fibe/main.py at master · Ankush-Kesri/fibe · GitHub

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
# Import necessary libraries
import pandas as pd
from sklearn import datasets
import numpy as np
from fibe_functions import fibe

#======================================
#      Regression Example
#======================================

# Load Boston dataset
boston = datasets.load_boston()

# Without any predefined fixed features
#--------------------------------------
data = pd.DataFrame(boston.data, columns=boston.feature_names)
target = pd.Series(boston.target)

final_features, actual_score, predicted_score, validationPerformance = fibe(data, target, verbose=True)
print(final_features, validationPerformance)

# With predefined fixed features
#-------------------------------
data = pd.DataFrame(boston.data, columns=boston.feature_names)
fixed = ['CRIM', 'ZN']
target = pd.Series(boston.target)

final_features, actual_score, predicted_score, validationPerformance = fibe(data, target, fixed_features=fixed, voting_strictness='both', verbose=True)
print(final_features, validationPerformance)


#======================================
#      Classification Example
#======================================

# Load UCI ML Breast Cancer Wisconsin (Diagnostic) dataset
cancer = datasets.load_breast_cancer()

# Without any predefined fixed features
#--------------------------------------
data = pd.DataFrame(cancer.data, columns=cancer.feature_names)
target = pd.Series(cancer.target)

final_features, actual_score, predicted_score, validationPerformance = fibe(data, target, task_type='classification', verbose=True)
print(final_features, actual_score, predicted_score, validationPerformance)

# With predefined fixed features
#-------------------------------
data = pd.DataFrame(cancer.data, columns=cancer.feature_names)
fixed = ['mean area', 'radius error']
target = pd.Series(cancer.target)

final_features, actual_score, predicted_score, validationPerformance = fibe(data, target, fixed_features=fixed, task_type='classification', voting_strictness='strict', verbose=True)
print(final_features, validationPerformance)

#============================================================
#      Classification Example with Imbalanced Data
#============================================================
# Load digit dataset
digit = datasets.load_digits()

# Making data imbalanced
data = pd.DataFrame(digit.data, columns=digit.feature_names)
target = digit.target

target = np.where(target < 4, 0, target) # Replacing all classes from 0 to 3 with 0
target = np.where(target > 1, 1, target) # Replacing all classes from 4 to 9 with 1
target = pd.DataFrame(target)
#print(target.value_counts())


# Without any predefined fixed features
#--------------------------------------
final_features, actual_score, predicted_score, validationPerformance = fibe(data, target, task_type='classification', balance=True, model_name='consensus', metric='F1-score', voting_strictness='strict', maxIter=1, verbose=True)
print(final_features, validationPerformance)

# With predefined fixed features
#-------------------------------
fixed = ['pixel_2_1', 'pixel_2_2']
final_features, actual_score, predicted_score, validationPerformance = fibe(data, target, fixed_features=fixed, task_type='classification', balance=True, metric='F1-score', voting_strictness='strict', maxIter=1, verbose=True)
print(final_features, validationPerformance)


# ----------- Ignore Below -------------
'''
df = pd.read_csv('/home/ch225256/iq_prediction/IQ_prediction/chd_fibe/chd_all_features_scores_standardized.csv') # Load your own features/targets


feature_df = pd.concat([df.iloc[:, 0:18], df.iloc[:, 33:]], axis=1)
feature_df = feature_df.drop(["ID", "site"], axis=1)
#feature_df = feature_df.drop(["ID", "site", "ageatMRI", "sex"], axis=1)
score_df = df[["FullScaleIQ"]]
#fixed=['ageatMRI', 'sex']
#fixed = df[["ageatMRI", "sex"]]
#fixed.drop(10, inplace=True)
#fixed = fixed.to_numpy()


final_features, actual_score, predicted_score, validationPerformance = fibe(feature_df, score_df, voting_strictness='strict', verbose=False)
print(final_features, actual_score, predicted_score, validationPerformance)
'''

'''# Fit data to a model with the selected features and predict
from sklearn.ensemble import RandomForestRegressor
from sklearn.model_selection import KFold

kf5 = KFold(n_splits = 5, shuffle = False)
actual_score = []
predicted_score = []

for inner_fold in kf5.split(feature_df):
    training = feature_df.iloc[inner_fold[0]]
    validation = feature_df.iloc[inner_fold[1]]

    tscore = score_df.iloc[inner_fold[0]]
    vscore = score_df.iloc[inner_fold[1]]

    model = RandomForestRegressor(n_estimators=100, random_state=42, max_depth=5)
    model.fit(training, tscore.values.ravel())
    y_pred = model.predict(validation)

    # Calculate the mean absolute error for the validation set
    actual_score.append(vscore.values.ravel().tolist())
    predicted_score.append(y_pred.tolist())

actual = [item for sublist in actual_score for item in sublist]
predicted = [item for sublist in predicted_score for item in sublist]
print(actual, predicted)'''