def scatter(pts, nsamples=100, colormap=None, scale=1, thresh=0.001, ax=None, store=False):
"""Create a scatter plot of x and y points from an array or an RDD (through sampling)
Can optionally use the values to determine colors"""
if ax is None:
ax = pyplot.gca()
if isrdd(pts):
if thresh is not None:
pts = array(pts.values().filter(lambda x: std(x) > thresh).takeSample(False, nsamples))
else:
pts = array(pts.values().takeSample(False, nsamples))
if len(pts) == 0:
raise Exception('no samples found, most likely your threshold is too low')
else:
pts = asarray(pts)
if colormap is not None:
# pass in strings or actual colormap objects
if isinstance(colormap, basestring):
clrs = Colorize(colormap, scale).calc(pts)
else:
clrs = colormap.calc(pts)
else:
clrs = 'indianred'
h = ax.scatter(pts[:, 0], pts[:, 1], s=100, c=clrs, alpha=0.6, edgecolor='black', linewidth=0.2)
if store is True:
return ax, h, pts
else:
return ax, h
def calc(self, data):
if isrdd(data):
self.checkargs(size(data.first()[1]))
return data.mapValues(lambda x: self.get(x))
else:
self.checkargs(size(data[0]))
return map(lambda line: self.get(line), data)
def calc(self, data):
if isrdd(data):
self.checkargs(size(data.first()[1]))
return data.mapValues(lambda x: self.get(x))
else:
self.checkargs(size(data[0]))
return map(lambda line: self.get(line), data)
def scatter(pts,
nsamples=100,
colormap=None,
scale=1,
thresh=0.001,
ax=None,
store=False):
"""Create a scatter plot of x and y points from an array or an RDD (through sampling)
Can optionally use the values to determine colors"""
if ax is None:
ax = pyplot.gca()
if isrdd(pts):
if thresh is not None:
pts = array(
pts.values().filter(lambda x: std(x) > thresh).takeSample(
False, nsamples))
else:
pts = array(pts.values().takeSample(False, nsamples))
if len(pts) == 0:
raise Exception(
'no samples found, most likely your threshold is too low')
else:
pts = asarray(pts)
if colormap is not None:
# pass in strings or actual colormap objects
if isinstance(colormap, basestring):
clrs = Colorize(colormap, scale).calc(pts)
else:
clrs = colormap.calc(pts)
else:
clrs = 'indianred'
h = ax.scatter(pts[:, 0],
pts[:, 1],
s=100,
c=clrs,
alpha=0.6,
edgecolor='black',
linewidth=0.2)
if store is True:
return ax, h, pts
else:
return ax, h
def calc(self, data, func):
"""Base function for making clustering predictions"""
# small optimization to avoid serializing full model
centers = self.centers
if isrdd(data):
return data.mapValues(lambda x: func(centers, x))
elif isinstance(data, list):
return map(lambda x: func(centers, x), data)
elif isinstance(data, ndarray):
if data.ndim == 1:
return func(centers, data)
else:
return map(lambda x: func(centers, x), data)
def calc(self, data, func):
"""Base function for making clustering predictions"""
# small optimization to avoid serializing full model
centers = self.centers
if isrdd(data):
return data.mapValues(lambda x: func(centers, x))
elif isinstance(data, list):
return map(lambda x: func(centers, x), data)
elif isinstance(data, ndarray):
if data.ndim == 1:
return func(centers, data)
else:
return map(lambda x: func(centers, x), data)
def pointmap(data, colormap='polar', scale=1.0, ax=None):
"""Create a spatial point map from a collection of key-value pairs, using the
keys as spatial indices, and the values to compute colors"""
if ax is None:
ax = pyplot.gca()
if isrdd(data):
pts = Colorize(colormap, scale).calc(data).collect()
else:
raise Exception('input must be an RDD')
clrs = array(map(lambda (k, v): v, pts))
x = map(lambda (k, v): k[0], pts)
y = map(lambda (k, v): k[1], pts)
z = map(lambda (k, v): k[2], pts) # currently unused
h = ax.scatter(x, y, s=100, c=clrs, alpha=0.5, edgecolor='black', linewidth=0.2)
return ax, h
def predict(self, data):
"""Predict the cluster that all data points belong to, and the similarity
Parameters
----------
data : RDD of (tuple, array) pairs, a list of arrays, or a single array
The data to predict cluster assignments on
Returns
-------
closest : RDD of (tuple, array) pairs, list of arrays, or a single array
For each data point, gives an array with the closest center for each data point,
and the correlation with that center
"""
if isrdd(data):
return data.mapValues(lambda x: KMeans.similarity(x, self.centers))
elif type(data) is list:
return map(lambda x: KMeans.similarity(x, self.centers), data)
else:
return KMeans.similarity(data, self.centers)
def predict(self, data):
"""Predict the cluster that all data points belong to, and the similarity
Parameters
----------
data : RDD of (tuple, array) pairs, a list of arrays, or a single array
The data to predict cluster assignments on
Returns
-------
closest : RDD of (tuple, array) pairs, list of arrays, or a single array
For each data point, gives an array with the closest center for each data point,
and the correlation with that center
"""
if isrdd(data):
return data.mapValues(lambda x: KMeans.similarity(x, self.centers))
elif type(data) is list:
return map(lambda x: KMeans.similarity(x, self.centers), data)
else:
return KMeans.similarity(data, self.centers)
def imagemap(data, colormap='polar', scale=1.0, ax=None):
"""Create an image from a collection of key-value pairs, using the
keys as spatial indices, and the values to compute colors"""
if ax is None:
ax = pyplot.gca()
if isrdd(data):
ndim = len(data.first()[0])
data = Colorize(colormap, scale).calc(data)
if ndim == 3:
pixels = pack(data, axes=2)
elif ndim == 2:
pixels = pack(data)
else:
raise Exception('number of spatial dimensions for images must be 2 or 3')
else:
raise Exception('input must be an RDD')
h = ax.imshow(transpose(pixels, [2, 1, 0]))
return ax, h
def imagemap(data, colormap='polar', scale=1.0, ax=None):
"""Create an image from a collection of key-value pairs, using the
keys as spatial indices, and the values to compute colors"""
if ax is None:
ax = pyplot.gca()
if isrdd(data):
ndim = len(data.first()[0])
data = Colorize(colormap, scale).calc(data)
if ndim == 3:
pixels = pack(data, axes=2)
elif ndim == 2:
pixels = pack(data)
else:
raise Exception(
'number of spatial dimensions for images must be 2 or 3')
else:
raise Exception('input must be an RDD')
h = ax.imshow(transpose(pixels, [2, 1, 0]))
return ax, h
def pointmap(data, colormap='polar', scale=1.0, ax=None):
"""Create a spatial point map from a collection of key-value pairs, using the
keys as spatial indices, and the values to compute colors"""
if ax is None:
ax = pyplot.gca()
if isrdd(data):
pts = Colorize(colormap, scale).calc(data).collect()
else:
raise Exception('input must be an RDD')
clrs = array(map(lambda (k, v): v, pts))
x = map(lambda (k, v): k[0], pts)
y = map(lambda (k, v): k[1], pts)
z = map(lambda (k, v): k[2], pts) # currently unused
h = ax.scatter(x,
y,
s=100,
c=clrs,
alpha=0.5,
edgecolor='black',
linewidth=0.2)
return ax, h
def save(data,
outputdir,
outputfile,
outputformat,
sorting=False,
dimsmax=None,
dimsmin=None):
"""
Save data to a variety of formats
Automatically determines whether data is an array
or an RDD and handle appropriately
Parameters
----------
data : RDD of (tuple, array) pairs, or numpy array
The data to save
outputdir : str
Output directory
outputfile : str
Output filename
outputformat : str
Output format ("matlab", "text", or "image")
"""
if not os.path.exists(outputdir):
os.makedirs(outputdir)
filename = os.path.join(outputdir, outputfile)
if isrdd(data):
nout = size(data.first()[1])
if dimsmax is not None:
dims = Dimensions()
dims.max = dimsmax
if dimsmin is not None:
dims.min = dimsmin
else:
dims.min = (1, 1, 1)
elif dimsmin is not None:
raise Exception('cannot provide dimsmin without dimsmax')
else:
dims = getdims(data)
if (outputformat == "matlab") | (outputformat == "text"):
if isrdd(data):
if nout > 1:
for iout in range(0, nout):
result = pack(data, ind=iout, dims=dims, sorting=sorting)
if outputformat == "matlab":
savemat(filename + "-" + str(iout) + ".mat",
mdict={outputfile + str(iout): result},
oned_as='column',
do_compression='true')
if outputformat == "text":
savetxt(filename + "-" + str(iout) + ".txt",
result,
fmt="%.6f")
else:
result = pack(data, dims=dims, sorting=sorting)
if outputformat == "matlab":
savemat(filename + ".mat",
mdict={outputfile: result},
oned_as='column',
do_compression='true')
if outputformat == "text":
savetxt(filename + ".txt", result, fmt="%.6f")
else:
if outputformat == "matlab":
savemat(filename + ".mat",
mdict={outputfile: data},
oned_as='column',
do_compression='true')
if outputformat == "text":
savetxt(filename + ".txt", data, fmt="%.6f")
if outputformat == "image":
if isrdd(data):
data = rescale(data)
if nout > 1:
for iout in range(0, nout):
result = pack(data, ind=iout, dims=dims, sorting=sorting)
arraytoim(result, filename + "-" + str(iout))
else:
result = pack(data, dims=dims, sorting=sorting)
arraytoim(result, filename)
else:
arraytoim(data, filename)
def save(data, outputdir, outputfile, outputformat, sorting=False, dimsmax=None, dimsmin=None):
"""
Save data to a variety of formats
Automatically determines whether data is an array
or an RDD and handle appropriately
Parameters
----------
data : RDD of (tuple, array) pairs, or numpy array
The data to save
outputdir : str
Output directory
outputfile : str
Output filename
outputformat : str
Output format ("matlab", "text", or "image")
"""
if not os.path.exists(outputdir):
os.makedirs(outputdir)
filename = os.path.join(outputdir, outputfile)
if isrdd(data):
nout = size(data.first()[1])
if dimsmax is not None:
dims = Dimensions()
dims.max = dimsmax
if dimsmin is not None:
dims.min = dimsmin
else:
dims.min = (1, 1, 1)
elif dimsmin is not None:
raise Exception('cannot provide dimsmin without dimsmax')
else:
dims = getdims(data)
if (outputformat == "matlab") | (outputformat == "text"):
if isrdd(data):
if nout > 1:
for iout in range(0, nout):
result = pack(data, ind=iout, dims=dims, sorting=sorting)
if outputformat == "matlab":
savemat(filename+"-"+str(iout)+".mat", mdict={outputfile+str(iout): result},
oned_as='column', do_compression='true')
if outputformat == "text":
savetxt(filename+"-"+str(iout)+".txt", result, fmt="%.6f")
else:
result = pack(data, dims=dims, sorting=sorting)
if outputformat == "matlab":
savemat(filename+".mat", mdict={outputfile: result},
oned_as='column', do_compression='true')
if outputformat == "text":
savetxt(filename+".txt", result, fmt="%.6f")
else:
if outputformat == "matlab":
savemat(filename+".mat", mdict={outputfile: data}, oned_as='column', do_compression='true')
if outputformat == "text":
savetxt(filename+".txt", data, fmt="%.6f")
if outputformat == "image":
if isrdd(data):
data = rescale(data)
if nout > 1:
for iout in range(0, nout):
result = pack(data, ind=iout, dims=dims, sorting=sorting)
arraytoim(result, filename+"-"+str(iout))
else:
result = pack(data, dims=dims, sorting=sorting)
arraytoim(result, filename)
else:
arraytoim(data, filename)
