【Kaggleチュートリアル】spaceship Titanicに参加してみた

import seaborn as sns
import numpy as np
import pandas as pd
import math

import os
for dirname, _, filenames in os.walk('/kaggle/input'):
    for filename in filenames:
        print(os.path.join(dirname, filename))

dataset=pd.read_csv("/kaggle/input/spaceship-titanic/train.csv")

sns.histplot(x="Age",hue="Transported",data=dataset)

sns.countplot(x="VIP",hue="Transported",data=dataset)

sns.countplot(x="Destination",hue="Transported",data=dataset)

sns.countplot(x="HomePlanet",hue="Transported",data=dataset)

df=dataset.assign(CabinDeck="XX").assign(CabinNum=-1).assign(CabinPort="XX")

for index,row in df[::].iterrows():
    cabin = str(row["Cabin"]).split("/")
    if(len(cabin)!=3):
        continue
    for i in range(3):
        df.iat[index,14+i]=cabin[i]

sns.countplot(x="CabinDeck",hue="Transported",data=df)

df.loc[df["CabinDeck"]=="T","CabinDeck"]="A"

df["CabinNum"]=df["CabinNum"].astype(int)
sns.histplot(x="CabinNum",hue="Transported",data=df)

df.loc[df["CabinNum"]==-1,"CabinNum"]=625

sns.countplot(x="CabinPort",hue="Transported",data=df)

MedianData=dataset.iloc[:,7:12].replace(0,float("nan")).median()
for index, row in df.iterrows():
    if row["RoomService"]>0 or row["FoodCourt"]>0 or row["ShoppingMall"]>0 or row["Spa"]>0 or row["VRDeck"]>0:
        df.iloc[index,:]=row.fillna(MedianData)
    else:
        df.iloc[index,7:12]=0
        
df["TotalCost"]=df["RoomService"]+df["FoodCourt"]+df["ShoppingMall"]+df["Spa"]+df["VRDeck"]

sns.histplot(x="TotalCost",hue="Transported",data=df)

testdata=pd.read_csv("/kaggle/input/spaceship-titanic/test.csv")
groupdf=pd.DataFrame(pd.concat([df,testdata])["PassengerId"])
groupdf["GroupSize"]=0
size=0
for index,row in groupdf[::-1].iterrows():
    if size==0:
        size=int(str(row["PassengerId"]).split("_")[1])
        groupdf.iloc[index-size+1:index+1,1]=size
    size=size-1
df=df.merge(groupdf,on="PassengerId")

sns.countplot(x="GroupSize",hue="Transported",data=df)

df=df.drop("Name",axis=1)
df=df.drop("Cabin",axis=1)
df=df.drop("PassengerId",axis=1)

df=df.fillna({"Age":dataset["Age"].median()})
df=df.fillna(df.mode().iloc[0])

df=df.astype({"CabinNum":int})
finaldf=pd.get_dummies(df, drop_first=False)
finaldf=finaldf.drop(["CabinPort_XX","CabinDeck_XX"],axis=1)

from sklearn.model_selection import train_test_split, GridSearchCV
from sklearn.svm import SVC
from sklearn.metrics import accuracy_score
from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import cross_val_score

X=df.drop(columns="Transported")
Y=df["Transported"]
X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size=0.2, random_state=1)
scaler = StandardScaler()
X_train = scaler.fit_transform(X_train)
X_test = scaler.transform(X_test)

# SVMモデルの設定
svm = SVC(kernel='rbf', random_state=1)
# k分割交差検証
k = 5 # 分割数
scores = cross_val_score(svm, X_train, y_train, cv=k, n_jobs=-1)
# 平均精度の表示
print(f'Average accuracy across {k} folds: {scores.mean()}')

# モデルのトレーニング
svm.fit(X_train, y_train)

# テストセットでの評価
y_pred = svm.predict(X_test)
accuracy = accuracy_score(y_test, y_pred)
print(f'Test set accuracy: {accuracy}')

testdata=testdata.assign(CabinDeck="XX").assign(CabinNum=625).assign(CabinPort="XX")

for index,row in testdata[::].iterrows():
    cabin = str(row["Cabin"]).split("/")
    if(len(cabin)!=3):
        continue
    for i in range(3):
        testdata.iat[index,13+i]=cabin[i]
testdata.loc[testdata["CabinDeck"]=="T","CabinDeck"]="A"
        
for index, row in testdata.iterrows():
    if row["RoomService"]>0 or row["FoodCourt"]>0 or row["ShoppingMall"]>0 or row["Spa"]>0 or row["VRDeck"]>0:
        testdata.iloc[index,:]=row.fillna(MedianData)
    else:
        testdata.iloc[index,7:12]=0
        
testdata["TotalCost"]=testdata["RoomService"]+testdata["FoodCourt"]+testdata["ShoppingMall"]+testdata["Spa"]+testdata["VRDeck"]

testdata=testdata.merge(groupdf,on="PassengerId")

testdata=testdata.drop("Name",axis=1)
testdata=testdata.drop("Cabin",axis=1)
testdata=testdata.drop("PassengerId",axis=1)

testdata=testdata.fillna({"Age":dataset["Age"].median()})
testdata=testdata.fillna(testdata.mode().iloc[0])

testdata=testdata.astype({"CabinNum":int})
testdata=pd.get_dummies(testdata, drop_first=False)
testdata=testdata.drop("CabinDeck_XX",axis=1)
testdata=testdata.drop("CabinPort_XX",axis=1)

testdata1=scaler.fit_transform(testdata)
pred=rfc.predict(pd.DataFrame(testdata1))

predd=pd.DataFrame({"Transported":pred})
tmp=pd.read_csv("/kaggle/input/spaceship-titanic/test.csv")
predd.assign(PassengerId=0)
predd["PassengerId"]=tmp["PassengerId"]
predd=predd.reindex(columns=['PassengerId', 'Transported'])
predd.to_csv("submission.csv", index = False)