Mercurial > pylearn

annotate _test_dataset.py @ 99:a8da709eb6a9

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
in ArrayDataSet.__init__ if a columns is an index, we change it to be a list that containt only this index. This way, we remove the special case where the columns is an index for all subsequent call. This was possing trouble with numpy.vstack() called by MinibatchWrapAroundIterator.next
author Frederic Bastien <bastienf@iro.umontreal.ca>
date Tue, 06 May 2008 13:57:36 -0400
parents 46c5c90019c2
children 5b3afda2f1ad
rev   line source
4
f7dcfb5f9d5b Added test for dataset.
bengioy@bengiomac.local
parents:
diff changeset
1 from dataset import *
f7dcfb5f9d5b Added test for dataset.
bengioy@bengiomac.local
parents:
diff changeset
2 from math import *
f7dcfb5f9d5b Added test for dataset.
bengioy@bengiomac.local
parents:
diff changeset
3 import unittest
f7dcfb5f9d5b Added test for dataset.
bengioy@bengiomac.local
parents:
diff changeset
4
f7dcfb5f9d5b Added test for dataset.
bengioy@bengiomac.local
parents:
diff changeset
5 def _sum_all(a):
f7dcfb5f9d5b Added test for dataset.
bengioy@bengiomac.local
parents:
diff changeset
6 s=a
f7dcfb5f9d5b Added test for dataset.
bengioy@bengiomac.local
parents:
diff changeset
7 while isinstance(s,numpy.ndarray):
f7dcfb5f9d5b Added test for dataset.
bengioy@bengiomac.local
parents:
diff changeset
8 s=sum(s)
f7dcfb5f9d5b Added test for dataset.
bengioy@bengiomac.local
parents:
diff changeset
9 return s
f7dcfb5f9d5b Added test for dataset.
bengioy@bengiomac.local
parents:
diff changeset
10
f7dcfb5f9d5b Added test for dataset.
bengioy@bengiomac.local
parents:
diff changeset
11 class T_arraydataset(unittest.TestCase):
f7dcfb5f9d5b Added test for dataset.
bengioy@bengiomac.local
parents:
diff changeset
12 def setUp(self):
f7dcfb5f9d5b Added test for dataset.
bengioy@bengiomac.local
parents:
diff changeset
13 numpy.random.seed(123456)
f7dcfb5f9d5b Added test for dataset.
bengioy@bengiomac.local
parents:
diff changeset
14
17
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
15
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
16 def test_ctor_len(self):
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
17 n = numpy.random.rand(8,3)
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
18 a=ArrayDataSet(n)
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
19 self.failUnless(a.data is n)
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
20 self.failUnless(a.fields is None)
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
21
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
22 self.failUnless(len(a) == n.shape[0])
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
23 self.failUnless(a[0].shape == (n.shape[1],))
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
24
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
25 def test_iter(self):
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
26 arr = numpy.random.rand(8,3)
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
27 a=ArrayDataSet(data=arr,fields={"x":slice(2),"y":slice(1,3)})
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
28 for i, example in enumerate(a):
21
fdf0abc490f7 Adapted _test_dataset.py to changes in LookupList
bengioy@bengiomac.local
parents: 19
diff changeset
29 self.failUnless(numpy.all( example['x'] == arr[i,:2]))
fdf0abc490f7 Adapted _test_dataset.py to changes in LookupList
bengioy@bengiomac.local
parents: 19
diff changeset
30 self.failUnless(numpy.all( example['y'] == arr[i,1:3]))
17
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
31
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
32 def test_zip(self):
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
33 arr = numpy.random.rand(8,3)
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
34 a=ArrayDataSet(data=arr,fields={"x":slice(2),"y":slice(1,3)})
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
35 for i, x in enumerate(a.zip("x")):
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
36 self.failUnless(numpy.all( x == arr[i,:2]))
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
37
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
38 def test_minibatch_basic(self):
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
39 arr = numpy.random.rand(10,4)
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
40 a=ArrayDataSet(data=arr,fields={"x":slice(2),"y":slice(1,4)})
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
41 for i, mb in enumerate(a.minibatches(minibatch_size=2)): #all fields
21
fdf0abc490f7 Adapted _test_dataset.py to changes in LookupList
bengioy@bengiomac.local
parents: 19
diff changeset
42 self.failUnless(numpy.all( mb['x'] == arr[i*2:i*2+2,0:2]))
fdf0abc490f7 Adapted _test_dataset.py to changes in LookupList
bengioy@bengiomac.local
parents: 19
diff changeset
43 self.failUnless(numpy.all( mb['y'] == arr[i*2:i*2+2,1:4]))
7
6f8f338686db Moved iterating counter into a FiniteDataSetIterator to allow embedded iterations and multiple threads iterating at the same time on a dataset.
bengioy@bengiomac.local
parents: 6
diff changeset
44
17
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
45 def test_getattr(self):
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
46 arr = numpy.random.rand(10,4)
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
47 a=ArrayDataSet(data=arr,fields={"x":slice(2),"y":slice(1,4)})
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
48 a_y = a.y
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
49 self.failUnless(numpy.all( a_y == arr[:,1:4]))
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
50
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
51 def test_minibatch_wraparound_even(self):
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
52 arr = numpy.random.rand(10,4)
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
53 arr2 = ArrayDataSet.Iterator.matcat(arr,arr)
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
54
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
55 a=ArrayDataSet(data=arr,fields={"x":slice(2),"y":slice(1,4)})
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
56
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
57 #print arr
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
58 for i, x in enumerate(a.minibatches(["x"], minibatch_size=2, n_batches=8)):
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
59 #print 'x' , x
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
60 self.failUnless(numpy.all( x == arr2[i*2:i*2+2,0:2]))
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
61
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
62 def test_minibatch_wraparound_odd(self):
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
63 arr = numpy.random.rand(10,4)
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
64 arr2 = ArrayDataSet.Iterator.matcat(arr,arr)
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
65
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
66 a=ArrayDataSet(data=arr,fields={"x":slice(2),"y":slice(1,4)})
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
67
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
68 for i, x in enumerate(a.minibatches(["x"], minibatch_size=3, n_batches=6)):
759d17112b23 more comments, looping ArrayDataSet iterator, bugfixes to lookup_list, more tests
bergstrj@iro.umontreal.ca
parents: 11
diff changeset
69 self.failUnless(numpy.all( x == arr2[i*3:i*3+3,0:2]))
19
57f4015e2e09 Iterators extend LookupList
bergstrj@iro.umontreal.ca
parents: 17
diff changeset
70
26
672fe4b23032 Fixed dataset errors so that _test_dataset.py works again.
bengioy@grenat.iro.umontreal.ca
parents: 22
diff changeset
71
672fe4b23032 Fixed dataset errors so that _test_dataset.py works again.
bengioy@grenat.iro.umontreal.ca
parents: 22
diff changeset
72 class T_renamingdataset(unittest.TestCase):
672fe4b23032 Fixed dataset errors so that _test_dataset.py works again.
bengioy@grenat.iro.umontreal.ca
parents: 22
diff changeset
73 def setUp(self):
672fe4b23032 Fixed dataset errors so that _test_dataset.py works again.
bengioy@grenat.iro.umontreal.ca
parents: 22
diff changeset
74 numpy.random.seed(123456)
672fe4b23032 Fixed dataset errors so that _test_dataset.py works again.
bengioy@grenat.iro.umontreal.ca
parents: 22
diff changeset
75
672fe4b23032 Fixed dataset errors so that _test_dataset.py works again.
bengioy@grenat.iro.umontreal.ca
parents: 22
diff changeset
76
672fe4b23032 Fixed dataset errors so that _test_dataset.py works again.
bengioy@grenat.iro.umontreal.ca
parents: 22
diff changeset
77 def test_hasfield(self):
672fe4b23032 Fixed dataset errors so that _test_dataset.py works again.
bengioy@grenat.iro.umontreal.ca
parents: 22
diff changeset
78 n = numpy.random.rand(3,8)
672fe4b23032 Fixed dataset errors so that _test_dataset.py works again.
bengioy@grenat.iro.umontreal.ca
parents: 22
diff changeset
79 a=ArrayDataSet(data=n,fields={"x":slice(2),"y":slice(1,4),"z":slice(4,6)})
672fe4b23032 Fixed dataset errors so that _test_dataset.py works again.
bengioy@grenat.iro.umontreal.ca
parents: 22
diff changeset
80 b=a.rename({'xx':'x','zz':'z'})
672fe4b23032 Fixed dataset errors so that _test_dataset.py works again.
bengioy@grenat.iro.umontreal.ca
parents: 22
diff changeset
81 self.failUnless(b.hasFields('xx','zz') and not b.hasFields('x') and not b.hasFields('y'))
672fe4b23032 Fixed dataset errors so that _test_dataset.py works again.
bengioy@grenat.iro.umontreal.ca
parents: 22
diff changeset
82
29
46c5c90019c2 Changed apply_function so that it propagates methods of the source.
bengioy@grenat.iro.umontreal.ca
parents: 28
diff changeset
83 class T_applyfunctiondataset(unittest.TestCase):
46c5c90019c2 Changed apply_function so that it propagates methods of the source.
bengioy@grenat.iro.umontreal.ca
parents: 28
diff changeset
84 def setUp(self):
46c5c90019c2 Changed apply_function so that it propagates methods of the source.
bengioy@grenat.iro.umontreal.ca
parents: 28
diff changeset
85 numpy.random.seed(123456)
46c5c90019c2 Changed apply_function so that it propagates methods of the source.
bengioy@grenat.iro.umontreal.ca
parents: 28
diff changeset
86
46c5c90019c2 Changed apply_function so that it propagates methods of the source.
bengioy@grenat.iro.umontreal.ca
parents: 28
diff changeset
87 def test_function(self):
46c5c90019c2 Changed apply_function so that it propagates methods of the source.
bengioy@grenat.iro.umontreal.ca
parents: 28
diff changeset
88 n = numpy.random.rand(3,8)
46c5c90019c2 Changed apply_function so that it propagates methods of the source.
bengioy@grenat.iro.umontreal.ca
parents: 28
diff changeset
89 a=ArrayDataSet(data=n,fields={"x":slice(2),"y":slice(1,4),"z":slice(4,6)})
46c5c90019c2 Changed apply_function so that it propagates methods of the source.
bengioy@grenat.iro.umontreal.ca
parents: 28
diff changeset
90 b=a.apply_function(lambda x,y: x+y,x+1, ['x','y'], ['x+y','x+1'], False,False,False)
46c5c90019c2 Changed apply_function so that it propagates methods of the source.
bengioy@grenat.iro.umontreal.ca
parents: 28
diff changeset
91 print b.fieldNames()
46c5c90019c2 Changed apply_function so that it propagates methods of the source.
bengioy@grenat.iro.umontreal.ca
parents: 28
diff changeset
92 print b('x+y')
46c5c90019c2 Changed apply_function so that it propagates methods of the source.
bengioy@grenat.iro.umontreal.ca
parents: 28
diff changeset
93
26
672fe4b23032 Fixed dataset errors so that _test_dataset.py works again.
bengioy@grenat.iro.umontreal.ca
parents: 22
diff changeset
94
4
f7dcfb5f9d5b Added test for dataset.
bengioy@bengiomac.local
parents:
diff changeset
95 if __name__ == '__main__':
f7dcfb5f9d5b Added test for dataset.
bengioy@bengiomac.local
parents:
diff changeset
96 unittest.main()
f7dcfb5f9d5b Added test for dataset.
bengioy@bengiomac.local
parents:
diff changeset
97