def logscale_spec(spec, sr=44100, factor=20.):
timebins, freqbins = np.shape(spec)
scale = np.linspace(0, 1, freqbins) ** factor
scale *= (freqbins-1)/max(scale)
scale = np.unique(np.round(scale))
# create spectrogram with new freq bins
newspec = np.complex128(np.zeros([timebins, len(scale)]))
for i in range(0, len(scale)):
if i == len(scale)-1:
newspec[:,i] = np.sum(spec[:,scale[i]:], axis=1)
else:
newspec[:,i] = np.sum(spec[:,scale[i]:scale[i+1]], axis=1)
# list center freq of bins
allfreqs = np.abs(np.fft.fftfreq(freqbins*2, 1./sr)[:freqbins+1])
freqs = []
for i in range(0, len(scale)):
if i == len(scale)-1:
freqs += [np.mean(allfreqs[scale[i]:])]
else:
freqs += [np.mean(allfreqs[scale[i]:scale[i+1]])]
return newspec, freqs
python类shape()的实例源码
def weightVariable(shape,std=1.0,name=None):
# Create a set of weights initialized with truncated normal random values
name = 'weights' if name is None else name
return tf.get_variable(name,shape,initializer=tf.truncated_normal_initializer(stddev=std/math.sqrt(shape[0])))
def biasVariable(shape,bias=0.1,name=None):
# create a set of bias nodes initialized with a constant 0.1
name = 'biases' if name is None else name
return tf.get_variable(name,shape,initializer=tf.constant_initializer(bias))
def max_pool(x,shape,name=None):
# return an op that performs max pooling across a 2D image
return tf.nn.max_pool(x,ksize=[1]+shape+[1],strides=[1]+shape+[1],padding='SAME',name=name)
def max_pool3d(x,shape,name=None):
# return an op that performs max pooling across a 2D image
return tf.nn.max_pool3d(x,ksize=[1]+shape+[1],strides=[1]+shape+[1],padding='SAME',name=name)
def plotFields(layer,fieldShape=None,channel=None,figOffset=1,cmap=None,padding=0.01):
# Receptive Fields Summary
try:
W = layer.W
except:
W = layer
wp = W.eval().transpose();
if len(np.shape(wp)) < 4: # Fully connected layer, has no shape
fields = np.reshape(wp,list(wp.shape[0:-1])+fieldShape)
else: # Convolutional layer already has shape
features, channels, iy, ix = np.shape(wp)
if channel is not None:
fields = wp[:,channel,:,:]
else:
fields = np.reshape(wp,[features*channels,iy,ix])
perRow = int(math.floor(math.sqrt(fields.shape[0])))
perColumn = int(math.ceil(fields.shape[0]/float(perRow)))
fig = mpl.figure(figOffset); mpl.clf()
# Using image grid
from mpl_toolkits.axes_grid1 import ImageGrid
grid = ImageGrid(fig,111,nrows_ncols=(perRow,perColumn),axes_pad=padding,cbar_mode='single')
for i in range(0,np.shape(fields)[0]):
im = grid[i].imshow(fields[i],cmap=cmap);
grid.cbar_axes[0].colorbar(im)
mpl.title('%s Receptive Fields' % layer.name)
# old way
# fields2 = np.vstack([fields,np.zeros([perRow*perColumn-fields.shape[0]] + list(fields.shape[1:]))])
# tiled = []
# for i in range(0,perColumn*perRow,perColumn):
# tiled.append(np.hstack(fields2[i:i+perColumn]))
#
# tiled = np.vstack(tiled)
# mpl.figure(figOffset); mpl.clf(); mpl.imshow(tiled,cmap=cmap); mpl.title('%s Receptive Fields' % layer.name); mpl.colorbar();
mpl.figure(figOffset+1); mpl.clf(); mpl.imshow(np.sum(np.abs(fields),0),cmap=cmap); mpl.title('%s Total Absolute Input Dependency' % layer.name); mpl.colorbar()
def __init__(self,input,shape,name,strides=[1,1,1,1],std=1.0,bias=0.1):
self.input = input
self.units = shape[-1]
self.shape = shape
self.strides = strides
self.name = name
self.initialize(std=std,bias=bias)
self.setupOutput()
self.setupSummary()
def initialize(self,std=1.0,bias=0.1):
with tf.variable_scope(self.name):
self.W = weightVariable(self.shape,std=std) # YxX patch, Z contrast, outputs to N neurons
self.b = biasVariable([self.shape[-1]],bias=bias) # N bias variables to go with the N neurons
def __init__(self,input,shape,name,strides=[1,1,1,1,1],std=1.0,bias=0.1):
super(Conv3D,self).__init__(input,shape,name,strides,std,bias)
def __init__(self,input,shape,name):
self.shape = shape
super(MaxPool,self).__init__(input,name)
def setupOutput(self):
with tf.variable_scope(self.name):
self.output = max_pool(self.input,shape=self.shape)
def dense_to_one_hot(labels_dense, num_classes=10):
"""Convert class labels from scalars to one-hot vectors."""
num_labels = labels_dense.shape[0]
index_offset = numpy.arange(num_labels) * num_classes
labels_one_hot = numpy.zeros((num_labels, num_classes))
labels_one_hot.flat[index_offset + labels_dense.ravel()] = 1
return labels_one_hot
def __init__(self, images, labels, fake_data=False):
if fake_data:
self._num_examples = 10000
else:
assert images.shape[0] == labels.shape[0], (
"images.shape: %s labels.shape: %s" % (images.shape,
labels.shape))
self._num_examples = images.shape[0]
# Convert shape from [num examples, rows, columns, depth]
# to [num examples, rows*columns] (assuming depth == 1)
self.imageShape = images.shape[1:]
self.imageChannels = self.imageShape[2]
images = images.reshape(images.shape[0],
images.shape[1] * images.shape[2] * images.shape[3])
# Convert from [0, 255] -> [0.0, 1.0].
images = images.astype(numpy.float32)
images = numpy.multiply(images, 1.0 / 255.0)
self._images = images
self._labels = labels
try:
if len(numpy.shape(self._labels)) == 1:
self._labels = dense_to_one_hot(self._labels,len(numpy.unique(self._labels)))
except:
traceback.print_exc()
self._epochs_completed = 0
self._index_in_epoch = 0
def weightVariable(shape,std=1.0,name=None):
# Create a set of weights initialized with truncated normal random values
name = 'weights' if name is None else name
return tf.get_variable(name,shape,initializer=tf.truncated_normal_initializer(stddev=std/math.sqrt(shape[0])))
def biasVariable(shape,bias=0.1,name=None):
# create a set of bias nodes initialized with a constant 0.1
name = 'biases' if name is None else name
return tf.get_variable(name,shape,initializer=tf.constant_initializer(bias))
def max_pool(x,shape,name=None):
# return an op that performs max pooling across a 2D image
return tf.nn.max_pool(x,ksize=[1]+shape+[1],strides=[1]+shape+[1],padding='SAME',name=name)
def plotFields(layer,fieldShape=None,channel=None,maxFields=25,figName='ReceptiveFields',cmap=None,padding=0.01):
# Receptive Fields Summary
W = layer.W
wp = W.eval().transpose();
if len(np.shape(wp)) < 4: # Fully connected layer, has no shape
fields = np.reshape(wp,list(wp.shape[0:-1])+fieldShape)
else: # Convolutional layer already has shape
features, channels, iy, ix = np.shape(wp)
if channel is not None:
fields = wp[:,channel,:,:]
else:
fields = np.reshape(wp,[features*channels,iy,ix])
fieldsN = min(fields.shape[0],maxFields)
perRow = int(math.floor(math.sqrt(fieldsN)))
perColumn = int(math.ceil(fieldsN/float(perRow)))
fig = mpl.figure(figName); mpl.clf()
# Using image grid
from mpl_toolkits.axes_grid1 import ImageGrid
grid = ImageGrid(fig,111,nrows_ncols=(perRow,perColumn),axes_pad=padding,cbar_mode='single')
for i in range(0,fieldsN):
im = grid[i].imshow(fields[i],cmap=cmap);
grid.cbar_axes[0].colorbar(im)
mpl.title('%s Receptive Fields' % layer.name)
# old way
# fields2 = np.vstack([fields,np.zeros([perRow*perColumn-fields.shape[0]] + list(fields.shape[1:]))])
# tiled = []
# for i in range(0,perColumn*perRow,perColumn):
# tiled.append(np.hstack(fields2[i:i+perColumn]))
#
# tiled = np.vstack(tiled)
# mpl.figure(figOffset); mpl.clf(); mpl.imshow(tiled,cmap=cmap); mpl.title('%s Receptive Fields' % layer.name); mpl.colorbar();
mpl.figure(figName+' Total'); mpl.clf(); mpl.imshow(np.sum(np.abs(fields),0),cmap=cmap); mpl.title('%s Total Absolute Input Dependency' % layer.name); mpl.colorbar()
def plotOutput(layer,feed_dict,fieldShape=None,channel=None,figOffset=1,cmap=None):
# Output summary
W = layer.output
wp = W.eval(feed_dict=feed_dict);
if len(np.shape(wp)) < 4: # Fully connected layer, has no shape
temp = np.zeros(np.product(fieldShape)); temp[0:np.shape(wp.ravel())[0]] = wp.ravel()
fields = np.reshape(temp,[1]+fieldShape)
else: # Convolutional layer already has shape
wp = np.rollaxis(wp,3,0)
features, channels, iy,ix = np.shape(wp)
if channel is not None:
fields = wp[:,channel,:,:]
else:
fields = np.reshape(wp,[features*channels,iy,ix])
perRow = int(math.floor(math.sqrt(fields.shape[0])))
perColumn = int(math.ceil(fields.shape[0]/float(perRow)))
fields2 = np.vstack([fields,np.zeros([perRow*perColumn-fields.shape[0]] + list(fields.shape[1:]))])
tiled = []
for i in range(0,perColumn*perRow,perColumn):
tiled.append(np.hstack(fields2[i:i+perColumn]))
tiled = np.vstack(tiled)
if figOffset is not None:
mpl.figure(figOffset); mpl.clf();
mpl.imshow(tiled,cmap=cmap); mpl.title('%s Output' % layer.name); mpl.colorbar();
def __init__(self,input,shape,name,std=1.0,bias=0.1):
self.input = input
self.units = shape[-1]
self.shape = shape
self.name = name
self.initialize(std=std,bias=bias)
self.setupOutput()
self.setupSummary()
def initialize(self,std=1.0,bias=0.1):
with tf.variable_scope(self.name):
self.W = weightVariable(self.shape,std=std) # YxX patch, Z contrast, outputs to N neurons
self.b = biasVariable([self.shape[-1]],bias=bias) # N bias variables to go with the N neurons
