-
Notifications
You must be signed in to change notification settings - Fork 0
/
majority_vote.py
72 lines (61 loc) · 2.22 KB
/
majority_vote.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
from utils import *
def main():
""" Returns the majority vote.
This code serves as an example how to load the data and
submit your result to kaggle.
:return: None
"""
# Load the train data.
train_data = load_train_csv("data")
correct_question_map = {}
total_question_map = {}
# Count how many questions were correct.
for i, q in enumerate(train_data["question_id"]):
if q in correct_question_map:
if train_data["is_correct"][i] == 1:
correct_question_map[q] += 1
total_question_map[q] += 1
else:
if train_data["is_correct"][i] == 1:
correct_question_map[q] = 1
total_question_map[q] = 1
# Load the validation data.
valid_data = load_valid_csv("data")
predictions = []
for i, q in enumerate(valid_data["question_id"]):
ratio = correct_question_map[q] / float(total_question_map[q])
# If the question was answered correctly more than half
# of the times, predict correct.
if ratio >= 0.5:
predictions.append(1.)
else:
predictions.append(0.)
# Evaluate your model using the new prediction.
acc = evaluate(valid_data, predictions)
print("Validation Accuracy: {}".format(acc))
# Load the public test data.
test_data = load_public_test_csv("data")
predictions = []
for i, q in enumerate(test_data["question_id"]):
ratio = correct_question_map[q] / float(total_question_map[q])
if ratio >= 0.5:
predictions.append(1.)
else:
predictions.append(0.)
acc = evaluate(test_data, predictions)
print("Test Accuracy: {}".format(acc))
# Load the private test dataset.
private_test = load_private_test_csv("data")
predictions = []
for i, q in enumerate(private_test["question_id"]):
ratio = correct_question_map[q] / float(total_question_map[q])
if ratio >= 0.5:
predictions.append(1.)
else:
predictions.append(0.)
private_test["is_correct"] = predictions
# Save your predictions to make it ready to submit to Kaggle.
save_private_test_csv(private_test)
return
if __name__ == "__main__":
main()