def test_extract_reassemble():
""" Tests that ExtractGridPatches and ReassembleGridPatches are
inverse of each other """
rng = np.random.RandomState([1, 3, 7])
topo = rng.randn(4, 3 * 5, 3 * 7, 2)
dataset = DenseDesignMatrix(topo_view=topo)
patch_shape = (3, 7)
extractor = ExtractGridPatches(patch_shape, patch_shape)
reassemblor = ReassembleGridPatches(patch_shape=patch_shape,
orig_shape=topo.shape[1:3])
dataset.apply_preprocessor(extractor)
dataset.apply_preprocessor(reassemblor)
new_topo = dataset.get_topological_view()
assert new_topo.shape == topo.shape
if not np.all(new_topo == topo):
assert False
else:
assert False
print 'compiling theano function'
f = function([V], feat)
print 'running theano function'
feat = f(X2)
feat_dataset = DenseDesignMatrix(X=feat,
view_converter=DefaultViewConverter(
[1, 1, feat.shape[1]]))
print 'reassembling features'
ns = 32 - size + 1
depatchifier = ReassembleGridPatches(orig_shape=(ns, ns),
patch_shape=(1, 1))
feat_dataset.apply_preprocessor(depatchifier)
print 'making topological view'
topo_feat = feat_dataset.get_topological_view()
assert topo_feat.shape[0] == X.shape[0]
print 'assembling visualizer'
n = np.ceil(np.sqrt(model.nhid))
pv3 = PatchViewer(grid_shape=(X.shape[0], num_filters),
patch_shape=(ns, ns),
is_color=False)
pv4 = PatchViewer(grid_shape=(n, n),
patch_shape=(size, size),
def __call__(self, full_X):
feature_type = self.feature_type
pooling_region_counts = self.pooling_region_counts
model = self.model
size = self.size
nan = 0
full_X = full_X.reshape(1, full_X.shape[0], full_X.shape[1],
full_X.shape[2])
if full_X.shape[3] == 1:
full_X = np.concatenate((full_X, full_X, full_X), axis=3)
print 'full_X.shape: ' + str(full_X.shape)
num_examples = full_X.shape[0]
assert num_examples == 1
pipeline = self.preprocessor
def average_pool(stride):
def point(p):
return p * ns / stride
rval = np.zeros(
(topo_feat.shape[0], stride, stride, topo_feat.shape[3]),
dtype='float32')
for i in xrange(stride):
for j in xrange(stride):
rval[:, i, j, :] = self.region_features(
topo_feat[:,
point(i):point(i + 1),
point(j):point(j + 1), :])
return rval
outputs = [
np.zeros((num_examples, count, count, model.nhid), dtype='float32')
for count in pooling_region_counts
]
assert len(outputs) > 0
fd = DenseDesignMatrix(X=np.zeros((1, 1), dtype='float32'),
view_converter=DefaultViewConverter(
[1, 1, model.nhid]))
ns = 32 - size + 1
depatchifier = ReassembleGridPatches(orig_shape=(ns, ns),
patch_shape=(1, 1))
batch_size = 1
for i in xrange(0, num_examples - batch_size + 1, batch_size):
print i
t1 = time.time()
d = DenseDesignMatrix(
topo_view=np.cast['float32'](full_X[i:i + batch_size, :]),
view_converter=DefaultViewConverter((32, 32, 3)))
t2 = time.time()
#print '\tapplying preprocessor'
d.apply_preprocessor(pipeline, can_fit=False)
X2 = d.get_design_matrix()
t3 = time.time()
#print '\trunning theano function'
feat = self.f(X2)
t4 = time.time()
assert feat.dtype == 'float32'
feat_dataset = copy.copy(fd)
if np.any(np.isnan(feat)):
nan += np.isnan(feat).sum()
feat[np.isnan(feat)] = 0
feat_dataset.set_design_matrix(feat)
#print '\treassembling features'
feat_dataset.apply_preprocessor(depatchifier)
#print '\tmaking topological view'
topo_feat = feat_dataset.get_topological_view()
assert topo_feat.shape[0] == batch_size
t5 = time.time()
#average pooling
for output, count in zip(outputs, pooling_region_counts):
output[i:i + batch_size, ...] = average_pool(count)
t6 = time.time()
print(t6 - t1, t2 - t1, t3 - t2, t4 - t3, t5 - t4, t6 - t5)
return outputs[0]
def _execute(self):
global pooling_matrix
save_path = self.save_path
batch_size = self.batch_size
feature_type = self.feature_type
dataset_family = self.dataset_family
which_set = self.which_set
model = self.model
size = self.size
nan = 0
dataset_descriptor = dataset_family[which_set][size]
dataset = dataset_descriptor.dataset_maker()
expected_num_examples = dataset_descriptor.num_examples
full_X = dataset.get_design_matrix()
num_examples = full_X.shape[0]
assert num_examples == expected_num_examples
if self.restrict is not None:
assert self.restrict[1] <= full_X.shape[0]
print 'restricting to examples ', self.restrict[
0], ' through ', self.restrict[1], ' exclusive'
full_X = full_X[self.restrict[0]:self.restrict[1], :]
assert self.restrict[1] > self.restrict[0]
#update for after restriction
num_examples = full_X.shape[0]
assert num_examples > 0
dataset.X = None
dataset.design_loc = None
dataset.compress = False
patchifier = ExtractGridPatches(patch_shape=(size, size),
patch_stride=(1, 1))
pipeline = serial.load(dataset_descriptor.pipeline_path)
assert isinstance(pipeline.items[0], ExtractPatches)
pipeline.items[0] = patchifier
print 'defining features'
V = T.matrix('V')
model.make_pseudoparams()
d = model.e_step.variational_inference(V=V)
H = d['H_hat']
Mu1 = d['S_hat']
assert H.dtype == 'float32'
assert Mu1.dtype == 'float32'
if self.feature_type == 'map_hs':
feat = (H > 0.5) * Mu1
elif self.feature_type == 'map_h':
feat = T.cast(H > 0.5, dtype='float32')
elif self.feature_type == 'exp_hs':
feat = H * Mu1
elif self.feature_type == 'exp_h':
feat = H
elif self.feature_type == 'exp_h_thresh':
feat = H * (H > .01)
else:
raise NotImplementedError()
assert feat.dtype == 'float32'
print 'compiling theano function'
f = function([V], feat)
if config.device.startswith('gpu') and model.nhid >= 4000:
f = halver(f, model.nhid)
topo_feat_var = T.TensorType(broadcastable=(False, False, False,
False),
dtype='float32')()
region_features = function([topo_feat_var],
topo_feat_var.mean(axis=(1, 2)))
def average_pool(stride):
def point(p):
return p * ns / stride
rval = np.zeros(
(topo_feat.shape[0], stride, stride, topo_feat.shape[3]),
dtype='float32')
for i in xrange(stride):
for j in xrange(stride):
rval[:, i, j, :] = region_features(
topo_feat[:,
point(i):point(i + 1),
point(j):point(j + 1), :])
return rval
num_superpixels = 7
output = np.zeros((num_examples, pooling_matrix.shape[0]),
dtype='float32')
fd = DenseDesignMatrix(X=np.zeros((1, 1), dtype='float32'),
view_converter=DefaultViewConverter(
[1, 1, model.nhid]))
ns = 32 - size + 1
depatchifier = ReassembleGridPatches(orig_shape=(ns, ns),
patch_shape=(1, 1))
if len(range(0, num_examples - batch_size + 1, batch_size)) <= 0:
print num_examples
print batch_size
for i in xrange(0, num_examples - batch_size + 1, batch_size):
print i
t1 = time.time()
d = copy.copy(dataset)
d.set_design_matrix(full_X[i:i + batch_size, :])
t2 = time.time()
#print '\tapplying preprocessor'
d.apply_preprocessor(pipeline, can_fit=False)
X2 = d.get_design_matrix()
t3 = time.time()
#print '\trunning theano function'
feat = f(X2)
t4 = time.time()
assert feat.dtype == 'float32'
feat_dataset = copy.copy(fd)
if np.any(np.isnan(feat)):
nan += np.isnan(feat).sum()
feat[np.isnan(feat)] = 0
feat_dataset.set_design_matrix(feat)
#print '\treassembling features'
feat_dataset.apply_preprocessor(depatchifier)
#print '\tmaking topological view'
topo_feat = feat_dataset.get_topological_view()
assert topo_feat.shape[0] == batch_size
t5 = time.time()
#average pooling
superpixels = average_pool(num_superpixels)
pooled = pooling_matrix.dot(superpixels.T).T
output[i:i + batch_size, :] = pooled
t6 = time.time()
print(t6 - t1, t2 - t1, t3 - t2, t4 - t3, t5 - t4, t6 - t5)
if self.chunk_size is not None:
assert save_path.endswith('.npy')
save_path_pieces = save_path.split('.npy')
assert len(save_path_pieces) == 2
assert save_path_pieces[1] == ''
save_path = save_path_pieces[0] + '_' + chr(
ord('A') + self.chunk_id) + '.npy'
np.save(save_path, output)
if nan > 0:
warnings.warn(str(nan) + ' features were nan')
def _execute(self):
global num_superpixels
num_output_features = self.num_output_features
idxs = self.idxs
top = self.top
bottom = self.bottom
left = self.left
right = self.right
save_path = self.save_path
batch_size = self.batch_size
dataset_family = self.dataset_family
which_set = self.which_set
model = self.model
size = self.size
nan = 0
dataset_descriptor = dataset_family[which_set][size]
dataset = dataset_descriptor.dataset_maker()
expected_num_examples = dataset_descriptor.num_examples
full_X = dataset.get_design_matrix()
num_examples = full_X.shape[0]
assert num_examples == expected_num_examples
if self.restrict is not None:
assert self.restrict[1] <= full_X.shape[0]
print 'restricting to examples ', self.restrict[
0], ' through ', self.restrict[1], ' exclusive'
full_X = full_X[self.restrict[0]:self.restrict[1], :]
assert self.restrict[1] > self.restrict[0]
#update for after restriction
num_examples = full_X.shape[0]
assert num_examples > 0
dataset.X = None
dataset.design_loc = None
dataset.compress = False
patchifier = ExtractGridPatches(patch_shape=(size, size),
patch_stride=(1, 1))
pipeline = serial.load(dataset_descriptor.pipeline_path)
assert isinstance(pipeline.items[0], ExtractPatches)
pipeline.items[0] = patchifier
print 'defining features'
V = T.matrix('V')
mu = model.mu
feat = triangle_code(V, mu)
assert feat.dtype == 'float32'
print 'compiling theano function'
f = function([V], feat)
nhid = model.mu.get_value().shape[0]
if config.device.startswith('gpu') and nhid >= 4000:
f = halver(f, model.nhid)
topo_feat_var = T.TensorType(broadcastable=(False, False, False,
False),
dtype='float32')()
if self.pool_mode == 'mean':
region_features = function([topo_feat_var],
topo_feat_var.mean(axis=(1, 2)))
elif self.pool_mode == 'max':
region_features = function([topo_feat_var],
topo_feat_var.max(axis=(1, 2)))
else:
assert False
def average_pool(stride):
def point(p):
return p * ns / stride
rval = np.zeros(
(topo_feat.shape[0], stride, stride, topo_feat.shape[3]),
dtype='float32')
for i in xrange(stride):
for j in xrange(stride):
rval[:, i, j, :] = region_features(
topo_feat[:,
point(i):point(i + 1),
point(j):point(j + 1), :])
return rval
output = np.zeros((num_examples, num_output_features), dtype='float32')
fd = DenseDesignMatrix(X=np.zeros((1, 1), dtype='float32'),
view_converter=DefaultViewConverter(
[1, 1, nhid]))
ns = 32 - size + 1
depatchifier = ReassembleGridPatches(orig_shape=(ns, ns),
patch_shape=(1, 1))
if len(range(0, num_examples - batch_size + 1, batch_size)) <= 0:
print num_examples
print batch_size
for i in xrange(0, num_examples - batch_size + 1, batch_size):
print i
t1 = time.time()
d = copy.copy(dataset)
d.set_design_matrix(full_X[i:i + batch_size, :])
t2 = time.time()
#print '\tapplying preprocessor'
d.apply_preprocessor(pipeline, can_fit=False)
X2 = d.get_design_matrix()
t3 = time.time()
#print '\trunning theano function'
feat = f(X2)
t4 = time.time()
assert feat.dtype == 'float32'
feat_dataset = copy.copy(fd)
if np.any(np.isnan(feat)):
nan += np.isnan(feat).sum()
feat[np.isnan(feat)] = 0
feat_dataset.set_design_matrix(feat)
#print '\treassembling features'
feat_dataset.apply_preprocessor(depatchifier)
#print '\tmaking topological view'
topo_feat = feat_dataset.get_topological_view()
assert topo_feat.shape[0] == batch_size
t5 = time.time()
#average pooling
superpixels = average_pool(num_superpixels)
assert batch_size == 1
if self.pool_mode == 'mean':
for j in xrange(num_output_features):
output[i:i + batch_size,
j] = superpixels[:, top[j]:bottom[j] + 1,
left[j]:right[j] + 1,
idxs[j]].mean()
elif self.pool_mode == 'max':
for j in xrange(num_output_features):
output[i:i + batch_size,
j] = superpixels[:, top[j]:bottom[j] + 1,
left[j]:right[j] + 1,
idxs[j]].max()
else:
assert False
assert output[i:i + batch_size, :].max() < 1e20
t6 = time.time()
print(t6 - t1, t2 - t1, t3 - t2, t4 - t3, t5 - t4, t6 - t5)
if self.chunk_size is not None:
assert save_path.endswith('.npy')
save_path_pieces = save_path.split('.npy')
assert len(save_path_pieces) == 2
assert save_path_pieces[1] == ''
save_path = save_path_pieces[0] + '_' + chr(
ord('A') + self.chunk_id) + '.npy'
np.save(save_path, output)
if nan > 0:
warnings.warn(str(nan) + ' features were nan')
def _execute(self):
batch_size = self.batch_size
feature_type = self.feature_type
pooling_region_counts = self.pooling_region_counts
dataset_family = self.dataset_family
which_set = self.which_set
model = self.model
size = self.size
nan = 0
dataset_descriptor = dataset_family[which_set][size]
dataset = dataset_descriptor.dataset_maker()
expected_num_examples = dataset_descriptor.num_examples
full_X = dataset.get_design_matrix()
assert full_X.dtype == 'float32'
num_examples = full_X.shape[0]
assert num_examples == expected_num_examples
print 'restricting to examples from classes 0 and 1'
full_X = full_X[dataset.y_fine < 2, :]
#update for after restriction
num_examples = full_X.shape[0]
assert num_examples > 0
dataset.X = None
dataset.design_loc = None
dataset.compress = False
patchifier = ExtractGridPatches(patch_shape=(size, size),
patch_stride=(1, 1))
pipeline = serial.load(dataset_descriptor.pipeline_path)
assert isinstance(pipeline.items[0], ExtractPatches)
pipeline.items[0] = patchifier
print 'defining features'
V = T.matrix('V')
assert V.type.dtype == 'float32'
model.make_pseudoparams()
d = model.infer(V=V)
H = d['H_hat']
Mu1 = d['S_hat']
G = d['G_hat']
if len(G) != 1:
raise NotImplementedError(
"only supports two layer pd-dbms for now")
G, = G
assert H.dtype == 'float32'
assert Mu1.dtype == 'float32'
nfeat = model.s3c.nhid + model.dbm.rbms[0].nhid
if self.feature_type == 'map_hs':
feat = (H > 0.5) * Mu1
raise NotImplementedError("doesn't support layer 2")
elif self.feature_type == 'map_h':
feat = T.cast(H > 0.5, dtype='float32')
raise NotImplementedError("doesn't support layer 2")
elif self.feature_type == 'exp_hs':
feat = H * Mu1
raise NotImplementedError("doesn't support layer 2")
elif self.feature_type == 'exp_hs_split':
Z = H * Mu1
pos = T.clip(Z, 0., 1e32)
neg = T.clip(-Z, 0, 1e32)
feat = T.concatenate((pos, neg), axis=1)
nfeat *= 2
raise NotImplementedError("doesn't support layer 2")
elif self.feature_type == 'exp_h,exp_g':
feat = T.concatenate((H, G), axis=1)
elif self.feature_type == 'exp_h_thresh':
feat = H * (H > .01)
raise NotImplementedError("doesn't support layer 2")
else:
raise NotImplementedError()
assert feat.dtype == 'float32'
print 'compiling theano function'
f = function([V], feat)
if config.device.startswith('gpu') and nfeat >= 4000:
f = halver(f, nfeat)
topo_feat_var = T.TensorType(broadcastable=(False, False, False,
False),
dtype='float32')()
if self.pool_mode == 'mean':
region_feat_var = topo_feat_var.mean(axis=(1, 2))
elif self.pool_mode == 'max':
region_feat_var = topo_feat_var.max(axis=(1, 2))
else:
raise ValueError("Unknown pool mode: " + self.pool_mode)
region_features = function([topo_feat_var], region_feat_var)
def average_pool(stride):
def point(p):
return p * ns / stride
rval = np.zeros(
(topo_feat.shape[0], stride, stride, topo_feat.shape[3]),
dtype='float32')
for i in xrange(stride):
for j in xrange(stride):
rval[:, i, j, :] = region_features(
topo_feat[:,
point(i):point(i + 1),
point(j):point(j + 1), :])
return rval
outputs = [
np.zeros((num_examples, count, count, nfeat), dtype='float32')
for count in pooling_region_counts
]
assert len(outputs) > 0
fd = DenseDesignMatrix(X=np.zeros((1, 1), dtype='float32'),
view_converter=DefaultViewConverter(
[1, 1, nfeat]))
ns = 32 - size + 1
depatchifier = ReassembleGridPatches(orig_shape=(ns, ns),
patch_shape=(1, 1))
if len(range(0, num_examples - batch_size + 1, batch_size)) <= 0:
print num_examples
print batch_size
for i in xrange(0, num_examples - batch_size + 1, batch_size):
print i
t1 = time.time()
d = copy.copy(dataset)
d.set_design_matrix(full_X[i:i + batch_size, :])
t2 = time.time()
#print '\tapplying preprocessor'
d.apply_preprocessor(pipeline, can_fit=False)
X2 = np.cast['float32'](d.get_design_matrix())
t3 = time.time()
#print '\trunning theano function'
feat = f(X2)
t4 = time.time()
assert feat.dtype == 'float32'
feat_dataset = copy.copy(fd)
if np.any(np.isnan(feat)):
nan += np.isnan(feat).sum()
feat[np.isnan(feat)] = 0
feat_dataset.set_design_matrix(feat)
#print '\treassembling features'
feat_dataset.apply_preprocessor(depatchifier)
#print '\tmaking topological view'
topo_feat = feat_dataset.get_topological_view()
assert topo_feat.shape[0] == batch_size
t5 = time.time()
#average pooling
for output, count in zip(outputs, pooling_region_counts):
output[i:i + batch_size, ...] = average_pool(count)
t6 = time.time()
print(t6 - t1, t2 - t1, t3 - t2, t4 - t3, t5 - t4, t6 - t5)
for output, save_path in zip(outputs, self.save_paths):
np.save(save_path, output)
if nan > 0:
warnings.warn(str(nan) + ' features were nan')
def _execute(self):
batch_size = self.batch_size
pooling_region_counts = self.pooling_region_counts
dataset_family = self.dataset_family
which_set = self.which_set
size = self.size
nan = 0
dataset_descriptor = dataset_family[which_set][size]
dataset = dataset_descriptor.dataset_maker()
expected_num_examples = dataset_descriptor.num_examples
full_X = dataset.get_design_matrix()
num_examples = full_X.shape[0]
assert num_examples == expected_num_examples
if self.restrict is not None:
assert self.restrict[1] <= full_X.shape[0]
print 'restricting to examples ',self.restrict[0],' through ',self.restrict[1],' exclusive'
full_X = full_X[self.restrict[0]:self.restrict[1],:]
assert self.restrict[1] > self.restrict[0]
#update for after restriction
num_examples = full_X.shape[0]
assert num_examples > 0
dataset.X = None
dataset.design_loc = None
dataset.compress = False
patchifier = ExtractGridPatches( patch_shape = (size,size), patch_stride = (1,1) )
pipeline = serial.load(dataset_descriptor.pipeline_path)
assert isinstance(pipeline.items[0], ExtractPatches)
pipeline.items[0] = patchifier
print 'defining features'
Z = T.matrix('Z')
if self.one_sided:
feat = abs(Z)
else:
pos = T.clip(Z,0.,1e30)
neg = T.clip(-Z,0.,1e30)
feat = T.concatenate((pos, neg), axis=1)
print 'compiling theano function'
f = function([Z],feat)
nfeat = self.W.shape[1] * (2 - self.one_sided)
if not (nfeat == 1600 or nfeat == 3200):
print nfeat
assert False
if config.device.startswith('gpu') and nfeat >= 4000:
f = halver(f, nfeat)
topo_feat_var = T.TensorType(broadcastable = (False,False,False,False), dtype='float32')()
region_features = function([topo_feat_var],
topo_feat_var.mean(axis=(1,2)) )
def average_pool( stride ):
def point( p ):
return p * ns / stride
rval = np.zeros( (topo_feat.shape[0], stride, stride, topo_feat.shape[3] ) , dtype = 'float32')
for i in xrange(stride):
for j in xrange(stride):
rval[:,i,j,:] = region_features( topo_feat[:,point(i):point(i+1), point(j):point(j+1),:] )
return rval
outputs = [ np.zeros((num_examples,count,count,nfeat),dtype='float32') for count in pooling_region_counts ]
assert len(outputs) > 0
fd = DenseDesignMatrix(X = np.zeros((1,1),dtype='float32'), view_converter = DefaultViewConverter([1, 1, nfeat] ) )
ns = 32 - size + 1
depatchifier = ReassembleGridPatches( orig_shape = (ns, ns), patch_shape=(1,1) )
if len(range(0,num_examples-batch_size+1,batch_size)) <= 0:
print num_examples
print batch_size
for i in xrange(0,num_examples-batch_size+1,batch_size):
print i
t1 = time.time()
d = copy.copy(dataset)
d.set_design_matrix(full_X[i:i+batch_size,:])
t2 = time.time()
#print '\tapplying preprocessor'
d.apply_preprocessor(pipeline, can_fit = False)
X2 = d.get_design_matrix()
t3 = time.time()
M.put(s,'batch',X2)
M.eval(s, 'Z = sparse_codes(batch, dictionary, lambda)')
Z = M.get(s, 'Z')
feat = f(np.cast['float32'](Z))
t4 = time.time()
assert feat.dtype == 'float32'
feat_dataset = copy.copy(fd)
if np.any(np.isnan(feat)):
nan += np.isnan(feat).sum()
feat[np.isnan(feat)] = 0
feat_dataset.set_design_matrix(feat)
#print '\treassembling features'
feat_dataset.apply_preprocessor(depatchifier)
#print '\tmaking topological view'
topo_feat = feat_dataset.get_topological_view()
assert topo_feat.shape[0] == batch_size
t5 = time.time()
#average pooling
for output, count in zip(outputs, pooling_region_counts):
output[i:i+batch_size,...] = average_pool(count)
t6 = time.time()
print (t6-t1, t2-t1, t3-t2, t4-t3, t5-t4, t6-t5)
for output, save_path in zip(outputs, self.save_paths):
if self.chunk_size is not None:
assert save_path.endswith('.npy')
save_path_pieces = save_path.split('.npy')
assert len(save_path_pieces) == 2
assert save_path_pieces[1] == ''
save_path = save_path_pieces[0] + '_' + chr(ord('A')+self.chunk_id)+'.npy'
np.save(save_path,output)
if nan > 0:
warnings.warn(str(nan)+' features were nan')