-
Notifications
You must be signed in to change notification settings - Fork 0
/
spaceship_titanic.py
166 lines (127 loc) · 5.08 KB
/
spaceship_titanic.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
from sklearn.preprocessing import StandardScaler, OneHotEncoder, OrdinalEncoder
from sklearn.pipeline import Pipeline
from sklearn.impute import SimpleImputer
from sklearn.compose import ColumnTransformer
from sklearn.ensemble import RandomForestClassifier
from sklearn.linear_model import LinearRegression, LogisticRegression
from sklearn.metrics import mean_squared_error, mean_absolute_error, r2_score
from lazypredict.Supervised import LazyRegressor, LazyClassifier
from sklearn.model_selection import GridSearchCV
from sklearn.model_selection import train_test_split, RandomizedSearchCV
from sklearn.metrics import classification_report
"""Import Dataset"""
test = pd.read_csv("test.csv")
train = pd.read_csv("train.csv")
# print(test.info())
print(train.info())
missing_train = train.isnull().sum()
missing_test = test.isnull().sum()
print("Percentage Missing train data\n", train.isnull().mean() * 100, )
print("=" * 20)
print("Percentage Missing of test data\n", test.isnull().mean() * 100)
# print(missing_train)
# print(missing_test)
# Vẽ heatmap cho dữ liệu bị thiếu|
# plt.figure(figsize=(12, 8))
# sns.heatmap(train.isnull(), cbar=False, cmap='viridis', yticklabels=False)
# plt.title('Visualization of Missing Data')
# plt.show()
"""# Checking balanced data or not"""
class_counts = train['Transported'].value_counts()
print(class_counts)
plt.figure(figsize=(10, 6))
sns.countplot(data=train, x='Transported')
plt.title('Class Distribution')
plt.show()
sns.heatmap(train.corr(), annot=True)
plt.show()
"""Biểu đồ corr yếu nên khả năng cao xài model tuyến tính sẽ cho kết quả không được tốt"""
target = 'Transported'
x = train.drop([target, "PassengerId", "Name", "Cabin"], axis=1)
y = train[target]
X_train, X_test, y_train, y_test = train_test_split(x, y, test_size=0.2, random_state=42)
print(y_train.shape)
print(X_train.shape)
"""#Pipeline:
- Preprocessing
- Imputer
- Transformer
"""
numerical_transformer = Pipeline(steps=[
('imputer', SimpleImputer(strategy="median")),
('scaler', StandardScaler())
])
print((X_train["HomePlanet"].unique()))
print((X_train["Destination"].unique()))
print((X_train["VIP"].unique()))
# HomePlanet_vals = ['Earth','Europa','Mars']
# CryoSleep_vals = X_train["CryoSleep"].unique()
# Destination_vals = X_train["Destination"].unique()
# VIP_vals = X_train["VIP"].unique()
HomePlanet_vals = ['Earth', 'Europa', 'Mars']
CryoSleep_vals = X_train["CryoSleep"].dropna().unique().tolist()
Destination_vals = X_train["Destination"].dropna().unique().tolist()
VIP_vals = X_train["VIP"].dropna().unique().tolist()
print(HomePlanet_vals)
print(CryoSleep_vals)
print(Destination_vals)
print(VIP_vals)
ordinal_transformer = Pipeline(steps=[
('imputer', SimpleImputer(strategy="most_frequent")), # Xử lý missing value
# ('scaler', OrdinalEncoder(categories=[HomePlanet_vals, X_train["CryoSleep"].unique(), X_train["Destination"].unique(), X_train["VIP"].unique()]))
('scaler', OrdinalEncoder(categories=[CryoSleep_vals, VIP_vals])) # Biến đổi dữ liệu
])
norminal_transformer = Pipeline(steps=[
('imputer', SimpleImputer(strategy="most_frequent")), # Xử lý missing value
('scaler', OneHotEncoder()) # Biến đổi dữ liệu
])
"""# Preprocessor"""
preprocessor = ColumnTransformer(transformers=[
("num_features", numerical_transformer, ["Age", 'RoomService', 'FoodCourt', 'ShoppingMall', 'Spa', 'VRDeck']),
('ord_features', ordinal_transformer, ["CryoSleep", 'VIP']),
('nom_features', norminal_transformer, ['HomePlanet', 'Destination'])
])
"""#pipeline"""
cls_lazy = Pipeline(steps=[
('preprocessor', preprocessor),
# ('model', RandomForestRegressor())
])
"""# LazyPredict"""
X_train = cls_lazy.fit_transform((X_train))
X_test = cls_lazy.transform(X_test)
clf = LazyClassifier(verbose=0, ignore_warnings=True, custom_metric=None)
models, predictions = clf.fit(X_train, X_test, y_train, y_test)
print(models)
"""# GridSearch"""
# Logistic Regression
# LGBMClassifier
# RandomForestClassifier
cls = Pipeline(steps=[
('preprocessor', preprocessor),
('model', LogisticRegression())
# ('model', RandomForestClassifier())
])
# Logistic regression parameters
lr_params = {
"model__penalty": ['l1', 'l2', 'elasticnet', None],
"model__solver": ['lbfgs', 'liblinear', 'newton-cg', 'newton-cholesky', 'sag', 'saga']
}
# RandomForestClassifier parameters
rf_params = {
"model__criterion": ["gini", "entropy", "log_loss"],
"model__max_features": ["sqrt", "log2", None],
}
grid_cls = GridSearchCV(cls, param_grid=lr_params, cv=6, verbose=2)
grid_cls.fit(X_train, y_train)
y_predict = grid_cls.predict(X_test)
print(classification_report(y_test, y_predict))
submission_pred = grid_cls.predict(test)
print(submission_pred.shape)
test_ids = test['PassengerId']
print(test_ids.shape)
submission_pred = submission_pred.astype(bool)
df = pd.DataFrame({'PassengerId': test_ids.values, 'Transported': submission_pred})
df.to_csv("/content/drive/MyDrive/Data Science/Kaggle Competition/Spaceship_titanic/Submission.csv", index=False)