-
Notifications
You must be signed in to change notification settings - Fork 0
/
runXGBoost.py
101 lines (76 loc) · 4.05 KB
/
runXGBoost.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
import xgboost as xgb
from sklearn.model_selection import train_test_split
import pandas as pd
import numpy as np
from util import generateSequence
from sklearn.metrics import roc_auc_score
def predictXGBoost(model_path, test_path, test_filename, result_path, variant, outcome):
generateSequence(test_path, test_filename)
test_seq_path = test_path + "data_poiFeat.csv"
result_model = model_path + "model" + outcome + ".txt"
result_prediction = result_path + "prediction-"+ variant +'-'+ outcome + ".csv"
test = pd.read_csv(test_path + test_filename)
# Check if "significance" column exists in the dataframe
has_significance_col = "significance" in test.columns
if has_significance_col:
y_test = test[["significance"]].values
else:
y_test = None
X_test2 = pd.read_csv(test_seq_path)
if variant == "seq_anno":
X_test = test[['OGEE_prop_Essential', 'deltagb', 'deltagh', "H3k27ac_CPM_1Kb_new", "ATAC_CPM_1Kb_new", "H3K4me3_CPM_1Kb_new"]]
X_test3 = pd.concat([X_test, X_test2], axis=1)
else:
X_test3 = X_test2
print("There are ", X_test3.shape[0], "gRNAs in test data.")
dtest = xgb.DMatrix(X_test3, label=y_test)
best_model = xgb.Booster()
best_model.load_model(result_model)
test_est = best_model.predict(dtest)
if has_significance_col:
print('\n> AUC score is', roc_auc_score(y_test, test_est))
PD = pd.DataFrame(np.column_stack((test['protospacer'], y_test, test_est)), columns=['grna', 'true', 'predict'])
else:
PD = pd.DataFrame(np.column_stack((test['protospacer'], test_est)), columns=['grna', 'predict'])
PD.to_csv(result_prediction, index=False)
print('\n> Prediction data saved in result path! \n')
def trainXGBoost(model_path, train_path, train_filename, variant, params, outcome):
# Assuming you have defined the generateSequence function somewhere
generateSequence(train_path, train_filename)
train_seq_path = train_path + "data_poiFeat.csv"
result_model = model_path + "model" + outcome + ".txt"
train = pd.read_csv(train_path+train_filename)
y = train[["significance"]].values
X2 = pd.read_csv(train_seq_path)
if variant == "seq_anno":
X = train[['OGEE_prop_Essential', 'deltagb','deltagh', "H3k27ac_CPM_1Kb_new", "ATAC_CPM_1Kb_new","H3K4me3_CPM_1Kb_new"]]
X_train = pd.concat([X, X2], axis=1)
else:
X_train = X2
print("There are ", X_train.shape[0], "gRNAs and ", X_train.shape[1], "features to train.")
# Splitting your data into a training set and a validation set
X_train_split, X_val_split, y_train_split, y_val_split = train_test_split(X_train, y, test_size=0.2, random_state=42)
# Converting datasets into DMatrix format for XGBoost
dtrain = xgb.DMatrix(X_train, label=y,enable_categorical=True)
dtrain_split = xgb.DMatrix(X_train_split, label=y_train_split, enable_categorical=True)
dval = xgb.DMatrix(X_val_split, label=y_val_split, enable_categorical=True)
num_boost_round = 999
evals_result = {} # This will store the evaluation results for each round
print('\n> Splitting train data into a training set and a validation set to select the best number of rounds \n')
model = xgb.train(
params,
dtrain_split,
num_boost_round= num_boost_round, # a large number, early stopping will determine the actual best number
evals=[(dtrain_split, "Train"), (dval, "Validation")],
early_stopping_rounds=10,
evals_result=evals_result,
verbose_eval=True
)
# Best number of rounds based on validation set performance
best_rounds = model.best_iteration
print(f"Best number of rounds based on validation set: {best_rounds}")
print('\n> Train a final model on the entire dataset using the best number of rounds \n')
# Now, if you want, you can train a final model on the entire dataset using the best number of rounds
final_model = xgb.train(params, dtrain, num_boost_round=best_rounds, verbose_eval=True)
final_model.save_model(result_model)
print('\n> Model Saved! \n')