-
Notifications
You must be signed in to change notification settings - Fork 1
/
folds.py
executable file
·47 lines (44 loc) · 1.7 KB
/
folds.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
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Thu Dec 19 14:21:35 2019
@author: wu
"""
import numpy as np
class FoldManager():
def __init__(self,
datas=["folds_isophonics.npy","folds_rwc.npy","folds_uspop.npy","folds_billboard.npy"],
names=["Isophonics","RWC","USPOP","Billboard"]):
self.folds = [np.load(d) for d in datas]
self.nfolds = len(self.folds[0])
self.nsets = len(datas)
self.names = names
def getTestFold(self,fold):
return np.concatenate([f[fold] for f in self.folds])
def getDatasetIdx(self,dset):
return np.concatenate(self.folds[dset]),self.names[dset]
def getTrainSupervisedFold(self,fold,supervise_size=None):
idxes = []
idx_train = list(range(self.nfolds))
idx_train.remove(fold)
idx_supervise = idx_train if supervise_size is None else idx_train[:supervise_size]
for n in range(self.nsets):
idxes.extend([self.folds[n][i] for i in idx_supervise])
idxes = np.concatenate(idxes)
return idxes
def getTrainUnsupervisedFold(self,fold,supervise_size):
idxes = []
idx_train = list(range(self.nfolds))
idx_train.remove(fold)
idx_supervise = idx_train[supervise_size:]
for n in range(self.nsets):
idxes.extend([self.folds[n][i] for i in idx_supervise])
idxes = np.concatenate(idxes)
return idxes
def getAll(self):
return np.concatenate([np.concatenate(f) for f in self.folds])
def retrieveFold(self,idx):
for f in range(self.nfolds):
for d in range(self.nsets):
if idx in self.folds[d][f]:
return f