def plotcoast(fn, pxlim, pylim, col='k'):
coasts = tabarray.tabarray(fn, comchar='s') # Read two columns from file
for segment in coasts.segments:
coastseg = coasts[segment].T
xw = coastseg[1]
yw = coastseg[0] # First one appears to be Latitude
xs = xw
ys = yw # Reset lists which store valid pos.
if 1:
# Mask arrays if outside plot box
xp, yp = annim.projection.topixel(
(numpy.array(xs), numpy.array(ys)))
xp = numpy.ma.masked_where(
numpy.isnan(xp) | (xp > pxlim[1]) | (xp < pxlim[0]), xp)
yp = numpy.ma.masked_where(
numpy.isnan(yp) | (yp > pylim[1]) | (yp < pylim[0]), yp)
# Mask array could be of type numpy.bool_ instead of numpy.ndarray
if numpy.isscalar(xp.mask):
xp.mask = numpy.array(xp.mask, 'bool')
if numpy.isscalar(yp.mask):
yp.mask = numpy.array(yp.mask, 'bool')
# Count the number of positions in these list that are inside the box
j = 0
for i in range(len(xp)):
if not xp.mask[i] and not yp.mask[i]:
j += 1
if j > 200: # Threshold to prevent too much detail and big pdf's
frame.plot(xp.data, yp.data, color=col)
def plotcoast(fn, pxlim, pylim, col='k'):
coasts = tabarray.tabarray(fn, comchar='s') # Read two columns from file
for segment in coasts.segments:
coastseg = coasts[segment].T
xw = coastseg[1]; yw = coastseg[0] # First one appears to be Latitude
xs = xw; ys = yw # Reset lists which store valid pos.
if 1:
# Mask arrays if outside plot box
xp, yp = annim.projection.topixel((numpy.array(xs),numpy.array(ys)))
xp = numpy.ma.masked_where(numpy.isnan(xp) |
(xp > pxlim[1]) | (xp < pxlim[0]), xp)
yp = numpy.ma.masked_where(numpy.isnan(yp) |
(yp > pylim[1]) | (yp < pylim[0]), yp)
# Mask array could be of type numpy.bool_ instead of numpy.ndarray
if numpy.isscalar(xp.mask):
xp.mask = numpy.array(xp.mask, 'bool')
if numpy.isscalar(yp.mask):
yp.mask = numpy.array(yp.mask, 'bool')
# Count the number of positions in these list that are inside the box
j = 0
for i in range(len(xp)):
if not xp.mask[i] and not yp.mask[i]:
j += 1
if j > 200: # Threshold to prevent too much detail and big pdf's
frame.plot(xp.data, yp.data, color=col)
def plotcoast(fn, frame, grat, col='k', lim=100, decim=5, plotsym=None, sign=1.0):
coasts = tabarray.tabarray(fn, comchar='s') # Read two columns from file
for segment in coasts.segments:
coastseg = coasts[segment].T
xw = sign * coastseg[1]; yw = coastseg[0] # First one appears to be Latitude
xs = xw; ys = yw # Reset lists which store valid pos.
if 1:
# Mask arrays if outside plot box
xp, yp = grat.gmap.topixel((numpy.array(xs),numpy.array(ys)))
# Be sure you understand
# the operator precedence: (a > 2) | (a < 5) is the proper syntax
# because a > 2 | a < 5 will result in an error due to the fact
# that 2 | a is evaluated first.
xp = numpy.ma.masked_where(numpy.isnan(xp) |
(xp > grat.pxlim[1]) | (xp < grat.pxlim[0]), xp)
yp = numpy.ma.masked_where(numpy.isnan(yp) |
(yp > grat.pylim[1]) | (yp < grat.pylim[0]), yp)
# Mask array could be of type numpy.bool_ instead of numpy.ndarray
if numpy.isscalar(xp.mask):
xp.mask = numpy.array(xp.mask, 'bool')
if numpy.isscalar(yp.mask):
yp.mask = numpy.array(yp.mask, 'bool')
# Count the number of positions in this list that are inside the box
xdc = []; ydc = []
for i in range(len(xp)):
if not xp.mask[i] and not yp.mask[i]:
if not i%decim:
xdc.append(xp.data[i])
ydc.append(yp.data[i])
if len(xdc) >= lim:
if plotsym == None:
frame.plot(xdc, ydc, color=col)
else:
frame.plot(xdc, ydc, '.', markersize=1, color=col)
def plotcoast(fn, frame, grat, col='k', lim=100, decim=5, plotsym=None, sign=1.0):
coasts = tabarray.tabarray(fn, comchar='s') # Read two columns from file
for segment in coasts.segments:
coastseg = coasts[segment].T
xw = sign * coastseg[1]; yw = coastseg[0] # First one appears to be Latitude
xs = xw; ys = yw # Reset lists which store valid pos.
if 1:
# Mask arrays if outside plot box
xp, yp = grat.gmap.topixel((numpy.array(xs),numpy.array(ys)))
# Be sure you understand
# the operator precedence: (a > 2) | (a < 5) is the proper syntax
# because a > 2 | a < 5 will result in an error due to the fact
# that 2 | a is evaluated first.
xp = numpy.ma.masked_where(numpy.isnan(xp) |
(xp > grat.pxlim[1]) | (xp < grat.pxlim[0]), xp)
yp = numpy.ma.masked_where(numpy.isnan(yp) |
(yp > grat.pylim[1]) | (yp < grat.pylim[0]), yp)
# Mask array could be of type numpy.bool_ instead of numpy.ndarray
if numpy.isscalar(xp.mask):
xp.mask = numpy.array(xp.mask, 'bool')
if numpy.isscalar(yp.mask):
yp.mask = numpy.array(yp.mask, 'bool')
# Count the number of positions in this list that are inside the box
xdc = []; ydc = []
for i in range(len(xp)):
if not xp.mask[i] and not yp.mask[i]:
if not i%decim:
xdc.append(xp.data[i])
ydc.append(yp.data[i])
if len(xdc) >= lim:
if plotsym == None:
frame.plot(xdc, ydc, color=col)
else:
frame.plot(xdc, ydc, '.', markersize=1, color=col)
def set_source(self, source):
"""
Define an alternative source for the colormap.
*source* can be any other matplotlib colormap object or its registered
name, a NumPy array with one RGB triplet per row or the name of a textfile
with one RGB triplet per line. Values should be between 0.0 and 1.0.
"""
self.source = source
try:
source = cm.get_cmap(source)
if source is None:
source = self.source # restore source
except:
pass
if isinstance(source, Colormap):
if not source._isinit:
source._init()
self.baselut = source._lut
self.N = source.N
elif isinstance(source, numpy.ndarray):
ncolors = source.shape[0]
self.baselut = numpy.ones((ncolors+3,4), numpy.float)
self.baselut[:ncolors,:3] = source
self.N = ncolors # may have changed
else:
try:
colors = tabarray(source)
except:
colors = tabarray(package_dir + '/lut/' + source)
ncolors = colors.shape[0]
self.baselut = numpy.ones((ncolors+3,4), numpy.float)
self.baselut[:ncolors,:3] = colors
self.N = ncolors # may have changed
self._i_under = self.N
self._i_over = self.N+1
self._i_bad = self.N+2
self.worklut = self.baselut.copy()
self._lut = self.worklut.copy() # existing may be inadequate
if self.bad_set:
badcolor, badalpha = self.bad_val
self.set_bad(badcolor, badalpha) # restore bad
if self.name is not None: # existing VariableColormap object?
self.set_scale(self.scale)
yr = xt * sinP + yt * cosP
return (xr / major)**2 + (yr / minor)**2
def residuals(p, data):
"""
Note that the function calculates the height z of the 3d landscape
of the function z = (x/maj)**2 + (y/min)**2.
An ellipse is defined where z=1. The residuals we use
for the least squares fit to minimize, is then 1-z
"""
x, y = data
return 1.0 - func(p, data)
x, y = tabarray('ellipse.dat').columns() # Read from file
x += numpy.random.normal(0.0, 0.05, len(x))
y += numpy.random.normal(0.0, 0.05, len(y))
# This ellipse data was made using a function. The parameters
# that were used are:
# xc = 5.0; yc = 4.0; major = 10.0; minor = 3.0; ang = 60.0
# Compare these values with the result of the esimate function.
p0 = getestimates(x, y)
fitter = kmpfit.Fitter(residuals=residuals, data=(x, y))
fitter.fit(params0=p0)
print("\n========= Fit results ellipse model ==========")
print("Initial params:", fitter.params0)
print("Params: ", fitter.params)
print("Iterations: ", fitter.niter)
print("Function ev: ", fitter.nfev)
yt = y - y0
xr = xt * cosP - yt * sinP
yr = xt * sinP + yt * cosP
return (xr/major)**2 + (yr/minor)**2
def residuals(p, data):
"""
Note that the function calculates the height z of the 3d landscape
of the function z = (x/maj)**2 + (y/min)**2.
An ellipse is defined where z=1. The residuals we use
for the least squares fit to minimize, is then 1-z
"""
x, y = data
return 1.0 - func(p,data)
x,y = tabarray('ellipse.dat').columns() # Read from file
x += numpy.random.normal(0.0, 0.05, len(x))
y += numpy.random.normal(0.0, 0.05, len(y))
# This ellipse data was made using a function. The parameters
# that were used are:
# xc = 5.0; yc = 4.0; major = 10.0; minor = 3.0; ang = 60.0
# Compare these values with the result of the esimate function.
p0 = getestimates(x, y)
fitter = kmpfit.Fitter(residuals=residuals, data=(x,y))
fitter.fit(params0=p0)
print("\n========= Fit results ellipse model ==========")
print("Initial params:", fitter.params0)
print("Params: ", fitter.params)
print("Iterations: ", fitter.niter)
print("Function ev: ", fitter.nfev)