-
Notifications
You must be signed in to change notification settings - Fork 0
/
Ensemble.py
38 lines (30 loc) · 1.18 KB
/
Ensemble.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
import os
import datetime
import sys
import time
import string
import random
import pandas as pd
import numpy as np
from sklearn.metrics import log_loss
def ensemblePrediction(predictionFiles, outputName):
namesDataFrame = pd.read_csv(predictionFiles[0], header=None)
numberOfImages = len(namesDataFrame)
names = []
for index in range(0, numberOfImages):
names.append(namesDataFrame.iloc[index][0])
for fileIndex, predictionFile in enumerate(predictionFiles):
prediction = pd.read_csv(predictionFile, header=None)
prediction.drop(prediction.columns[[0]], axis = 1, inplace = True)
predictionMatrix = prediction.as_matrix().astype(float)
if fileIndex == 0:
averagePrediction = predictionMatrix
else:
averagePrediction += predictionMatrix
averagePrediction /= len(predictionFiles)
outputData = np.column_stack([names, averagePrediction])
np.savetxt(outputName, outputData, delimiter=',', fmt='%s')
if(len(sys.argv) < 2):
print('Usage: Ensemble.py a.csv b.csv c.csv output.csv')
sys.exit(1)
ensemblePrediction(sys.argv[1:len(sys.argv)-1], sys.argv[len(sys.argv) - 1])