kvnkho

from fugue import transform
import pandas as pd

schema = """Model:str, Accuracy:float, AUC:float, Recall:float, Prec:float, 
F1:float, Kappa:float, MCC:float, TT_Sec:float"""

def wrapper(df: pd.DataFrame) -> pd.DataFrame:
    clf = setup(data = df, 
                target = 'Survived', 
                session_id=123, 

kvnkho

import fugue_spark

schema = """Model:str, Accuracy:float, AUC:float, Recall:float, Prec:float, 
F1:float, Kappa:float, MCC:float, TT_Sec:float, Sex:str"""

def wrapper(df: pd.DataFrame) -> pd.DataFrame:
    clf = setup(data = df, 
                target = 'Survived', 
                session_id=123, 
                silent = True, 

kvnkho

df = df.drop(["Name", "PassengerId", "Ticket", "Cabin"], axis = 1)
df["Sex"] = pd.factorize(df["Sex"])[0]
dummy = pd.get_dummies(df['Embarked'], prefix='Cabin')
df = pd.concat([df.drop("Embarked", axis=1), dummy], axis = 1)

kvnkho

from sklearn.model_selection import train_test_split

y = df["Survived"]
X = df.drop("Survived", axis = 1)
X_train, X_test, y_train, y_test = train_test_split(X, y, 
                                                    test_size=0.2, 
                                                    random_state=42)

# fill age variable
fill_age = X_train["Age"].mean()

kvnkho

from typing import Any
from sklearn.linear_model import LogisticRegression

def train_model(model: Any, X_train, X_test, y_train, y_test):
    clf = model.fit(X_train, y_train)
    y_pred = clf.predict(X_test)
    acc = accuracy_score(y_test, y_pred)
    return {"model": model.__class__.__name__, "params": model.get_params(), "accuracy": acc}

train_model(LogisticRegression(), X_train, X_test, y_train, y_test)

kvnkho

from sklearn.neighbors import KNeighborsClassifier
train_model(KNeighborsClassifier(), X_train, X_test, y_train, y_test)

kvnkho

from prefect import task

@task(nout=4)
def create_data():
    df = get_data("titanic")
    df = df.drop(["Name", "PassengerId", "Ticket", "Cabin"], axis = 1)
    df["Sex"] = pd.factorize(df["Sex"])[0]
    dummy = pd.get_dummies(df['Embarked'], prefix='Cabin')
    df = pd.concat([df.drop("Embarked", axis=1), dummy], axis = 1)
    y = df["Survived"]

kvnkho

from sklearn.linear_model import LogisticRegression
from sklearn.neighbors import KNeighborsClassifier
from sklearn.tree import DecisionTreeClassifier
from sklearn.svm import SVC
from sklearn.ensemble import RandomForestClassifier

@task
def get_models():
    return [LogisticRegression(random_state=42),
            KNeighborsClassifier(), DecisionTreeClassifier(), SVC(), 

kvnkho

@task
def train_model(model: Any, X_train, X_test, y_train, y_test):
    clf = model.fit(X_train, y_train)
    y_pred = clf.predict(X_test)
    acc = accuracy_score(y_test, y_pred)
    return {"model": model.__class__.__name__, 
            "params": model.get_params(), 
            "accuracy": acc}

kvnkho

import prefect

@task
def get_results(results):
    res = pd.DataFrame(results)
    prefect.context.logger.info(res)
    return res