def h5_plot(h5file,
channel='channel00',
label=None,
xrange=None,
normtochan=None,
yrange=None):
# read the h5 file, formatted according to the XMI format
# If fit was performed, also read in the PyMCA config file to figure out detector calibration
h5file = np.array(h5file)
plt.figure(figsize=(20, 15))
for j in range(0, h5file.size):
if h5file.size == 1:
savename = str(h5file).split('.')[0] + '.png'
h5 = str(h5file)
else:
savename = str(h5file[0]).split('.')[0] + '-' + str(
h5file[-1]).split('.')[0].split('/')[-1] + '.png'
h5 = str(h5file[j])
fileid = str(h5.split('.')[0].split('/')[-1])
if channel == 'all':
chnl = ['channel00', 'channel02']
for i in range(0, 2):
spe, names, cfg = gplot_rh5(h5, chnl[i])
if normtochan is not None:
spe = spe[:] / spe[normtochan]
if cfg is None:
zero = 0.
gain = 1.
xtitle = "Channel Number"
else:
zero = cfg[0]
gain = cfg[1]
xtitle = "Energy [keV]"
# plot the spectrum, Ylog, X-axis converted to Energy (keV) if PyMca cfg provided
plt.plot(
np.linspace(0, spe.shape[0] - 1, num=spe.shape[0]) * gain +
zero,
spe,
label=fileid + '_' + str(chnl[i]),
linestyle='-')
else:
spe, names, cfg = gplot_rh5(h5, channel)
if normtochan is not None:
spe = spe[:] / spe[normtochan]
if cfg is None:
zero = 0.
gain = 1.
xtitle = "Channel Number"
else:
zero = cfg[0]
gain = cfg[1]
xtitle = "Energy [keV]"
# plot the spectrum, Ylog, X-axis converted to Energy (keV) if PyMca cfg provided
if xrange is None:
xrange = (zero, (spe.shape[0] - 1) * gain + zero)
plt.plot(
np.linspace(0, spe.shape[0] - 1, num=spe.shape[0]) * gain +
zero,
spe,
label=fileid + '_' + str(channel),
linestyle='-')
ax = plt.gca()
handles, labels = ax.get_legend_handles_labels()
plt.yscale('log')
plt.xlim(xrange)
if yrange is not None:
plt.ylim(yrange)
plt.legend(handles, labels, loc='best')
ax.set_xlabel(xtitle, fontsize=16)
ax.set_ylabel("Intensity [counts]", fontsize=16)
ax.tick_params(axis='both', which='major', labelsize=14)
ax.tick_params(axis='x', which='minor', bottom=True)
# add peak annotation if names and cfg provided
if cfg is not None and names is not None:
# x axis of plot is in Energy (keV)
# determine peak energy values (Rayl energy = cfg[2])
from PyMca5.PyMcaPhysics.xrf import Elements
for n in names:
if n != 'Rayl' and n != 'Compt':
if n.split(" ")[1][0] == 'K':
energy = Elements.getxrayenergy(
n.split(" ")[0],
n.split(" ")[1] + 'L3') #Ka1
elif n.split(" ")[1][0] == 'L':
energy = Elements.getxrayenergy(
n.split(" ")[0],
n.split(" ")[1] + '3M5') #La1
elif n.split(" ")[1][0] == 'M':
energy = Elements.getxrayenergy(
n.split(" ")[0],
n.split(" ")[1] + '5N7') #Ma1
elif n == 'Rayl':
energy = cfg[2]
else:
energy = None
# if energy not None, plot label at this energy value
if energy is not None:
idx = max(np.where(handles[0].get_xdata() <= energy)[-1])
yval = 10**(
np.log10(max([hand.get_ydata()[idx]
for hand in handles])) * 1.025)
# plot the text label X% above this value
plt.text(energy,
yval,
n,
horizontalalignment='center',
fontsize=14)
#TODO: Increase fontsizes, add minor tickmarks on x-axis
plt.show()
plt.savefig(savename) #, bbox_inches='tight', pad_inches=0)
plt.close()
def plot(data,
labels=None,
cfg=None,
xrange=None,
yrange=None,
normtochan=None,
savename='plot.png'):
data = np.array(data) #expected dimensions: [N, channels]
if len(data.shape) == 1:
data = np.reshape((1, data.shape[0]))
elif len(data.shape) >= 3:
print("Error: expected data dimensions: [N, channels]; ", data.shape)
return
if cfg is not None: # we're looking for the energy calibration values...
from PyMca5.PyMca import ConfigDict
config = ConfigDict.ConfigDict()
config.read(cfg)
cfg = [
config['detector']['zero'], config['detector']['gain'],
config['fit']['energy'][0]
] #zero, gain, E_Rayleigh
names = []
peaks = list(config['peaks'].keys())
for peak in peaks:
for linetype in config['peaks'][peak]:
names.append(peak + ' ' + str(linetype))
plt.figure(figsize=(20, 15))
for j in range(0, data.shape[0]):
spe = data[j, :]
if labels is None:
lbl = 'spectrum ' + str(j)
else:
lbl = labels[j]
if normtochan is not None:
spe = spe[:] / spe[normtochan]
if cfg is None:
zero = 0.
gain = 1.
xtitle = "Channel Number"
else:
zero = cfg[0]
gain = cfg[1]
xtitle = "Energy [keV]"
# plot the spectrum, Ylog, X-axis converted to Energy (keV) if PyMca cfg provided
if xrange is None:
xrange = (zero, (spe.shape[0] - 1) * gain + zero)
plt.plot(np.linspace(0, spe.shape[0] - 1, num=spe.shape[0]) * gain +
zero,
spe,
label=lbl,
linestyle='-')
ax = plt.gca()
handles, labels = ax.get_legend_handles_labels()
plt.yscale('log')
plt.xlim(xrange)
if yrange is not None:
plt.ylim(yrange)
plt.legend(handles, labels, loc='best')
ax.set_xlabel(xtitle, fontsize=16)
ax.set_ylabel("Intensity [counts]", fontsize=16)
ax.tick_params(axis='both', which='major', labelsize=14)
ax.tick_params(axis='x', which='minor', bottom=True)
# add peak annotation if names and cfg provided
if cfg is not None:
# x axis of plot is in Energy (keV)
# determine peak energy values (Rayl energy = cfg[2])
from PyMca5.PyMcaPhysics.xrf import Elements
for n in names:
if n != 'Rayl' and n != 'Compt':
if n.split(" ")[1][0] == 'K':
energy = Elements.getxrayenergy(
n.split(" ")[0],
n.split(" ")[1] + 'L3') #Ka1
elif n.split(" ")[1][0] == 'L':
energy = Elements.getxrayenergy(
n.split(" ")[0],
n.split(" ")[1] + '3M5') #La1
elif n.split(" ")[1][0] == 'M':
energy = Elements.getxrayenergy(
n.split(" ")[0],
n.split(" ")[1] + '5N7') #Ma1
elif n == 'Rayl':
energy = cfg[2]
else:
energy = None
# if energy not None, plot label at this energy value
if energy is not None:
idx = max(np.where(handles[0].get_xdata() <= energy)[-1])
yval = 10**(
np.log10(max([hand.get_ydata()[idx]
for hand in handles])) * 1.025)
# plot the text label X% above this value
plt.text(energy,
yval,
n,
horizontalalignment='center',
fontsize=14)
plt.show()
plt.savefig(savename) #, bbox_inches='tight', pad_inches=0)
plt.close()
