def plot2d(metaAry, size=(10, 7.5), dpi=75, fontsize=15, cmap=None, \
nticks=5, aspect_ratio=1.0, corient='vertical', cformat=None, \
show_cbar=True, vmin=None, vmax=None, interpolation='sinc', \
fig=None, ax=None):
"""
metaArray function to do a simple 2D plot.
size Plot size, default to (10, 7.5)
dpi Dot Per Inch for raster graphics
fontsize Norminal font size
cmap Colour map, default is pyplot.cm.hot
nticks Number of ticks in the colour bar
aspect_ratio Aspect ratio of the plot {float|'ij'|'xy'}
float: Fixed aspect ratio by the given number
'ij': Same aspect ratio as ij space
'xy': Same aspect ratio as xy space
corient Colorbar orientation ('vertical'|'horizontal')
cformat Colorbar format [ None | format string | Formatter object ]
vmin Minimum value for the colour scale
vmax Maximum value for the coloir scale
interpolation Colour interpolation methods
[None, 'none', 'nearest', 'bilinear', 'bicubic',
'spline16', 'spline36', 'hanning', 'hamming', 'hermite',
'kaiser', 'quadric', 'catrom', 'gaussian', 'bessel',
'mitchell', 'sinc', 'lanczos']
"""
if cmap is None:
#cmap = cm.spectral
try:
cmap = cm.viridis
except AttributeError:
cmap = cm.hot
if corient is not 'horizontal':
corient = 'vertical'
axis = metaAry['range']
data = metaAry.data
x0 = axis['begin'][0]
x1 = axis['end'][0]
y0 = axis['begin'][1]
y1 = axis['end'][1]
# Try to work out the aspect ratio and plot size before scailing the axis
# Aspect ratio of the plot
if aspect_ratio == 'ij':
ratio = data.shape
ratio = float(ratio[1]) / ratio[0]
elif aspect_ratio == 'xy':
ratio = float(y1 - y0) / (float(x1 - x0))
else:
try:
ratio = float(aspect_ratio)
except:
print(
"*** Warning! Unrecognisable aspect ratio spec. Using the default instead."
)
ratio = 1.0
## Make plot with vertical (default) colorbar
#if size == 'default':
## Default size aspect ratio is 4:3, and size to be (10, 7.5)
## Adjust size according to image aspect ratio
#if corient == 'vertical':
#size_ratio = ratio * 0.75
#size = (10, 10 * size_ratio)
#else:
#size_ratio = ratio / 0.75
#size = (7.5 / size_ratio, 7.5)
##diagonal = 12.5**2
##size_x = (diagonal / (1 + size_ratio**2))**0.5
##size_y = (diagonal / size_x**2)**0.5
##size = (size_x, size_y)
# Try to work out the colour scale
if vmin is None:
v0 = metaAry.data.min()
else:
v0 = vmin
if vmax is None:
v1 = metaAry.data.max()
else:
v1 = vmax
# In case the values are all the same
v0, v1 = scale_check(v0, v1)
xunit = axis['unit'][0]
yunit = axis['unit'][1]
vunit = metaAry['unit']
# Apply unit prefix if unit is defined
xunit, x0, x1, xscale = prettyunit(xunit, x0, x1)
yunit, y0, y1, yscale = prettyunit(yunit, y0, y1)
vunit, v0, v1, vscale = prettyunit(vunit, v0, v1)
if vscale != 1:
data = data.copy() * vscale
xlabl = lbl_repr(axis['label'][0], xunit)
ylabl = lbl_repr(axis['label'][1], yunit)
vlabl = lbl_repr(metaAry['label'], vunit)
ticks = np.linspace(v0, v1, nticks)
ticks_lbl = []
# Matplotlib inshow data in transposed from metaArray convension
# And it adjust the aspect ratio based on the prefix corrected number
ratio /= float(y1 - y0) / float(x1 - x0)
ratio = float(np.abs(ratio)) # This is the number fed to matplotlib
for i in range(nticks):
ticks_lbl.append("%(val)0.4g" % {'val': ticks[i]})
if fig == None: fig = figure(figsize=size, dpi=dpi)
if ax == None: ax = fig.add_subplot(111)
extent = (x0, x1, y0, y1)
cax = ax.imshow(data.transpose()[::-1], cmap=cmap, extent=extent, \
interpolation=interpolation, vmin=v0, vmax=v1, aspect=ratio)
if show_cbar:
cbar = fig.colorbar(cax,
ticks=ticks,
orientation=corient,
format=cformat)
# Add colorbar, make sure to specify tick locations to match desired ticklabels
cbar.ax.set_yticklabels(ticks_lbl)
cbar.set_label(vlabl, fontsize=fontsize)
# ax.set_size(fontsize)
ax.set_xlabel(xlabl, fontsize=fontsize) # Label font size
ax.set_ylabel(ylabl, fontsize=fontsize)
rcParams.update({'font.size': fontsize}) # Value font size
if fontsize is not None:
ax.set_title(metaAry['name'], fontsize=int(fontsize * 1.3))
else:
ax.set_title(metaAry['name'])
# fig.tight_layout()
return fig, ax
def multiplot(metaAry, size=(10, 7.5), dpi=75, grid=True, \
legend=0, fontsize=15, real_label=None, imag_label=None, \
fig=None, host=None, par=None):
"""
metaArray function to do a simple 1D plot of complex array as real and imaginary parts.
legend:
'best' 0
'upper right' 1
'upper left' 2
'lower left' 3
'lower right' 4
'right' 5
'center left' 6
'center right' 7
'lower center' 8
'upper center' 9
'center' 10
"""
if legend is None:
legend = 0
if real_label is None:
real_label = "Real"
if imag_label is None:
imag_label = "Imaginary"
axis = metaAry['range']
rdata = metaAry.data.real
idata = metaAry.data.imag
# Load the plotting ranges and units
x0 = axis['begin'][0]
x1 = axis['end'][0]
ry0 = min(rdata)
ry1 = max(rdata)
iy0 = min(idata)
iy1 = max(idata)
xunit = axis['unit'][0]
ryunit = metaAry['unit']
iyunit = metaAry['unit']
# Leave 10% margin in the y axis
rmean = np.average((ry0, ry1))
rreach = np.abs(ry0 - ry1) / 2 / 0.9
ry0 = np.sign(ry0 - rmean) * rreach + rmean
ry1 = np.sign(ry1 - rmean) * rreach + rmean
imean = np.average((iy0, iy1))
ireach = np.abs(iy0 - iy1) / 2 / 0.9
iy0 = np.sign(iy0 - imean) * ireach + imean
iy1 = np.sign(iy1 - imean) * ireach + imean
# Apply unit prefix if unit is defined
xunit, x0, x1, xscale = prettyunit(xunit, x0, x1)
ryunit, ry0, ry1, ryscale = prettyunit(ryunit, ry0, ry1)
iyunit, iy0, iy1, iyscale = prettyunit(iyunit, iy0, iy1)
if ryscale != 1:
rdata = rdata.copy() * ryscale
if iyscale != 1:
idata = idata.copy() * iyscale
xlabl = lbl_repr(axis['label'][0], xunit)
rylabl = lbl_repr(metaAry['label'], ryunit, real_label + ' part')
iylabl = lbl_repr(metaAry['label'], iyunit, imag_label + ' part')
title = metaAry['name']
fig = figure(figsize=size, dpi=dpi)
host = SubplotHost(fig, 111)
fig.add_subplot(host)
par = host.twinx()
#if axis['log'][0] == False:
# x = linspace(x0, x1, len(metaAry))
#else:
# raise NotImplemented, "Log axis is not yet implemented."
x = metaAry.get_axis()
host.plot(x, rdata, 'b-', label=lbl_repr(axis['label'][0], '', real_label))
par.plot(x, idata, 'r--', label=lbl_repr(axis['label'][0], '', real_label))
host.grid(grid)
host.set_xlabel(xlabl, fontsize=fontsize)
host.set_ylabel(rylabl, fontsize=fontsize)
par.set_ylabel(iylabl, fontsize=fontsize)
host.set_xlim([x0, x1])
host.set_ylim([ry0, ry1])
par.set_ylim([iy0, iy1])
if fontsize is not None:
host.set_title(title, fontsize=int(fontsize * 1.3))
else:
host.set_title(title)
if legend >= 0:
host.legend(loc=legend)
return fig, host, par
def plot1d(metaAry, size=(10, 7.5), dpi=75, grid=True, legend=None, fontsize=15,\
fig=None, ax=None, label=None):
"""
metaArray function to do a simple 1D plot.
legend:
'best' 0
'upper right' 1
'upper left' 2
'lower left' 3
'lower right' 4
'right' 5
'center left' 6
'center right' 7
'lower center' 8
'upper center' 9
'center' 10
label Label for the legend display, default to metaAry['range']['label'][0]
"""
if metaAry.dtype is np.dtype('complex'):
return plotcomplex(metaAry,
size=size,
dpi=dpi,
grid=grid,
legend=legend,
fontsize=fontsize)
if legend is None:
legend = -1
axis = metaAry['range']
data = metaAry.data
# Load the plotting ranges and units
x0 = axis['begin'][0]
x1 = axis['end'][0]
y0 = min(metaAry.data)
y1 = max(metaAry.data)
xunit = axis['unit'][0]
yunit = metaAry['unit']
# Leave 10% margin in the y axis
mean = np.average((y0, y1))
reach = np.abs(y0 - y1) / 2 / 0.9
y0 = np.sign(y0 - mean) * reach + mean
y1 = np.sign(y1 - mean) * reach + mean
y0, y1 = scale_check(y0, y1)
# Apply unit prefix if unit is defined
xunit, x0, x1, xscale = prettyunit(xunit, x0, x1)
yunit, y0, y1, yscale = prettyunit(yunit, y0, y1)
if yscale != 1:
data = data.copy() * yscale
xlabl = lbl_repr(axis['label'][0], xunit)
ylabl = lbl_repr(metaAry['label'], yunit)
title = metaAry['name']
# check if object is 1D metaArray object
if fig is None:
fig = figure(figsize=size, dpi=dpi)
if ax is None:
ax = fig.add_subplot(111)
else:
x00, x01 = ax.get_xlim()
y00, y01 = ax.get_ylim()
x0 = min((x0, x00))
y0 = min((y0, y00))
x1 = max((x1, x01))
y1 = max((y1, y01))
if axis['log'][0] == False:
x = np.linspace(x0, x1, len(metaAry))
# x1 = metaAry.get_axis()
else:
x = metaAry.get_axis()
raise NotImplemented
if label is None:
label = axis['label'][0]
ax.plot(x, data, label=label)
ax.grid(grid)
ax.set_xlabel(xlabl, fontsize=fontsize)
ax.set_ylabel(ylabl, fontsize=fontsize)
ax.set_xlim([x0, x1])
ax.set_ylim([y0, y1])
if fontsize is not None:
ax.set_title(title, fontsize=int(fontsize * 1.3))
else:
ax.set_title(title)
if legend >= 0:
ax.legend(loc=legend)
return fig, ax
def plotcomplexpolar(metaAry, axis=-1, size=(10, 7.5), dpi=75, \
grid=True, legend=0, fontsize=14, linewidth=2.0, \
log_mag=False, mag_label=None, mag_unit=None, \
pha_label=None, pha_unit=None, degree=False):
"""
metaArray function to do a simple 1D plot of complex array (metaAry[axis]) as magnitude and phase angle.
legend:
'best' 0
'upper right' 1
'upper left' 2
'lower left' 3
'lower right' 4
'right' 5
'center left' 6
'center right' 7
'lower center' 8
'upper center' 9
'center' 10
"""
assert type(axis) is int, "Axis is not an integer: {!r}".format(axis)
fontsize = float(fontsize)
if legend is None:
legend = 0
if mag_label is None:
mag_label = "Magnitude"
if pha_label is None:
pha_label = "Phase"
mag = np.abs(metaAry.data)
pha = np.angle(metaAry.data)
# Load the plotting ranges and units
x0 = metaAry.get_range(axis, 'begin')
x1 = metaAry.get_range(axis, 'end')
my0 = min(mag)
my1 = max(mag)
py0 = np.floor(4 * min(pha) / np.pi) # Round to the nearest pi/4
py1 = np.ceil(4 * max(pha) / np.pi)
pticks = np.arange(py0, py1 + 1) # Ticks in pi/4 interval
xunit = metaAry.get_range(axis, 'unit')
if mag_unit == None:
myunit = metaAry['unit']
else:
myunit = str(mag_unit)
if pha_unit == None:
if degree == True:
pyunit = 'Deg.'
else:
pyunit = 'Rad.'
else:
pyunit = pha_unit
# Leave 10% margin in the y axis
if log_mag == True:
my0 = np.log10(my0)
my1 = np.log10(my1)
mmean = np.average((my0, my1))
mreach = np.abs(my0 - my1) / 2 / 0.9
my0 = np.sign(my0 - mmean) * mreach + mmean
my1 = np.sign(my1 - mmean) * mreach + mmean
if log_mag == True:
my0 = 10**my0
my1 = 10**my1
pmean = np.average((py0, py1))
preach = np.abs(py0 - py1) / 2 / 0.9
py0 = np.sign(py0 - pmean) * preach + pmean
py1 = np.sign(py1 - pmean) * preach + pmean
my0, my1 = scale_check(my0, my1)
py0, py1 = scale_check(py0, py1)
# Apply unit prefix if unit is defined
xunit, x0, x1, xscale = prettyunit(xunit, x0, x1)
myunit, my0, my1, myscale = prettyunit(myunit, my0, my1)
pyunit, py0, py1, pyscale = prettyunit(pyunit, py0, py1)
x = metaAry.get_axis(axis=axis)
if xscale != 1:
x *= xscale
if myscale != 1:
mag *= myscale
if pyscale != 1:
pha *= pyscale
xlabl = lbl_repr(metaAry.get_range(axis, 'label'), xunit)
mylabl = lbl_repr(metaAry['label'], myunit, mag_label)
pylabl = lbl_repr(metaAry['label'], pyunit, pha_label)
title = metaAry['name']
# Done the preparation, do the actual plotting
fig, ax1 = plt.subplots(figsize=size, dpi=dpi)
ax2 = ax1.twinx()
ax1.grid(True, which="both", ls="--", color='g')
######
ax1.plot(x, mag, 'b-', linewidth=linewidth, label=mag_label)
if degree == True:
ax2.plot(x,
pha * 180 / np.pi,
'r--',
linewidth=linewidth,
label=pha_label)
else:
ax2.plot(x, pha, 'r--', linewidth=linewidth, label=pha_label)
######
ax1.set_ylabel(mylabl, color='b', fontsize=fontsize)
for tl in ax1.get_yticklabels():
tl.set_color('b')
tl.set_fontsize(fontsize)
if log_mag == True:
ax1.set_yscale('log', nonposy='clip')
ax1.set_ylim([my0, my1])
######
ax2.set_ylabel(pylabl, color='r', fontsize=fontsize)
if degree == True:
py0 *= 45
py1 *= 45
ax2.set_ylim([py0, py1])
ax2.set_yticks(pticks * 45)
else:
py0 = py0 * np.pi / 4
py1 = py1 * np.pi / 4
ax2.set_ylim([py0, py1])
ax2.set_yticks(pticks * np.pi / 4)
pticks_lbl = []
for pt in pticks:
if np.abs(pt) == 1:
val = r'\pi/4'
elif np.abs(pt) == 2:
val = r'\pi/2'
elif np.abs(pt) == 3:
val = r'3\pi/4'
elif np.abs(pt) == 4:
val = r'\pi'
if np.abs(pt) == 0:
val = '0'
elif np.sign(pt) == -1:
val = '-' + val
pticks_lbl.append('$' + val + '$')
ax2.set_yticklabels(pticks_lbl)
for tl in ax2.get_yticklabels():
tl.set_color('r')
tl.set_fontsize(fontsize)
######
ax1.set_title(title, fontsize=fontsize * 1.3)
if metaAry.get_range(axis, 'log') == True:
ax1.set_xscale("log", nonposx='clip')
ax2.set_xscale("log", nonposx='clip')
for tl in ax1.get_xticklabels():
tl.set_fontsize(fontsize)
ax1.set_xlabel(xlabl, fontsize=fontsize)
ax1.set_xlim([x0, x1])
if legend >= 0:
lns1, lbl1 = ax1.get_legend_handles_labels()
lns2, lbl2 = ax2.get_legend_handles_labels()
ax1.legend(lns1 + lns2, lbl1 + lbl2, loc=legend)
######
## fig = figure(figsize=size, dpi = dpi)
## host = SubplotHost(fig, 111)
#fig.add_subplot(host)
#par = host.twinx()
#host.plot(x, mag, 'b-', label=mag_label)
#par.plot(x, pha, 'r--', label=pha_label)
#host.grid(grid, which="both", color='g')
#host.set_xlabel(xlabl)
#host.set_ylabel(mylabl, color='b')
#par.set_ylabel(pylabl, color='r')
#if log_mag == True:
#host.set_yscale('log', nonposy='clip')
#host.set_xlim([x0, x1])
#host.set_ylim([my0, my1])
#par.set_ylim([py0, py1])
#for tl in host.get_xticklabels():
#tl.set_fontsize(fontsize)
#for tl in host.get_yticklabels():
#tl.set_color('b')
#tl.set_fontsize(fontsize)
#for tl in par.get_yticklabels():
#tl.set_color('r')
#tl.set_fontsize(fontsize)
## host.set_title(title, fontsize=int(fontsize*1.3))
#host.set_title(title, fontsize=fontsize)
#if legend >= 0:
#host.legend(loc=legend)
return fig, ax1, ax2
def plot2d(metaAry, size=(10, 7.5), dpi=75, fontsize=15, cmap=None, \
nticks=5, aspect_ratio=1.0, corient='vertical', cformat=None, \
show_cbar=True, vmin=None, vmax=None, interpolation='sinc', \
fig=None, ax=None):
"""
metaArray function to do a simple 2D plot.
size Plot size, default to (10, 7.5)
dpi Dot Per Inch for raster graphics
fontsize Norminal font size
cmap Colour map, default is pyplot.cm.hot
nticks Number of ticks in the colour bar
aspect_ratio Aspect ratio of the plot {float|'ij'|'xy'}
float: Fixed aspect ratio by the given number
'ij': Same aspect ratio as ij space
'xy': Same aspect ratio as xy space
corient Colorbar orientation ('vertical'|'horizontal')
cformat Colorbar format [ None | format string | Formatter object ]
vmin Minimum value for the colour scale
vmax Maximum value for the coloir scale
interpolation Colour interpolation methods
[None, 'none', 'nearest', 'bilinear', 'bicubic',
'spline16', 'spline36', 'hanning', 'hamming', 'hermite',
'kaiser', 'quadric', 'catrom', 'gaussian', 'bessel',
'mitchell', 'sinc', 'lanczos']
"""
if cmap is None:
#cmap = cm.spectral
try:
cmap = cm.viridis
except AttributeError:
cmap = cm.hot
if corient is not 'horizontal':
corient = 'vertical'
axis = metaAry['range']
data = metaAry.data
x0 = axis['begin'][0]
x1 = axis['end'][0]
y0 = axis['begin'][1]
y1 = axis['end'][1]
# Try to work out the aspect ratio and plot size before scailing the axis
# Aspect ratio of the plot
if aspect_ratio == 'ij':
ratio = data.shape
ratio = float(ratio[1]) / ratio[0]
elif aspect_ratio == 'xy':
ratio = float(y1 - y0) / (float(x1 - x0))
else:
try:
ratio = float(aspect_ratio)
except:
print("*** Warning! Unrecognisable aspect ratio spec. Using the default instead.")
ratio = 1.0
## Make plot with vertical (default) colorbar
#if size == 'default':
## Default size aspect ratio is 4:3, and size to be (10, 7.5)
## Adjust size according to image aspect ratio
#if corient == 'vertical':
#size_ratio = ratio * 0.75
#size = (10, 10 * size_ratio)
#else:
#size_ratio = ratio / 0.75
#size = (7.5 / size_ratio, 7.5)
##diagonal = 12.5**2
##size_x = (diagonal / (1 + size_ratio**2))**0.5
##size_y = (diagonal / size_x**2)**0.5
##size = (size_x, size_y)
# Try to work out the colour scale
if vmin is None:
v0 = metaAry.data.min()
else:
v0 = vmin
if vmax is None:
v1 = metaAry.data.max()
else:
v1 = vmax
# In case the values are all the same
v0, v1 = scale_check(v0, v1)
xunit = axis['unit'][0]
yunit = axis['unit'][1]
vunit = metaAry['unit']
# Apply unit prefix if unit is defined
xunit, x0, x1, xscale = prettyunit(xunit, x0, x1)
yunit, y0, y1, yscale = prettyunit(yunit, y0, y1)
vunit, v0, v1, vscale = prettyunit(vunit, v0, v1)
if vscale != 1:
data = data.copy() * vscale
xlabl = lbl_repr(axis['label'][0], xunit)
ylabl = lbl_repr(axis['label'][1], yunit)
vlabl = lbl_repr(metaAry['label'], vunit)
ticks = np.linspace(v0, v1, nticks)
ticks_lbl = []
# Matplotlib inshow data in transposed from metaArray convension
# And it adjust the aspect ratio based on the prefix corrected number
ratio /= float(y1 - y0) / float(x1 - x0)
ratio = float(np.abs(ratio)) # This is the number fed to matplotlib
for i in range(nticks):
ticks_lbl.append("%(val)0.4g" % {'val':ticks[i]})
if fig == None: fig = figure(figsize=size, dpi=dpi)
if ax == None: ax = fig.add_subplot(111)
extent = (x0, x1, y0, y1)
cax = ax.imshow(data.transpose()[::-1], cmap=cmap, extent=extent, \
interpolation=interpolation, vmin=v0, vmax=v1, aspect=ratio)
if show_cbar:
cbar = fig.colorbar(cax, ticks=ticks, orientation=corient, format=cformat)
# Add colorbar, make sure to specify tick locations to match desired ticklabels
cbar.ax.set_yticklabels(ticks_lbl)
cbar.set_label(vlabl, fontsize=fontsize)
# ax.set_size(fontsize)
ax.set_xlabel(xlabl, fontsize=fontsize) # Label font size
ax.set_ylabel(ylabl, fontsize=fontsize)
rcParams.update({'font.size': fontsize}) # Value font size
if fontsize is not None:
ax.set_title(metaAry['name'], fontsize=int(fontsize*1.3))
else:
ax.set_title(metaAry['name'])
# fig.tight_layout()
return fig, ax
def plot1d(metaAry, size=(10, 7.5), dpi=75, grid=True, legend=None, fontsize=15,\
fig=None, ax=None, label=None):
"""
metaArray function to do a simple 1D plot.
legend:
'best' 0
'upper right' 1
'upper left' 2
'lower left' 3
'lower right' 4
'right' 5
'center left' 6
'center right' 7
'lower center' 8
'upper center' 9
'center' 10
label Label for the legend display, default to metaAry['range']['label'][0]
"""
if metaAry.dtype is np.dtype('complex'):
return plotcomplex(metaAry, size=size, dpi=dpi, grid=grid, legend=legend, fontsize=fontsize)
if legend is None:
legend = -1
axis = metaAry['range']
data = metaAry.data
# Load the plotting ranges and units
x0 = axis['begin'][0]
x1 = axis['end'][0]
y0 = min(metaAry.data)
y1 = max(metaAry.data)
xunit = axis['unit'][0]
yunit = metaAry['unit']
# Leave 10% margin in the y axis
mean = np.average((y0, y1))
reach = np.abs(y0-y1) / 2 / 0.9
y0 = np.sign(y0-mean) * reach + mean
y1 = np.sign(y1-mean) * reach + mean
y0, y1 = scale_check(y0, y1)
# Apply unit prefix if unit is defined
xunit, x0, x1, xscale = prettyunit(xunit, x0, x1)
yunit, y0, y1, yscale = prettyunit(yunit, y0, y1)
if yscale != 1:
data = data.copy() * yscale
xlabl = lbl_repr(axis['label'][0], xunit)
ylabl = lbl_repr(metaAry['label'], yunit)
title = metaAry['name']
# check if object is 1D metaArray object
if fig is None:
fig = figure(figsize=size, dpi=dpi)
if ax is None:
ax = fig.add_subplot(111)
else:
x00, x01 = ax.get_xlim()
y00, y01 = ax.get_ylim()
x0 = min((x0, x00))
y0 = min((y0, y00))
x1 = max((x1, x01))
y1 = max((y1, y01))
if axis['log'][0] == False:
x = np.linspace(x0, x1, len(metaAry))
# x1 = metaAry.get_axis()
else:
x = metaAry.get_axis()
raise NotImplemented
if label is None:
label = axis['label'][0]
ax.plot(x, data, label=label)
ax.grid(grid)
ax.set_xlabel(xlabl, fontsize=fontsize)
ax.set_ylabel(ylabl, fontsize=fontsize)
ax.set_xlim([x0, x1])
ax.set_ylim([y0, y1])
if fontsize is not None:
ax.set_title(title, fontsize=int(fontsize*1.3))
else:
ax.set_title(title)
if legend >= 0:
ax.legend(loc=legend)
return fig, ax
def multiplot(metaAry, size=(10, 7.5), dpi=75, grid=True, \
legend=0, fontsize=15, real_label=None, imag_label=None, \
fig=None, host=None, par=None):
"""
metaArray function to do a simple 1D plot of complex array as real and imaginary parts.
legend:
'best' 0
'upper right' 1
'upper left' 2
'lower left' 3
'lower right' 4
'right' 5
'center left' 6
'center right' 7
'lower center' 8
'upper center' 9
'center' 10
"""
if legend is None:
legend = 0
if real_label is None:
real_label = "Real"
if imag_label is None:
imag_label = "Imaginary"
axis = metaAry['range']
rdata = metaAry.data.real
idata = metaAry.data.imag
# Load the plotting ranges and units
x0 = axis['begin'][0]
x1 = axis['end'][0]
ry0 = min(rdata)
ry1 = max(rdata)
iy0 = min(idata)
iy1 = max(idata)
xunit = axis['unit'][0]
ryunit = metaAry['unit']
iyunit = metaAry['unit']
# Leave 10% margin in the y axis
rmean = np.average((ry0, ry1))
rreach = np.abs(ry0-ry1) / 2 / 0.9
ry0 = np.sign(ry0-rmean) * rreach + rmean
ry1 = np.sign(ry1-rmean) * rreach + rmean
imean = np.average((iy0, iy1))
ireach = np.abs(iy0-iy1) / 2 / 0.9
iy0 = np.sign(iy0-imean) * ireach + imean
iy1 = np.sign(iy1-imean) * ireach + imean
# Apply unit prefix if unit is defined
xunit, x0, x1, xscale = prettyunit(xunit, x0, x1)
ryunit, ry0, ry1, ryscale = prettyunit(ryunit, ry0, ry1)
iyunit, iy0, iy1, iyscale = prettyunit(iyunit, iy0, iy1)
if ryscale != 1:
rdata = rdata.copy() * ryscale
if iyscale != 1:
idata = idata.copy() * iyscale
xlabl = lbl_repr(axis['label'][0], xunit)
rylabl = lbl_repr(metaAry['label'], ryunit, real_label + ' part')
iylabl = lbl_repr(metaAry['label'], iyunit, imag_label + ' part')
title = metaAry['name']
fig = figure(figsize=size, dpi=dpi)
host = SubplotHost(fig, 111)
fig.add_subplot(host)
par = host.twinx()
#if axis['log'][0] == False:
# x = linspace(x0, x1, len(metaAry))
#else:
# raise NotImplemented, "Log axis is not yet implemented."
x = metaAry.get_axis()
host.plot(x, rdata, 'b-', label=lbl_repr(axis['label'][0], '', real_label))
par.plot(x, idata, 'r--', label=lbl_repr(axis['label'][0], '', real_label))
host.grid(grid)
host.set_xlabel(xlabl, fontsize=fontsize)
host.set_ylabel(rylabl, fontsize=fontsize)
par.set_ylabel(iylabl, fontsize=fontsize)
host.set_xlim([x0, x1])
host.set_ylim([ry0, ry1])
par.set_ylim([iy0, iy1])
if fontsize is not None:
host.set_title(title, fontsize=int(fontsize*1.3))
else:
host.set_title(title)
if legend >= 0:
host.legend(loc=legend)
return fig, host, par
def plotcomplexpolar(metaAry, axis=-1, size=(10, 7.5), dpi=75, \
grid=True, legend=0, fontsize=14, linewidth=2.0, \
log_mag=False, mag_label=None, mag_unit=None, \
pha_label=None, pha_unit=None, degree=False):
"""
metaArray function to do a simple 1D plot of complex array (metaAry[axis]) as magnitude and phase angle.
legend:
'best' 0
'upper right' 1
'upper left' 2
'lower left' 3
'lower right' 4
'right' 5
'center left' 6
'center right' 7
'lower center' 8
'upper center' 9
'center' 10
"""
assert type(axis) is int, "Axis is not an integer: {!r}".format(axis)
fontsize = float(fontsize)
if legend is None:
legend = 0
if mag_label is None:
mag_label = "Magnitude"
if pha_label is None:
pha_label = "Phase"
mag = np.abs(metaAry.data)
pha = np.angle(metaAry.data)
# Load the plotting ranges and units
x0 = metaAry.get_range(axis, 'begin')
x1 = metaAry.get_range(axis, 'end')
my0 = min(mag)
my1 = max(mag)
py0 = np.floor(4 * min(pha) / np.pi) # Round to the nearest pi/4
py1 = np.ceil(4 * max(pha) / np.pi)
pticks = np.arange(py0, py1+1) # Ticks in pi/4 interval
xunit = metaAry.get_range(axis, 'unit')
if mag_unit == None:
myunit = metaAry['unit']
else:
myunit = str(mag_unit)
if pha_unit == None:
if degree == True:
pyunit = 'Deg.'
else:
pyunit = 'Rad.'
else:
pyunit = pha_unit
# Leave 10% margin in the y axis
if log_mag == True:
my0 = np.log10(my0)
my1 = np.log10(my1)
mmean = np.average((my0, my1))
mreach = np.abs(my0-my1) / 2 / 0.9
my0 = np.sign(my0-mmean) * mreach + mmean
my1 = np.sign(my1-mmean) * mreach + mmean
if log_mag == True:
my0 = 10**my0
my1 = 10**my1
pmean = np.average((py0, py1))
preach = np.abs(py0-py1) / 2 / 0.9
py0 = np.sign(py0-pmean) * preach + pmean
py1 = np.sign(py1-pmean) * preach + pmean
my0, my1 = scale_check(my0, my1)
py0, py1 = scale_check(py0, py1)
# Apply unit prefix if unit is defined
xunit, x0, x1, xscale = prettyunit(xunit, x0, x1)
myunit, my0, my1, myscale = prettyunit(myunit, my0, my1)
pyunit, py0, py1, pyscale = prettyunit(pyunit, py0, py1)
x = metaAry.get_axis(axis=axis)
if xscale != 1:
x *= xscale
if myscale != 1:
mag *= myscale
if pyscale != 1:
pha *= pyscale
xlabl = lbl_repr(metaAry.get_range(axis, 'label'), xunit)
mylabl = lbl_repr(metaAry['label'], myunit, mag_label)
pylabl = lbl_repr(metaAry['label'], pyunit, pha_label)
title = metaAry['name']
# Done the preparation, do the actual plotting
fig, ax1 = plt.subplots(figsize=size, dpi=dpi)
ax2 = ax1.twinx()
ax1.grid(True, which="both", ls="--", color='g')
######
ax1.plot(x, mag, 'b-', linewidth=linewidth, label=mag_label)
if degree == True:
ax2.plot(x, pha * 180 / np.pi, 'r--', linewidth=linewidth, label=pha_label)
else:
ax2.plot(x, pha, 'r--', linewidth=linewidth, label=pha_label)
######
ax1.set_ylabel(mylabl, color='b', fontsize=fontsize)
for tl in ax1.get_yticklabels():
tl.set_color('b')
tl.set_fontsize(fontsize)
if log_mag == True:
ax1.set_yscale('log', nonposy='clip')
ax1.set_ylim([my0, my1])
######
ax2.set_ylabel(pylabl, color='r', fontsize=fontsize)
if degree == True:
py0 *= 45
py1 *= 45
ax2.set_ylim([py0, py1])
ax2.set_yticks(pticks*45)
else:
py0 = py0 * np.pi / 4
py1 = py1 * np.pi / 4
ax2.set_ylim([py0, py1])
ax2.set_yticks(pticks * np.pi / 4)
pticks_lbl = []
for pt in pticks:
if np.abs(pt) == 1:
val = r'\pi/4'
elif np.abs(pt) == 2:
val = r'\pi/2'
elif np.abs(pt) == 3:
val = r'3\pi/4'
elif np.abs(pt) == 4:
val = r'\pi'
if np.abs(pt) == 0:
val = '0'
elif np.sign(pt) == -1:
val = '-' + val
pticks_lbl.append('$' + val + '$')
ax2.set_yticklabels(pticks_lbl)
for tl in ax2.get_yticklabels():
tl.set_color('r')
tl.set_fontsize(fontsize)
######
ax1.set_title(title, fontsize=fontsize*1.3)
if metaAry.get_range(axis, 'log') == True:
ax1.set_xscale("log", nonposx='clip')
ax2.set_xscale("log", nonposx='clip')
for tl in ax1.get_xticklabels():
tl.set_fontsize(fontsize)
ax1.set_xlabel(xlabl, fontsize=fontsize)
ax1.set_xlim([x0, x1])
if legend >= 0:
lns1, lbl1 = ax1.get_legend_handles_labels()
lns2, lbl2 = ax2.get_legend_handles_labels()
ax1.legend(lns1 + lns2, lbl1 + lbl2, loc=legend)
######
## fig = figure(figsize=size, dpi = dpi)
## host = SubplotHost(fig, 111)
#fig.add_subplot(host)
#par = host.twinx()
#host.plot(x, mag, 'b-', label=mag_label)
#par.plot(x, pha, 'r--', label=pha_label)
#host.grid(grid, which="both", color='g')
#host.set_xlabel(xlabl)
#host.set_ylabel(mylabl, color='b')
#par.set_ylabel(pylabl, color='r')
#if log_mag == True:
#host.set_yscale('log', nonposy='clip')
#host.set_xlim([x0, x1])
#host.set_ylim([my0, my1])
#par.set_ylim([py0, py1])
#for tl in host.get_xticklabels():
#tl.set_fontsize(fontsize)
#for tl in host.get_yticklabels():
#tl.set_color('b')
#tl.set_fontsize(fontsize)
#for tl in par.get_yticklabels():
#tl.set_color('r')
#tl.set_fontsize(fontsize)
## host.set_title(title, fontsize=int(fontsize*1.3))
#host.set_title(title, fontsize=fontsize)
#if legend >= 0:
#host.legend(loc=legend)
return fig, ax1, ax2
