自动化机器学习pycaret： PyCaret Basic Auto Classification LightGBM

from IPython.display import clear_output
!pip3 install pycaret --user
clear_output()

!pip install numpy==1.20.0

import numpy as np
import pandas as pd 

import random
import os 

from pycaret.classification import *

TRAIN_PATH = "../input/titanic/train.csv"
TEST_PATH = "../input/titanic/test.csv"
SAMPLE_SUBMISSION_PATH = "../input/titanic/gender_submission.csv"
SUBMISSION_PATH = "submission.csv"

ID = "PassengerId"
TARGET = "Survived"

SEED = 2022

def seed_everything(seed: int = SEED):
    random.seed(seed)
    np.random.seed(seed)
    os.environ["PYTHONHASHSEED"] = str(seed)
    
seed_everything()

import pandas as pd
train = pd.read_csv(TRAIN_PATH)
test = pd.read_csv(TEST_PATH)
test

 

 

 

setup(
    data=train,
    target=TARGET,
    silent=True
)

 

({'lr': <pycaret.containers.models.classification.LogisticRegressionClassifierContainer at 0x7f2b6faa2950>,
  'knn': <pycaret.containers.models.classification.KNeighborsClassifierContainer at 0x7f2b6faa2ad0>,
  'nb': <pycaret.containers.models.classification.GaussianNBClassifierContainer at 0x7f2b6faa2790>,
  'dt': <pycaret.containers.models.classification.DecisionTreeClassifierContainer at 0x7f2b6faa27d0>,
  'svm': <pycaret.containers.models.classification.SGDClassifierContainer at 0x7f2b6faac990>,
  'rbfsvm': <pycaret.containers.models.classification.SVCClassifierContainer at 0x7f2b6faac6d0>,
  'gpc': <pycaret.containers.models.classification.GaussianProcessClassifierContainer at 0x7f2b6faac910>,
  'mlp': <pycaret.containers.models.classification.MLPClassifierContainer at 0x7f2b6faac510>,
  'ridge': <pycaret.containers.models.classification.RidgeClassifierContainer at 0x7f2b6fb3f750>,
  'rf': <pycaret.containers.models.classification.RandomForestClassifierContainer at 0x7f2b6faa26d0>,
  'qda': <pycaret.containers.models.classification.QuadraticDiscriminantAnalysisContainer at 0x7f2b6fb3f2d0>,
  'ada': <pycaret.containers.models.classification.AdaBoostClassifierContainer at 0x7f2b6fb3f210>,
  'gbc': <pycaret.containers.models.classification.GradientBoostingClassifierContainer at 0x7f2b6fb3ce10>,
  'lda': <pycaret.containers.models.classification.LinearDiscriminantAnalysisContainer at 0x7f2b6faac610>,
  'et': <pycaret.containers.models.classification.ExtraTreesClassifierContainer at 0x7f2b6fb3c910>,
  'xgboost': <pycaret.containers.models.classification.XGBClassifierContainer at 0x7f2b6fb3ca10>,
  'lightgbm': <pycaret.containers.models.classification.LGBMClassifierContainer at 0x7f2b6fb3c250>,
  'catboost': <pycaret.containers.models.classification.CatBoostClassifierContainer at 0x7f2b6fb3c1d0>,
  'dummy': <pycaret.containers.models.classification.DummyClassifierContainer at 0x7f2b6facd490>},
 True,
 150    0
 547    1
 125    1
 779    1
 183    1
       ..
 370    1
 317    0
 351    0
 339    0
 289    1
 Name: Survived, Length: 623, dtype: int64,
 10,
 8860,
 '88e1',
 Pipeline(memory=None, steps=[('empty_step', 'passthrough')], verbose=False),
 False,
            Age       Fare  Pclass_1  Pclass_2  Pclass_3  \
 0    22.000000   7.250000       0.0       0.0       1.0   
 1    38.000000  71.283302       1.0       0.0       0.0   
 2    26.000000   7.925000       0.0       0.0       1.0   
 3    35.000000  53.099998       1.0       0.0       0.0   
 4    35.000000   8.050000       0.0       0.0       1.0   
 ..         ...        ...       ...       ...       ...   
 886  27.000000  13.000000       0.0       1.0       0.0   
 887  19.000000  30.000000       1.0       0.0       0.0   
 888  29.466112  23.450001       0.0       0.0       1.0   
 889  26.000000  30.000000       1.0       0.0       0.0   
 890  32.000000   7.750000       0.0       0.0       1.0   
 
      Name_Aks Mrs. Sam (Leah Rosen)  Name_Albimona Mr. Nassef Cassem  \
 0                               0.0                              0.0   
 1                               0.0                              0.0   
 2                               0.0                              0.0   
 3                               0.0                              0.0   
 4                               0.0                              0.0   
 ..                              ...                              ...   
 886                             0.0                              0.0   
 887                             0.0                              0.0   
 888                             0.0                              0.0   
 889                             0.0                              0.0   
 890                             0.0                              0.0   
 
      Name_Ali Mr. Ahmed  Name_Allen Mr. William Henry  \
 0                   0.0                           0.0   
 1                   0.0                           0.0   
 2                   0.0                           0.0   
 3                   0.0                           0.0   
 4                   0.0                           1.0   
 ..                  ...                           ...   
 886                 0.0                           0.0   
 887                 0.0                           0.0   
 888                 0.0                           0.0   
 889                 0.0                           0.0   
 890                 0.0                           0.0   

model = create_model('lightgbm')

tuneModel = tune_model(model,optimize = 'AUC')

 

plot_model(tuneModel)

 

plot_model(tuneModel, plot='feature')