def apply_minibatches_function(self, f, X, Z):
"""Apply a function to batches of the input.
The convolutional neural networks class needs the input to be the same
size as the batch size. This function slices the input so that it
can be processed correctly.
If the batch size is not a divisor of the input size, an exception is
raised.
Parameters
----------
f : Callable
Function to use for all the batches.
X : Numpy array
Input of the function
Z : numpy array
Target of the function
Returns
-------
exprs : The average of the results of the function over all the batches.
"""
data = [minibatches(i, self.batch_size, d)
for i, d in zip([X, Z], self.sample_dim)]
if theano.config.device == 'gpu':
total = [f(*element).asndarray()
for element in zip(data[0], data[1])]
else:
total = [f(*element) for element in zip(data[0], data[1])]
return sum(total)/float(len(total))
def test_minibatches():
"""Test if minibatches are correctly generated if given a size."""
D = np.random.random((13, 5))
batches = minibatches(D, batch_size=5)
assert batches[0].shape[0] == 5
assert batches[1].shape[0] == 5
assert batches[2].shape[0] == 3
def test_minibatches():
"""Test if minibatches are correctly generated if given a size."""
D = np.random.random((13, 5))
batches = minibatches(D, batch_size=5)
assert batches[0].shape[0] == 5
assert batches[1].shape[0] == 5
assert batches[2].shape[0] == 3
def apply_minibatches_function(self, f, X, Z):
"""Apply a function to batches of the input.
The convolutional neural networks class needs the input to be the same
size as the batch size. This function slices the input so that it can be
processed correctly.
If the batch size is not a divisor of the input size, an exception is
raised.
:param f: theano function
Function to use for all the batches.
:param X: numpy array
Input of the function
:param Z: numpy array
Target of the function
:returns: The average of the results of the function over all the batches.
"""
data = [
minibatches(i, self.batch_size, d)
for i, d in zip([X, Z], self.sample_dim)
]
total = [f(*element) for element in zip(data[0], data[1])]
return sum(total) / float(len(total))
def predict(self, X):
"""Override the predict function.
:param X: numpy array
Input
:returns: The predictions of the network.
"""
data = minibatches(X, self.batch_size, 0)
if theano.config.device == 'gpu':
raise NotImplementedError(
'prediction not possible on gpu with conv net yet, please implement :)')
total = np.concatenate([super(Cnn, self).predict(element) for element in data], axis=0)
return total
def visualize_tsne(args):
model_dir = os.path.abspath(args['<model>'])
data_dir = os.path.abspath(args['<data>'])
os.chdir(model_dir)
cps = contrib.find_checkpoints('.')
if cps:
with gzip.open(cps[-1], 'rb') as fp:
trainer = cPickle.load(fp)
trainer.model.parameters.data[...] = trainer.best_pars
data = h5.File(data_dir,'r')
TX = data['test_set/test_set'][:5000]
TZ = data['test_labels/real_test_labels'][:5000]
TZ = one_hot(TZ,13)
print 'data loaded.'
if args['<mode>'] == 'cnn':
f_transformed = trainer.model.function(['inpt'],'mlp-layer-2-inpt')
print 'transform-function generated.'
data = minibatches(TX, trainer.model.batch_size, 0)
trans_TX = np.concatenate([f_transformed(element) for element in data], axis=0)
else:
f_transformed = trainer.model.function(['inpt'],'layer-2-inpt')
print 'transform-function generated.'
trans_TX = f_transformed(TX)
trans_TX = np.array(trans_TX, dtype=np.float32)
print 'data transformed'
trans_n_input = trans_TX.shape[1]
trans_tsne = Tsne(trans_n_input, 2, perplexity=5)
print 'TSNE initialized.'
trans_TX_r = trans_tsne.fit_transform(trans_TX)
print 'data TSNEd'
fig = plt.figure(figsize=(16, 16))
ax = fig.add_subplot(111)
TZ_am = TZ.argmax(axis=1)
ax.scatter(trans_TX_r[TZ_am==0, 0], trans_TX_r[TZ_am==0, 1], c='g', lw=0, alpha=1, s=100, marker='o')
ax.scatter(trans_TX_r[TZ_am==1, 0], trans_TX_r[TZ_am==1, 1], c='b', lw=0, alpha=1, s=100, marker='v')
ax.scatter(trans_TX_r[TZ_am==2, 0], trans_TX_r[TZ_am==2, 1], c='yellow', lw=0, alpha=1, s=100, marker='^')
ax.scatter(trans_TX_r[TZ_am==3, 0], trans_TX_r[TZ_am==3, 1], c='r', lw=0, alpha=1, s=100, marker='<')
ax.scatter(trans_TX_r[TZ_am==4, 0], trans_TX_r[TZ_am==4, 1], c='g', lw=0, alpha=1, s=100, marker='>')
ax.scatter(trans_TX_r[TZ_am==5, 0], trans_TX_r[TZ_am==5, 1], c='m', lw=0, alpha=1, s=100, marker='8')
ax.scatter(trans_TX_r[TZ_am==6, 0], trans_TX_r[TZ_am==6, 1], c='crimson', lw=0, alpha=1, s=100, marker='s')
ax.scatter(trans_TX_r[TZ_am==7, 0], trans_TX_r[TZ_am==7, 1], c='lawngreen', lw=0, alpha=1, s=100, marker='p')
ax.scatter(trans_TX_r[TZ_am==8, 0], trans_TX_r[TZ_am==8, 1], c='gold', lw=0, alpha=1, s=100, marker='*')
ax.scatter(trans_TX_r[TZ_am==9, 0], trans_TX_r[TZ_am==9, 1], c='darkorange', lw=0, alpha=1, s=100, marker='h')
ax.scatter(trans_TX_r[TZ_am==10, 0], trans_TX_r[TZ_am==10, 1], c='k', lw=0, alpha=1, s=100, marker='H')
ax.scatter(trans_TX_r[TZ_am==11, 0], trans_TX_r[TZ_am==11, 1], c='magenta', lw=0, alpha=1, s=100, marker='d')
ax.scatter(trans_TX_r[TZ_am==12, 0], trans_TX_r[TZ_am==12, 1], c='turquoise', lw=0, alpha=1, s=100, marker='D')
plt.savefig(os.path.join('/nthome/maugust/thesis',args['<output>']))
def predict(self, X):
"""Override the predict function.
:param X: numpy array
Input
:returns: The predictions of the network.
"""
data = minibatches(X, self.batch_size, 0)
if theano.config.device == 'gpu':
raise NotImplementedError(
'prediction not possible on gpu with conv net yet, please implement :)'
)
total = np.concatenate(
[super(Cnn, self).predict(element) for element in data], axis=0)
return total
def iter_minibatches(self, lst, batch_size, dims, n_cycles=False, random_state=None):
batches = [minibatches(i, batch_size, d) for i, d in zip(lst, dims)]
if len(batches) > 1:
if any(len(i) != len(batches[0]) for i in batches[1:]):
raise ValueError("containers to be batched have different lengths")
counter = itertools.count()
if random_state is not None:
random.seed(random_state.normal())
while True:
indices = [i for i, _ in enumerate(batches[0])]
while True:
random.shuffle(indices)
for i in indices:
yield (self.transformedData(batches[0][i]), batches[1][i])
count = counter.next()
if n_cycles and count >= n_cycles:
raise StopIteration()
def predict(self, X):
"""Override the predict function.
Parameters
----------
X : Numpy array
Input of the function
Returns
-------
exprs : The predictions of the network
"""
data = minibatches(X, self.batch_size, 0)
total = np.concatenate([super(Cnn, self).predict(element)
for element in data], axis=0)
return total
def apply_minibatches_function(self, f, X, Z):
"""Apply a function to batches of the input.
The convolutional neural networks class needs the input to be the same
size as the batch size. This function slices the input so that it can be
processed correctly.
If the batch size is not a divisor of the input size, an exception is
raised.
:param f: theano function
Function to use for all the batches.
:param X: numpy array
Input of the function
:param Z: numpy array
Target of the function
:returns: The average of the results of the function over all the batches.
"""
data = [minibatches(i, self.batch_size, d) for i, d in zip([X, Z], self.sample_dim)]
total = [f(*element) for element in zip(data[0], data[1])]
return sum(total) / float(len(total))
def iter_minibatches(self,
lst,
batch_size,
dims,
n_cycles=False,
random_state=None):
batches = [minibatches(i, batch_size, d) for i, d in zip(lst, dims)]
if len(batches) > 1:
if any(len(i) != len(batches[0]) for i in batches[1:]):
raise ValueError(
"containers to be batched have different lengths")
counter = itertools.count()
if random_state is not None:
random.seed(random_state.normal())
while True:
indices = [i for i, _ in enumerate(batches[0])]
while True:
random.shuffle(indices)
for i in indices:
yield (self.transformedData(batches[0][i]),
batches[1][i])
count = counter.next()
if n_cycles and count >= n_cycles:
raise StopIteration()