrHardOfHearing        44514 non-null  float64 
 20  BlindOrVisionDifficulty    44514 non-null  float64 
 21  DifficultyConcentrating    44514 non-null  float64 
 22  DifficultyWalking          44514 non-null  float64 
 23  DifficultyDressingBathing  44514 non-null  float64 
 24  DifficultyErrands          44514 non-null  float64 
 25  SmokerStatus               44514 non-null  float64 
 26  ECigaretteUsage            44514 non-null  float64 
 27  ChestScan                  44514 non-null  float64 
 28  RaceEthnicityCategory      44514 non-null  category
 29  AgeCategory                44514 non-null  category
 30  HeightInMeters             44514 non-null  float64 
 31  WeightInKilograms          44514 non-null  float64 
 32  BMI                        44514 non-null  float64 
 33  AlcoholDrinkers            44514 non-null  float64 
 34  HIVTesting                 44514 non-null  float64 
 35  FluVaxLast12               44514 non-null  float64 
 36  PneumoVaxEver              44514 non-null  float64 
 37  TetanusLast10Tdap          44514 non-null  float64 
 38  HighRiskLastYear           44514 non-null  float64 
 39  CovidPos                   44514 non-null  float64 
dtypes: category(4), float64(36)
memory usage: 12.7 MB