class XASNormPanel(TaskPanel):
"""XAS normalization Panel"""
def __init__(self, parent, controller=None, **kws):
TaskPanel.__init__(self,
parent,
controller,
configname='xasnorm_config',
config=defaults,
**kws)
def build_display(self):
panel = self.panel
self.wids = {}
self.plotone_op = Choice(panel,
choices=list(PlotOne_Choices.keys()),
action=self.onPlotOne,
size=(200, -1))
self.plotsel_op = Choice(panel,
choices=list(PlotSel_Choices.keys()),
action=self.onPlotSel,
size=(200, -1))
self.plotone_op.SetSelection(1)
self.plotsel_op.SetSelection(1)
plot_one = Button(panel,
'Plot Current Group',
size=(170, -1),
action=self.onPlotOne)
plot_sel = Button(panel,
'Plot Selected Groups',
size=(170, -1),
action=self.onPlotSel)
e0panel = wx.Panel(panel)
self.wids['auto_e0'] = Check(e0panel,
default=True,
label='auto?',
action=self.onSet_XASE0)
self.wids['showe0'] = Check(e0panel,
default=True,
label='show?',
action=self.onSet_XASE0)
sx = wx.BoxSizer(wx.HORIZONTAL)
sx.Add(self.wids['auto_e0'], 0, LEFT, 4)
sx.Add(self.wids['showe0'], 0, LEFT, 4)
pack(e0panel, sx)
self.wids['auto_step'] = Check(panel,
default=True,
label='auto?',
action=self.onNormMethod)
self.wids['nvict'] = Choice(panel,
choices=('0', '1', '2', '3'),
size=(100, -1),
action=self.onNormMethod,
default=0)
self.wids['nnorm'] = Choice(panel,
choices=list(Nnorm_choices.values()),
size=(100, -1),
action=self.onNormMethod,
default=0)
opts = {
'size': (100, -1),
'digits': 2,
'increment': 5.0,
'action': self.onSet_Ranges
}
xas_pre1 = self.add_floatspin('pre1', value=defaults['pre1'], **opts)
xas_pre2 = self.add_floatspin('pre2', value=defaults['pre2'], **opts)
xas_norm1 = self.add_floatspin('norm1',
value=defaults['norm1'],
**opts)
xas_norm2 = self.add_floatspin('norm2',
value=defaults['norm2'],
**opts)
opts = {'digits': 3, 'increment': 0.1, 'value': 0}
plot_voff = self.add_floatspin('plot_voff',
with_pin=False,
size=(80, -1),
action=self.onVoffset,
**opts)
xas_e0 = self.add_floatspin('e0', action=self.onSet_XASE0Val, **opts)
xas_step = self.add_floatspin('step',
action=self.onSet_XASStep,
with_pin=False,
min_val=0.0,
**opts)
opts['value'] = 1.0
scale = self.add_floatspin('scale', action=self.onSet_Scale, **opts)
self.wids['norm_method'] = Choice(
panel,
choices=('polynomial', 'mback'), # , 'area'),
size=(120, -1),
action=self.onNormMethod)
self.wids['norm_method'].SetSelection(0)
atsyms = ['?'] + self.larch.symtable._xray._xraydb.atomic_symbols
edges = ('K', 'L3', 'L2', 'L1', 'M5')
self.wids['atsym'] = Choice(panel, choices=atsyms, size=(75, -1))
self.wids['edge'] = Choice(panel, choices=edges, size=(60, -1))
self.wids['is_frozen'] = Check(panel,
default=False,
label='Freeze Group',
action=self.onFreezeGroup)
saveconf = Button(panel,
'Save as Default Settings',
size=(200, -1),
action=self.onSaveConfigBtn)
use_auto = Button(panel,
'Use Default Settings',
size=(200, -1),
action=self.onAutoNorm)
copy_auto = Button(panel,
'Copy',
size=(60, -1),
action=self.onCopyAuto)
def CopyBtn(name):
return Button(panel,
'Copy',
size=(60, -1),
action=partial(self.onCopyParam, name))
add_text = self.add_text
HLINEWID = 575
panel.Add(SimpleText(panel,
'XAS Pre-edge subtraction and Normalization',
**self.titleopts),
dcol=4)
panel.Add(SimpleText(panel, 'Copy to Selected Groups:'),
style=RIGHT,
dcol=2)
panel.Add(plot_sel, newrow=True)
panel.Add(self.plotsel_op, dcol=3)
panel.Add(SimpleText(panel, 'Y Offset:'), style=RIGHT)
panel.Add(plot_voff, style=RIGHT)
panel.Add(plot_one, newrow=True)
panel.Add(self.plotone_op, dcol=4)
panel.Add(CopyBtn('plotone_op'), dcol=1, style=RIGHT)
panel.Add(HLine(panel, size=(HLINEWID, 3)), dcol=6, newrow=True)
add_text('Non-XAS Data Scale:')
panel.Add(scale, dcol=2)
panel.Add(HLine(panel, size=(HLINEWID, 3)), dcol=6, newrow=True)
add_text('XAS Data:')
panel.Add(use_auto, dcol=4)
panel.Add(copy_auto, dcol=1, style=RIGHT)
add_text('Element and Edge: ', newrow=True)
panel.Add(self.wids['atsym'])
panel.Add(self.wids['edge'], dcol=3)
panel.Add(CopyBtn('atsym'), dcol=1, style=RIGHT)
add_text('E0 : ')
panel.Add(xas_e0)
panel.Add(e0panel, dcol=3)
panel.Add(CopyBtn('xas_e0'), dcol=1, style=RIGHT)
add_text('Edge Step: ')
panel.Add(xas_step)
panel.Add(self.wids['auto_step'], dcol=3)
panel.Add(CopyBtn('xas_step'), dcol=1, style=RIGHT)
panel.Add((5, 5), newrow=True)
panel.Add(HLine(panel, size=(HLINEWID, 3)), dcol=6, newrow=True)
add_text('Pre-edge range: ')
panel.Add(xas_pre1)
add_text(' : ', newrow=False)
panel.Add(xas_pre2, dcol=2)
panel.Add(CopyBtn('xas_pre'), dcol=1, style=RIGHT)
panel.Add(SimpleText(panel, 'Victoreen order:'), newrow=True)
panel.Add(self.wids['nvict'], dcol=4)
panel.Add((5, 5), newrow=True)
panel.Add(HLine(panel, size=(HLINEWID, 3)), dcol=6, newrow=True)
add_text('Normalization method: ')
panel.Add(self.wids['norm_method'], dcol=4)
panel.Add(CopyBtn('xas_norm'), dcol=1, style=RIGHT)
add_text('Normalization range: ')
panel.Add(xas_norm1)
add_text(' : ', newrow=False)
panel.Add(xas_norm2, dcol=2)
panel.Add(SimpleText(panel, 'Polynomial Type:'), newrow=True)
panel.Add(self.wids['nnorm'], dcol=4)
panel.Add(HLine(panel, size=(HLINEWID, 3)), dcol=6, newrow=True)
panel.Add((5, 5), newrow=True)
panel.Add(self.wids['is_frozen'], newrow=True)
panel.Add(saveconf, dcol=5)
panel.Add((5, 5), newrow=True)
panel.Add(HLine(panel, size=(HLINEWID, 3)), dcol=6, newrow=True)
panel.pack()
sizer = wx.BoxSizer(wx.VERTICAL)
sizer.Add((5, 5), 0, LEFT, 3)
sizer.Add(panel, 0, LEFT, 3)
sizer.Add((5, 5), 0, LEFT, 3)
pack(self, sizer)
def get_config(self, dgroup=None):
"""custom get_config to possibly inherit from Athena settings"""
if dgroup is None:
dgroup = self.controller.get_group()
if dgroup is None:
return self.get_defaultconfig()
if hasattr(dgroup, self.configname):
conf = getattr(dgroup, self.configname)
else:
conf = self.get_defaultconfig()
if hasattr(dgroup, 'bkg_params'): # from Athena
for attr in ('e0', 'pre1', 'pre2', 'nnorm'):
conf[attr] = getattr(dgroup.bkg_params, attr, conf[attr])
for attr, aattr in (('norm1', 'nor1'), ('norm2', 'nor2')):
conf[attr] = getattr(dgroup.bkg_params, aattr, conf[attr])
conf['auto_step'] = (float(
getattr(dgroup.bkg_params, 'fixstep', 0.0)) < 0.5)
conf['edge_step'] = getattr(dgroup.bkg_params, 'step',
conf['edge_step'])
if conf['edge_step'] is None:
conf['edge_step'] = getattr(dgroup, 'edge_step', conf['edge_step'])
conf['atsym'] = getattr(dgroup, 'atsym', conf['atsym'])
conf['edge'] = getattr(dgroup, 'edge', conf['edge'])
if hasattr(dgroup, 'e0') and conf['atsym'] == '?':
atsym, edge = guess_edge(dgroup.e0)
conf['atsym'] = atsym
conf['edge'] = edge
if hasattr(dgroup, 'mback_params'):
conf['atsym'] = getattr(dgroup.mback_params, 'atsym',
conf['atsym'])
conf['edge'] = getattr(dgroup.mback_params, 'edge', conf['edge'])
setattr(dgroup, self.configname, conf)
return conf
def fill_form(self, dgroup):
"""fill in form from a data group"""
opts = self.get_config(dgroup)
self.skip_process = True
if dgroup.datatype == 'xas':
self.plotone_op.SetChoices(list(PlotOne_Choices.keys()))
self.plotsel_op.SetChoices(list(PlotSel_Choices.keys()))
self.plotone_op.SetStringSelection(opts['plotone_op'])
self.plotsel_op.SetStringSelection(opts['plotsel_op'])
self.wids['e0'].SetValue(opts['e0'])
edge_step = opts.get('edge_step', None)
if edge_step is None:
edge_step = 1.0
if hasattr(dgroup, 'e0') and opts['atsym'] == '?':
atsym, edge = guess_edge(dgroup.e0)
opts['atsym'] = atsym
opts['edge'] = edge
self.wids['step'].SetValue(edge_step)
autoset_fs_increment(self.wids['step'], edge_step)
for attr in ('pre1', 'pre2', 'norm1', 'norm2'):
val = opts.get(attr, None)
if val is not None:
self.wids[attr].SetValue(val)
self.set_nnorm_widget(opts.get('nnorm'))
self.wids['nvict'].SetSelection(opts['nvict'])
self.wids['showe0'].SetValue(opts['show_e0'])
self.wids['auto_e0'].SetValue(opts['auto_e0'])
self.wids['auto_step'].SetValue(opts['auto_step'])
self.wids['edge'].SetStringSelection(opts['edge'].title())
self.wids['atsym'].SetStringSelection(opts['atsym'].title())
self.wids['norm_method'].SetStringSelection(
opts['norm_method'].lower())
for attr in ('pre1', 'pre2', 'norm1', 'norm2', 'nnorm', 'edge',
'atsym', 'step', 'norm_method'):
self.wids[attr].Enable()
self.wids['scale'].Disable()
else:
self.plotone_op.SetChoices(list(PlotOne_Choices_nonxas.keys()))
self.plotsel_op.SetChoices(list(PlotSel_Choices_nonxas.keys()))
self.wids['scale'].SetValue(opts['scale'])
for attr in ('pre1', 'pre2', 'norm1', 'norm2', 'nnorm', 'edge',
'atsym', 'step', 'norm_method'):
self.wids[attr].Disable()
self.wids['scale'].Enable()
frozen = opts.get('is_frozen', False)
if hasattr(dgroup, 'is_frozen'):
frozen = dgroup.is_frozen
self.wids['is_frozen'].SetValue(frozen)
self._set_frozen(frozen)
wx.CallAfter(self.unset_skip_process)
def set_nnorm_widget(self, nnorm=None):
if nnorm is None:
nnorm_str = 'auto'
else:
try:
nnorm = int(nnorm)
except ValueError:
nnorm = None
nnorm_str = Nnorm_choices.get(nnorm, 'auto')
self.wids['nnorm'].SetStringSelection(nnorm_str)
def unset_skip_process(self):
self.skip_process = False
def read_form(self):
"read form, return dict of values"
form_opts = {}
form_opts['e0'] = self.wids['e0'].GetValue()
form_opts['edge_step'] = self.wids['step'].GetValue()
for attr in ('pre1', 'pre2', 'norm1', 'norm2'):
val = self.wids[attr].GetValue()
if val == 0: val = None
form_opts[attr] = val
form_opts['nnorm'] = Nnorm_names.get(
self.wids['nnorm'].GetStringSelection(), None)
form_opts['nvict'] = int(self.wids['nvict'].GetSelection())
form_opts['plotone_op'] = self.plotone_op.GetStringSelection()
form_opts['plotsel_op'] = self.plotsel_op.GetStringSelection()
form_opts['plot_voff'] = self.wids['plot_voff'].GetValue()
form_opts['show_e0'] = self.wids['showe0'].IsChecked()
form_opts['auto_e0'] = self.wids['auto_e0'].IsChecked()
form_opts['auto_step'] = self.wids['auto_step'].IsChecked()
form_opts['norm_method'] = self.wids['norm_method'].GetStringSelection(
).lower()
form_opts['edge'] = self.wids['edge'].GetStringSelection().title()
form_opts['atsym'] = self.wids['atsym'].GetStringSelection().title()
form_opts['scale'] = self.wids['scale'].GetValue()
return form_opts
def onNormMethod(self, evt=None):
method = self.wids['norm_method'].GetStringSelection().lower()
self.update_config({'norm_method': method})
if method.startswith('mback'):
dgroup = self.controller.get_group()
cur_elem = self.wids['atsym'].GetStringSelection()
if hasattr(dgroup, 'e0') and cur_elem == 'H':
atsym, edge = guess_edge(dgroup.e0)
self.wids['edge'].SetStringSelection(edge)
self.wids['atsym'].SetStringSelection(atsym)
self.update_config({'edge': edge, 'atsym': atsym})
time.sleep(0.01)
wx.CallAfter(self.onReprocess)
def _set_frozen(self, frozen):
try:
dgroup = self.controller.get_group()
dgroup.is_frozen = frozen
except:
pass
for wattr in ('e0', 'step', 'pre1', 'pre2', 'norm1', 'norm2', 'nvict',
'nnorm', 'showe0', 'auto_e0', 'auto_step', 'norm_method',
'edge', 'atsym'):
self.wids[wattr].Enable(not frozen)
def onFreezeGroup(self, evt=None):
self._set_frozen(evt.IsChecked())
def onPlotOne(self, evt=None):
self.plot(self.controller.get_group())
def onVoffset(self, evt=None):
time.sleep(0.01)
wx.CallAfter(self.onPlotSel)
def onPlotSel(self, evt=None):
newplot = True
group_ids = self.controller.filelist.GetCheckedStrings()
if len(group_ids) < 1:
return
last_id = group_ids[-1]
groupname = self.controller.file_groups[str(last_id)]
dgroup = self.controller.get_group(groupname)
plot_choices = PlotSel_Choices
if dgroup.datatype != 'xas':
plot_choices = PlotSel_Choices_nonxas
ytitle = self.plotsel_op.GetStringSelection()
yarray_name = plot_choices[ytitle]
ylabel = getattr(plotlabels, yarray_name, ytitle)
if yarray_name == 'norm':
norm_method = self.wids['norm_method'].GetStringSelection().lower()
if norm_method.startswith('mback'):
yarray_name = 'norm_mback'
ylabel = "%s (MBACK)" % ylabel
elif norm_method.startswith('area'):
yarray_name = 'norm_area'
ylabel = "%s (Area)" % ylabel
voff = self.wids['plot_voff'].GetValue()
for ix, checked in enumerate(group_ids):
yoff = ix * voff
groupname = self.controller.file_groups[str(checked)]
dgroup = self.controller.get_group(groupname)
plot_yarrays = [(yarray_name, PLOTOPTS_1, dgroup.filename)]
if dgroup is not None:
dgroup.plot_extras = []
self.plot(dgroup,
title='',
new=newplot,
multi=True,
yoff=yoff,
plot_yarrays=plot_yarrays,
with_extras=False,
delay_draw=True)
newplot = False
ppanel = self.controller.get_display(stacked=False).panel
ppanel.conf.show_legend = True
ppanel.conf.draw_legend()
ppanel.unzoom_all()
def onAutoNorm(self, evt=None):
dgroup = self.controller.get_group()
try:
norm2 = max(dgroup.energy) - dgroup.e0
norm1 = 5.0 * int(norm2 / 15.0)
nnorm = 2
if (norm2 - norm1 < 350): nnorm = 1
if (norm2 - norm1 < 50): nnorm = 0
except:
nnorm = None
self.wids['auto_step'].SetValue(1)
self.wids['auto_e0'].SetValue(1)
self.wids['nvict'].SetSelection(0)
self.wids['pre1'].SetValue(0)
self.wids['pre2'].SetValue(0)
self.wids['norm1'].SetValue(0)
self.wids['norm2'].SetValue(0)
if nnorm is not None:
self.set_nnorm_widget(nnorm)
self.wids['norm_method'].SetSelection(0)
self.onReprocess()
def onCopyAuto(self, evt=None):
opts = dict(pre1=0,
pre2=0,
nvict=0,
norm1=0,
norm2=0,
norm_method='polynomial',
nnorm=2,
auto_e0=1,
auto_step=1)
for checked in self.controller.filelist.GetCheckedStrings():
groupname = self.controller.file_groups[str(checked)]
grp = self.controller.get_group(groupname)
if grp != self.controller.group and not grp.is_frozen:
# try:
# norm2 = max(grp.energy) - grp.e0
# norm1 = 5.0*int(norm2/15.0)
# nnorm = 2
# if (norm2-norm1 < 350): nnorm = 1
# if (norm2-norm1 < 50): nnorm = 0
# except:
# nnorm = 1
# opts['nnorm'] = nnorm
self.update_config(opts, dgroup=grp)
self.fill_form(grp)
self.process(grp, noskip=True)
def onSaveConfigBtn(self, evt=None):
conf = self.get_config()
conf.update(self.read_form())
self.set_defaultconfig(conf)
def onCopyParam(self, name=None, evt=None):
conf = self.get_config()
form = self.read_form()
conf.update(form)
dgroup = self.controller.get_group()
self.update_config(conf)
self.fill_form(dgroup)
opts = {}
name = str(name)
def copy_attrs(*args):
for a in args:
opts[a] = conf[a]
if name == 'plotone_op':
copy_attrs('plotone_op')
elif name == 'xas_e0':
copy_attrs('e0', 'show_e0', 'auto_e0')
elif name == 'xas_step':
copy_attrs('edge_step', 'auto_step')
elif name == 'xas_pre':
copy_attrs('pre1', 'pre2', 'nvict')
elif name == 'atsym':
copy_attrs('atsym', 'edge')
elif name == 'xas_norm':
copy_attrs('norm_method', 'nnorm', 'norm1', 'norm2')
for checked in self.controller.filelist.GetCheckedStrings():
groupname = self.controller.file_groups[str(checked)]
grp = self.controller.get_group(groupname)
if grp != self.controller.group and not grp.is_frozen:
self.update_config(opts, dgroup=grp)
self.fill_form(grp)
self.process(grp, noskip=True)
def onSet_XASE0(self, evt=None, value=None):
"handle setting auto e0 / show e0"
auto_e0 = self.wids['auto_e0'].GetValue()
self.update_config({
'e0': self.wids['e0'].GetValue(),
'auto_e0': self.wids['auto_e0'].GetValue()
})
time.sleep(0.01)
wx.CallAfter(self.onReprocess)
def onSet_XASE0Val(self, evt=None, value=None):
"handle setting e0"
self.wids['auto_e0'].SetValue(0)
self.update_config({
'e0': self.wids['e0'].GetValue(),
'auto_e0': self.wids['auto_e0'].GetValue()
})
time.sleep(0.01)
wx.CallAfter(self.onReprocess)
def onSet_XASStep(self, evt=None, value=None):
"handle setting edge step"
edge_step = self.wids['step'].GetValue()
if edge_step < 0:
self.wids['step'].SetValue(abs(edge_step))
self.wids['auto_step'].SetValue(0)
self.update_config({'edge_step': abs(edge_step), 'auto_step': False})
autoset_fs_increment(self.wids['step'], abs(edge_step))
time.sleep(0.01)
wx.CallAfter(self.onReprocess)
def onSet_Scale(self, evt=None, value=None):
"handle setting non-XAFS scale value"
self.update_config({'scale': self.wids['scale'].GetValue()})
time.sleep(0.01)
wx.CallAfter(self.onReprocess)
def onSet_Ranges(self, evt=None, **kws):
conf = {}
for attr in ('pre1', 'pre2', 'norm1', 'norm2'):
conf[attr] = self.wids[attr].GetValue()
self.update_config(conf)
time.sleep(0.01)
wx.CallAfter(self.onReprocess)
def onSelPoint(self, evt=None, opt='__', relative_e0=True, win=None):
"""
get last selected point from a specified plot window
and fill in the value for the widget defined by `opt`.
by default it finds the latest cursor position from the
cursor history of the first 20 plot windows.
"""
if opt not in self.wids:
return None
_x, _y = last_cursor_pos(win=win, _larch=self.larch)
if _x is None:
return
e0 = self.wids['e0'].GetValue()
if opt == 'e0':
self.wids['e0'].SetValue(_x)
self.wids['auto_e0'].SetValue(0)
elif opt in ('pre1', 'pre2', 'norm1', 'norm2'):
self.wids[opt].SetValue(_x - e0)
time.sleep(0.01)
wx.CallAfter(self.onReprocess)
def onReprocess(self, evt=None, value=None, **kws):
"handle request reprocess"
if self.skip_process:
return
try:
dgroup = self.controller.get_group()
except TypeError:
return
if not hasattr(dgroup, self.configname):
return
form = self.read_form()
self.process(dgroup=dgroup)
self.plot(dgroup)
def make_dnormde(self, dgroup):
form = dict(group=dgroup.groupname)
self.larch_eval(
"{group:s}.dnormde={group:s}.dmude/{group:s}.edge_step".format(
**form))
def process(self, dgroup=None, force_mback=False, noskip=False, **kws):
""" handle process (pre-edge/normalize) of XAS data from XAS form
"""
if self.skip_process and not noskip:
return
if dgroup is None:
dgroup = self.controller.get_group()
if dgroup is None:
return
self.skip_process = True
conf = self.get_config(dgroup)
dgroup.custom_plotopts = {}
form = self.read_form()
form['group'] = dgroup.groupname
if dgroup.datatype != 'xas':
self.skip_process = False
dgroup.mu = dgroup.ydat * 1.0
opts = {'group': dgroup.groupname, 'scale': conf.get('scale', 1.0)}
self.larch_eval("{group:s}.scale = {scale:.8f}".format(**opts))
self.larch_eval(
"{group:s}.norm = {scale:.8f}*{group:s}.ydat".format(**opts))
return
en_units = getattr(dgroup, 'energy_units', None)
if en_units is None:
en_units = guess_energy_units(dgroup.energy)
if en_units != 'eV':
mono_dspace = getattr(dgroup, 'mono_dspace', 1)
dlg = EnergyUnitsDialog(self.parent,
dgroup.energy,
unitname=en_units,
dspace=mono_dspace)
res = dlg.GetResponse()
dlg.Destroy()
if res.ok:
en_units = res.units
dgroup.mono_dspace = res.dspace
dgroup.xdat = dgroup.energy = res.energy
dgroup.energy_units = en_units
e0 = form['e0']
edge_step = form['edge_step']
copts = [dgroup.groupname]
if not form['auto_e0']:
if e0 < max(dgroup.energy) and e0 > min(dgroup.energy):
copts.append("e0=%.4f" % float(e0))
if not form['auto_step']:
copts.append("step=%s" % gformat(float(edge_step)))
for attr in ('pre1', 'pre2', 'nvict', 'nnorm', 'norm1', 'norm2'):
if form[attr] is None:
copts.append("%s=None" % attr)
else:
copts.append("%s=%.2f" % (attr, form[attr]))
self.larch_eval("pre_edge(%s)" % (', '.join(copts)))
self.larch_eval(
"{group:s}.norm_poly = 1.0*{group:s}.norm".format(**form))
norm_method = form['norm_method'].lower()
form['normmeth'] = 'poly'
if force_mback or norm_method.startswith('mback'):
form['normmeth'] = 'mback'
copts = [dgroup.groupname]
copts.append("z=%d" % atomic_number(form['atsym']))
copts.append("edge='%s'" % form['edge'])
for attr in ('pre1', 'pre2', 'nvict', 'nnorm', 'norm1', 'norm2'):
if form[attr] is None:
copts.append("%s=None" % attr)
else:
copts.append("%s=%.2f" % (attr, form[attr]))
self.larch_eval("mback_norm(%s)" % (', '.join(copts)))
if form['auto_step']:
norm_expr = """{group:s}.norm = 1.0*{group:s}.norm_{normmeth:s}
{group:s}.edge_step = 1.0*{group:s}.edge_step_{normmeth:s}"""
self.larch_eval(norm_expr.format(**form))
else:
norm_expr = """{group:s}.norm = 1.0*{group:s}.norm_{normmeth:s}
{group:s}.norm *= {group:s}.edge_step_{normmeth:s}/{edge_step:.8f}"""
self.larch_eval(norm_expr.format(**form))
if norm_method.startswith('area'):
form['normmeth'] = 'area'
expr = """{group:s}.norm = 1.0*{group:s}.norm_{normmeth:s}
{group:s}.edge_step = 1.0*{group:s}.edge_step_{normmeth:s}"""
self.larch_eval(expr.format(**form))
self.make_dnormde(dgroup)
if form['auto_e0']:
self.wids['e0'].SetValue(dgroup.e0)
if form['auto_step']:
self.wids['step'].SetValue(dgroup.edge_step)
autoset_fs_increment(self.wids['step'], dgroup.edge_step)
self.wids['atsym'].SetStringSelection(dgroup.atsym)
self.wids['edge'].SetStringSelection(dgroup.edge)
self.set_nnorm_widget(dgroup.pre_edge_details.nnorm)
for attr in ('e0', 'edge_step'):
conf[attr] = getattr(dgroup, attr)
for attr in ('pre1', 'pre2', 'norm1', 'norm2'):
conf[attr] = val = getattr(dgroup.pre_edge_details, attr, None)
if val is not None:
self.wids[attr].SetValue(val)
if hasattr(dgroup, 'mback_params'): # from mback
conf['atsym'] = getattr(dgroup.mback_params, 'atsym')
conf['edge'] = getattr(dgroup.mback_params, 'edge')
self.update_config(conf, dgroup=dgroup)
wx.CallAfter(self.unset_skip_process)
def get_plot_arrays(self, dgroup):
lab = plotlabels.norm
if dgroup is None:
return
dgroup.plot_y2label = None
dgroup.plot_xlabel = plotlabels.energy
dgroup.plot_yarrays = [('norm', PLOTOPTS_1, lab)]
if dgroup.datatype != 'xas':
pchoice = PlotOne_Choices_nonxas[
self.plotone_op.GetStringSelection()]
dgroup.plot_xlabel = 'x'
dgroup.plot_ylabel = 'y'
dgroup.plot_yarrays = [('ydat', PLOTOPTS_1, 'ydat')]
dgroup.dmude = np.gradient(dgroup.ydat) / np.gradient(dgroup.xdat)
if not hasattr(dgroup, 'scale'):
dgroup.scale = 1.0
dgroup.norm = dgroup.ydat * dgroup.scale
if pchoice == 'dmude':
dgroup.plot_ylabel = 'dy/dx'
dgroup.plot_yarrays = [('dmude', PLOTOPTS_1, 'dy/dx')]
elif pchoice == 'norm':
dgroup.plot_ylabel = 'scaled y'
dgroup.plot_yarrays = [('norm', PLOTOPTS_1, 'y/scale')]
elif pchoice == 'norm+dnormde':
lab = plotlabels.norm
dgroup.plot_y2label = 'dy/dx'
dgroup.plot_yarrays = [('ydat', PLOTOPTS_1, 'y'),
('dnormde', PLOTOPTS_D, 'dy/dx')]
return
req_attrs = ['e0', 'norm', 'dmude', 'pre_edge']
pchoice = PlotOne_Choices[self.plotone_op.GetStringSelection()]
if pchoice in ('mu', 'norm', 'flat', 'dmude'):
lab = getattr(plotlabels, pchoice)
dgroup.plot_yarrays = [(pchoice, PLOTOPTS_1, lab)]
elif pchoice == 'prelines':
dgroup.plot_yarrays = [('mu', PLOTOPTS_1, plotlabels.mu),
('pre_edge', PLOTOPTS_2, 'pre edge'),
('post_edge', PLOTOPTS_2, 'post edge')]
elif pchoice == 'preedge':
lab = r'pre-edge subtracted $\mu$'
dgroup.pre_edge_sub = dgroup.norm * dgroup.edge_step
dgroup.plot_yarrays = [('pre_edge_sub', PLOTOPTS_1, lab)]
elif pchoice == 'mu+dmude':
lab = plotlabels.mu
lab2 = plotlabels.dmude
dgroup.plot_yarrays = [('mu', PLOTOPTS_1, lab),
('dmude', PLOTOPTS_D, lab2)]
dgroup.plot_y2label = lab2
elif pchoice == 'norm+dnormde':
lab = plotlabels.norm
lab2 = plotlabels.dmude + ' (normalized)'
dgroup.plot_yarrays = [('norm', PLOTOPTS_1, lab),
('dnormde', PLOTOPTS_D, lab2)]
dgroup.plot_y2label = lab2
elif pchoice == 'mback_norm':
req_attrs.append('mback_norm')
lab = r'$\mu$'
if not hasattr(dgroup, 'mback_mu'):
self.process(dgroup=dgroup, force_mback=True)
dgroup.plot_yarrays = [('mu', PLOTOPTS_1, lab),
('mback_mu', PLOTOPTS_2,
r'tabulated $\mu(E)$')]
elif pchoice == 'mback_poly':
req_attrs.append('mback_norm')
lab = plotlabels.norm
if not hasattr(dgroup, 'mback_mu'):
self.process(dgroup=dgroup, force_mback=True)
dgroup.plot_yarrays = [('norm_mback', PLOTOPTS_1, 'mback'),
('norm_poly', PLOTOPTS_2, 'polynomial')]
elif pchoice == 'area_norm':
dgroup.plot_yarrays = [('norm_area', PLOTOPTS_1, 'area'),
('norm_poly', PLOTOPTS_2, 'polynomial')]
dgroup.plot_ylabel = lab
needs_proc = False
for attr in req_attrs:
needs_proc = needs_proc or (not hasattr(dgroup, attr))
if needs_proc:
self.process(dgroup=dgroup, noskip=True)
y4e0 = dgroup.ydat = getattr(dgroup, dgroup.plot_yarrays[0][0],
dgroup.mu)
dgroup.plot_extras = []
if self.wids['showe0'].IsChecked():
ie0 = index_of(dgroup.energy, dgroup.e0)
dgroup.plot_extras.append(('marker', dgroup.e0, y4e0[ie0], {}))
def plot(self,
dgroup,
title=None,
plot_yarrays=None,
yoff=0,
delay_draw=False,
multi=False,
new=True,
zoom_out=True,
with_extras=True,
**kws):
if self.skip_plotting:
return
ppanel = self.controller.get_display(stacked=False).panel
viewlims = ppanel.get_viewlimits()
plotcmd = ppanel.oplot
if new:
plotcmd = ppanel.plot
groupname = getattr(dgroup, 'groupname', None)
if groupname is None:
return
if not hasattr(dgroup, 'xdat'):
print("Cannot plot group ", groupname)
if ((getattr(dgroup, 'plot_yarrays', None) is None
or getattr(dgroup, 'energy', None) is None
or getattr(dgroup, 'mu', None) is None
or getattr(dgroup, 'e0', None) is None
or getattr(dgroup, 'dmude', None) is None
or getattr(dgroup, 'norm', None) is None)):
self.process(dgroup=dgroup)
self.get_plot_arrays(dgroup)
if plot_yarrays is None and hasattr(dgroup, 'plot_yarrays'):
plot_yarrays = dgroup.plot_yarrays
popts = kws
path, fname = os.path.split(dgroup.filename)
if 'label' not in popts:
popts['label'] = dgroup.plot_ylabel
zoom_out = (zoom_out or min(dgroup.xdat) >= viewlims[1]
or max(dgroup.xdat) <= viewlims[0]
or min(dgroup.ydat) >= viewlims[3]
or max(dgroup.ydat) <= viewlims[2])
if not zoom_out:
popts['xmin'] = viewlims[0]
popts['xmax'] = viewlims[1]
popts['ymin'] = viewlims[2]
popts['ymax'] = viewlims[3]
popts['xlabel'] = dgroup.plot_xlabel
popts['ylabel'] = dgroup.plot_ylabel
if getattr(dgroup, 'plot_y2label', None) is not None:
popts['y2label'] = dgroup.plot_y2label
plot_choices = PlotSel_Choices
if dgroup.datatype != 'xas':
plot_choices = PlotSel_Choices_nonxas
if multi:
ylabel = self.plotsel_op.GetStringSelection()
yarray_name = plot_choices[ylabel]
if dgroup.datatype == 'xas':
ylabel = getattr(plotlabels, yarray_name, ylabel)
popts['ylabel'] = ylabel
plot_extras = None
if new:
if title is None:
title = fname
plot_extras = getattr(dgroup, 'plot_extras', None)
popts['title'] = title
popts['delay_draw'] = delay_draw
if hasattr(dgroup, 'custom_plotopts'):
popts.update(dgroup.custom_plotopts)
popts['show_legend'] = len(plot_yarrays) > 1
narr = len(plot_yarrays) - 1
for i, pydat in enumerate(plot_yarrays):
yaname, yopts, yalabel = pydat
popts.update(yopts)
if yalabel is not None:
popts['label'] = yalabel
popts['delay_draw'] = delay_draw or (i != narr)
if yaname == 'dnormde' and not hasattr(dgroup, yaname):
self.make_dnormde(dgroup)
if yaname == 'norm_mback' and not hasattr(dgroup, yaname):
self.process(dgroup=dgroup, noskip=True, force_mback=True)
plotcmd(dgroup.xdat, getattr(dgroup, yaname) + yoff, **popts)
plotcmd = ppanel.oplot
if with_extras and plot_extras is not None:
axes = ppanel.axes
for etype, x, y, opts in plot_extras:
if etype == 'marker':
xpopts = {
'marker': 'o',
'markersize': 4,
'label': '_nolegend_',
'markerfacecolor': 'red',
'markeredgecolor': '#884444'
}
xpopts.update(opts)
axes.plot([x], [y], **xpopts)
elif etype == 'vline':
xpopts = {
'ymin': 0,
'ymax': 1.0,
'label': '_nolegend_',
'color': '#888888'
}
xpopts.update(opts)
axes.axvline(x, **xpopts)
if not popts['delay_draw']:
ppanel.canvas.draw()
def build_display(self):
panel = self.panel
self.wids = {}
self.plotone_op = Choice(panel,
choices=list(PlotOne_Choices.keys()),
action=self.onPlotOne,
size=(200, -1))
self.plotsel_op = Choice(panel,
choices=list(PlotSel_Choices.keys()),
action=self.onPlotSel,
size=(200, -1))
self.plotone_op.SetSelection(1)
self.plotsel_op.SetSelection(1)
plot_one = Button(panel,
'Plot Current Group',
size=(170, -1),
action=self.onPlotOne)
plot_sel = Button(panel,
'Plot Selected Groups',
size=(170, -1),
action=self.onPlotSel)
e0panel = wx.Panel(panel)
self.wids['auto_e0'] = Check(e0panel,
default=True,
label='auto?',
action=self.onSet_XASE0)
self.wids['showe0'] = Check(e0panel,
default=True,
label='show?',
action=self.onSet_XASE0)
sx = wx.BoxSizer(wx.HORIZONTAL)
sx.Add(self.wids['auto_e0'], 0, LEFT, 4)
sx.Add(self.wids['showe0'], 0, LEFT, 4)
pack(e0panel, sx)
self.wids['auto_step'] = Check(panel,
default=True,
label='auto?',
action=self.onNormMethod)
self.wids['nvict'] = Choice(panel,
choices=('0', '1', '2', '3'),
size=(100, -1),
action=self.onNormMethod,
default=0)
self.wids['nnorm'] = Choice(panel,
choices=list(Nnorm_choices.values()),
size=(100, -1),
action=self.onNormMethod,
default=0)
opts = {
'size': (100, -1),
'digits': 2,
'increment': 5.0,
'action': self.onSet_Ranges
}
xas_pre1 = self.add_floatspin('pre1', value=defaults['pre1'], **opts)
xas_pre2 = self.add_floatspin('pre2', value=defaults['pre2'], **opts)
xas_norm1 = self.add_floatspin('norm1',
value=defaults['norm1'],
**opts)
xas_norm2 = self.add_floatspin('norm2',
value=defaults['norm2'],
**opts)
opts = {'digits': 3, 'increment': 0.1, 'value': 0}
plot_voff = self.add_floatspin('plot_voff',
with_pin=False,
size=(80, -1),
action=self.onVoffset,
**opts)
xas_e0 = self.add_floatspin('e0', action=self.onSet_XASE0Val, **opts)
xas_step = self.add_floatspin('step',
action=self.onSet_XASStep,
with_pin=False,
min_val=0.0,
**opts)
opts['value'] = 1.0
scale = self.add_floatspin('scale', action=self.onSet_Scale, **opts)
self.wids['norm_method'] = Choice(
panel,
choices=('polynomial', 'mback'), # , 'area'),
size=(120, -1),
action=self.onNormMethod)
self.wids['norm_method'].SetSelection(0)
atsyms = ['?'] + self.larch.symtable._xray._xraydb.atomic_symbols
edges = ('K', 'L3', 'L2', 'L1', 'M5')
self.wids['atsym'] = Choice(panel, choices=atsyms, size=(75, -1))
self.wids['edge'] = Choice(panel, choices=edges, size=(60, -1))
self.wids['is_frozen'] = Check(panel,
default=False,
label='Freeze Group',
action=self.onFreezeGroup)
saveconf = Button(panel,
'Save as Default Settings',
size=(200, -1),
action=self.onSaveConfigBtn)
use_auto = Button(panel,
'Use Default Settings',
size=(200, -1),
action=self.onAutoNorm)
copy_auto = Button(panel,
'Copy',
size=(60, -1),
action=self.onCopyAuto)
def CopyBtn(name):
return Button(panel,
'Copy',
size=(60, -1),
action=partial(self.onCopyParam, name))
add_text = self.add_text
HLINEWID = 575
panel.Add(SimpleText(panel,
'XAS Pre-edge subtraction and Normalization',
**self.titleopts),
dcol=4)
panel.Add(SimpleText(panel, 'Copy to Selected Groups:'),
style=RIGHT,
dcol=2)
panel.Add(plot_sel, newrow=True)
panel.Add(self.plotsel_op, dcol=3)
panel.Add(SimpleText(panel, 'Y Offset:'), style=RIGHT)
panel.Add(plot_voff, style=RIGHT)
panel.Add(plot_one, newrow=True)
panel.Add(self.plotone_op, dcol=4)
panel.Add(CopyBtn('plotone_op'), dcol=1, style=RIGHT)
panel.Add(HLine(panel, size=(HLINEWID, 3)), dcol=6, newrow=True)
add_text('Non-XAS Data Scale:')
panel.Add(scale, dcol=2)
panel.Add(HLine(panel, size=(HLINEWID, 3)), dcol=6, newrow=True)
add_text('XAS Data:')
panel.Add(use_auto, dcol=4)
panel.Add(copy_auto, dcol=1, style=RIGHT)
add_text('Element and Edge: ', newrow=True)
panel.Add(self.wids['atsym'])
panel.Add(self.wids['edge'], dcol=3)
panel.Add(CopyBtn('atsym'), dcol=1, style=RIGHT)
add_text('E0 : ')
panel.Add(xas_e0)
panel.Add(e0panel, dcol=3)
panel.Add(CopyBtn('xas_e0'), dcol=1, style=RIGHT)
add_text('Edge Step: ')
panel.Add(xas_step)
panel.Add(self.wids['auto_step'], dcol=3)
panel.Add(CopyBtn('xas_step'), dcol=1, style=RIGHT)
panel.Add((5, 5), newrow=True)
panel.Add(HLine(panel, size=(HLINEWID, 3)), dcol=6, newrow=True)
add_text('Pre-edge range: ')
panel.Add(xas_pre1)
add_text(' : ', newrow=False)
panel.Add(xas_pre2, dcol=2)
panel.Add(CopyBtn('xas_pre'), dcol=1, style=RIGHT)
panel.Add(SimpleText(panel, 'Victoreen order:'), newrow=True)
panel.Add(self.wids['nvict'], dcol=4)
panel.Add((5, 5), newrow=True)
panel.Add(HLine(panel, size=(HLINEWID, 3)), dcol=6, newrow=True)
add_text('Normalization method: ')
panel.Add(self.wids['norm_method'], dcol=4)
panel.Add(CopyBtn('xas_norm'), dcol=1, style=RIGHT)
add_text('Normalization range: ')
panel.Add(xas_norm1)
add_text(' : ', newrow=False)
panel.Add(xas_norm2, dcol=2)
panel.Add(SimpleText(panel, 'Polynomial Type:'), newrow=True)
panel.Add(self.wids['nnorm'], dcol=4)
panel.Add(HLine(panel, size=(HLINEWID, 3)), dcol=6, newrow=True)
panel.Add((5, 5), newrow=True)
panel.Add(self.wids['is_frozen'], newrow=True)
panel.Add(saveconf, dcol=5)
panel.Add((5, 5), newrow=True)
panel.Add(HLine(panel, size=(HLINEWID, 3)), dcol=6, newrow=True)
panel.pack()
sizer = wx.BoxSizer(wx.VERTICAL)
sizer.Add((5, 5), 0, LEFT, 3)
sizer.Add(panel, 0, LEFT, 3)
sizer.Add((5, 5), 0, LEFT, 3)
pack(self, sizer)
class PrePeakPanel(wx.Panel):
def __init__(self, parent=None, controller=None, **kws):
wx.Panel.__init__(self, parent, -1, size=(550, 625), **kws)
self.parent = parent
self.controller = controller
self.larch = controller.larch
self.fit_components = OrderedDict()
self.fit_model = None
self.fit_params = None
self.user_added_params = None
self.summary = None
self.sizer = wx.GridBagSizer(10, 6)
self.build_display()
self.pick2_timer = wx.Timer(self)
self.pick2_group = None
self.Bind(wx.EVT_TIMER, self.onPick2Timer, self.pick2_timer)
self.pick2_t0 = 0.
self.pick2_timeout = 15.
self.pick2erase_timer = wx.Timer(self)
self.pick2erase_panel = None
self.Bind(wx.EVT_TIMER, self.onPick2EraseTimer, self.pick2erase_timer)
def onPanelExposed(self, **kws):
# called when notebook is selected
try:
fname = self.controller.filelist.GetStringSelection()
gname = self.controller.file_groups[fname]
dgroup = self.controller.get_group(gname)
# print(" Fill prepeak panel from group ", fname, gname, dgroup)
self.fill_form(dgroup)
except:
pass # print(" Cannot Fill prepeak panel from group ")
def larch_eval(self, cmd):
"""eval"""
self.controller.larch.eval(cmd)
def build_display(self):
self.mod_nb = flat_nb.FlatNotebook(self, -1, agwStyle=FNB_STYLE)
self.mod_nb.SetTabAreaColour(wx.Colour(250,250,250))
self.mod_nb.SetActiveTabColour(wx.Colour(254,254,195))
self.mod_nb.SetNonActiveTabTextColour(wx.Colour(10,10,128))
self.mod_nb.SetActiveTabTextColour(wx.Colour(128,0,0))
self.mod_nb.Bind(wx.EVT_NOTEBOOK_PAGE_CHANGED, self.onNBChanged)
pan = self.panel = GridPanel(self, ncols=4, nrows=4, pad=2, itemstyle=LCEN)
self.wids = {}
def FloatSpinWithPin(name, value, **kws):
s = wx.BoxSizer(wx.HORIZONTAL)
self.wids[name] = FloatSpin(pan, value=value, **kws)
bb = BitmapButton(pan, get_icon('pin'), size=(25, 25),
action=partial(self.onSelPoint, opt=name),
tooltip='use last point selected from plot')
s.Add(self.wids[name])
s.Add(bb)
return s
opts = dict(digits=2, increment=0.1)
ppeak_e0 = FloatSpinWithPin('ppeak_e0', value=0, **opts)
ppeak_elo = FloatSpinWithPin('ppeak_elo', value=-15, **opts)
ppeak_ehi = FloatSpinWithPin('ppeak_ehi', value=-5, **opts)
ppeak_emin = FloatSpinWithPin('ppeak_emin', value=-30, **opts)
ppeak_emax = FloatSpinWithPin('ppeak_emax', value=0, **opts)
self.fitbline_btn = Button(pan,'Fit Baseline', action=self.onFitBaseline,
size=(150, 25))
self.fitmodel_btn = Button(pan, 'Fit Model',
action=self.onFitModel, size=(150, 25))
self.fitsel_btn = Button(pan, 'Fit Selected Groups',
action=self.onFitSelected, size=(150, 25))
self.fitmodel_btn.Disable()
self.fitsel_btn.Disable()
self.array_choice = Choice(pan, size=(150, -1),
choices=list(Array_Choices.keys()))
self.array_choice.SetSelection(1)
models_peaks = Choice(pan, size=(150, -1),
choices=ModelChoices['peaks'],
action=self.addModel)
models_other = Choice(pan, size=(150, -1),
choices=ModelChoices['other'],
action=self.addModel)
self.plot_choice = Choice(pan, size=(150, -1),
choices=PlotChoices,
action=self.onPlot)
self.message = SimpleText(pan,
'first fit baseline, then add peaks to fit model.')
self.msg_centroid = SimpleText(pan, '----')
opts = dict(default=True, size=(75, -1), action=self.onPlot)
self.show_centroid = Check(pan, label='show?', **opts)
self.show_peakrange = Check(pan, label='show?', **opts)
self.show_fitrange = Check(pan, label='show?', **opts)
self.show_e0 = Check(pan, label='show?', **opts)
opts = dict(default=False, size=(200, -1), action=self.onPlot)
self.plot_sub_bline = Check(pan, label='Subtract Baseline?', **opts)
def add_text(text, dcol=1, newrow=True):
pan.Add(SimpleText(pan, text), dcol=dcol, newrow=newrow)
titleopts = dict(font=Font(12), colour='#AA0000')
pan.Add(SimpleText(pan, ' Pre-edge Peak Fitting', **titleopts), dcol=5)
add_text(' Run Fit:', newrow=False)
add_text('Array to fit: ')
pan.Add(self.array_choice, dcol=3)
pan.Add((10, 10))
pan.Add(self.fitbline_btn)
add_text('E0: ')
pan.Add(ppeak_e0)
pan.Add((10, 10), dcol=2)
pan.Add(self.show_e0)
pan.Add(self.fitmodel_btn)
add_text('Fit Energy Range: ')
pan.Add(ppeak_emin)
add_text(' : ', newrow=False)
pan.Add(ppeak_emax)
pan.Add(self.show_fitrange)
pan.Add(self.fitsel_btn)
t = SimpleText(pan, 'Pre-edge Peak Range: ')
t.SetToolTip('Range used as mask for background')
pan.Add(t, newrow=True)
pan.Add(ppeak_elo)
add_text(' : ', newrow=False)
pan.Add(ppeak_ehi)
pan.Add(self.show_peakrange)
add_text( 'Peak Centroid: ')
pan.Add(self.msg_centroid, dcol=3)
pan.Add(self.show_centroid, dcol=1)
# plot buttons
ts = wx.BoxSizer(wx.HORIZONTAL)
ts.Add(self.plot_choice)
ts.Add(self.plot_sub_bline)
pan.Add(SimpleText(pan, 'Plot: '), newrow=True)
pan.Add(ts, dcol=7)
# add model
ts = wx.BoxSizer(wx.HORIZONTAL)
ts.Add(models_peaks)
ts.Add(models_other)
pan.Add(SimpleText(pan, 'Add Component: '), newrow=True)
pan.Add(ts, dcol=7)
pan.Add(SimpleText(pan, 'Messages: '), newrow=True)
pan.Add(self.message, dcol=7)
pan.pack()
sizer = wx.BoxSizer(wx.VERTICAL)
sizer.Add((5,5), 0, LCEN, 3)
sizer.Add(HLine(self, size=(550, 2)), 0, LCEN, 3)
sizer.Add(pan, 0, LCEN, 3)
sizer.Add((5,5), 0, LCEN, 3)
sizer.Add(HLine(self, size=(550, 2)), 0, LCEN, 3)
sizer.Add((5,5), 0, LCEN, 3)
sizer.Add(self.mod_nb, 1, LCEN|wx.GROW, 10)
pack(self, sizer)
def get_config(self, dgroup=None):
"""get processing configuration for a group"""
if dgroup is None:
dgroup = self.controller.get_group()
conf = getattr(dgroup, 'prepeak_config', {})
if 'e0' not in conf:
conf = dict(e0 = dgroup.e0, elo=-10, ehi=-5,
emin=-40, emax=0, yarray='norm')
dgroup.prepeak_config = conf
if not hasattr(dgroup, 'prepeaks'):
dgroup.prepeaks = Group()
return conf
def fill_form(self, dat):
if isinstance(dat, Group):
self.wids['ppeak_e0'].SetValue(dat.e0)
if hasattr(dat, 'prepeaks'):
self.wids['ppeak_emin'].SetValue(dat.prepeaks.emin)
self.wids['ppeak_emax'].SetValue(dat.prepeaks.emax)
self.wids['ppeak_elo'].SetValue(dat.prepeaks.elo)
self.wids['ppeak_ehi'].SetValue(dat.prepeaks.ehi)
elif instance(dat, dict):
self.wids['ppeak_e0'].SetValue(dat['e0'])
self.wids['ppeak_emin'].SetValue(dat['emin'])
self.wids['ppeak_emax'].SetValue(dat['emax'])
self.wids['ppeak_elo'].SetValue(dat['elo'])
self.wids['ppeak_ehi'].SetValue(dat['ehi'])
self.array_choice.SetStringSelection(dat['array_desc'])
self.show_e0.Enable(dat['show_e0'])
self.show_centroid.Enable(dat['show_centroid'])
self.show_fitrange.Enable(dat['show_fitrange'])
self.show_peakrange.Enable(dat['show_peakrange'])
self.plot_sub_bline.Enable(dat['plot_sub_bline'])
def read_form(self):
"read for, returning dict of values"
dgroup = self.controller.get_group()
array_desc = self.array_choice.GetStringSelection()
form_opts = {'gname': dgroup.groupname,
'array_desc': array_desc.lower(),
'array_name': Array_Choices[array_desc],
'baseline_form': 'lorentzian'}
form_opts['e0'] = self.wids['ppeak_e0'].GetValue()
form_opts['emin'] = self.wids['ppeak_emin'].GetValue()
form_opts['emax'] = self.wids['ppeak_emax'].GetValue()
form_opts['elo'] = self.wids['ppeak_elo'].GetValue()
form_opts['ehi'] = self.wids['ppeak_ehi'].GetValue()
form_opts['plot_sub_bline'] = self.plot_sub_bline.IsChecked()
form_opts['show_centroid'] = self.show_centroid.IsChecked()
form_opts['show_peakrange'] = self.show_peakrange.IsChecked()
form_opts['show_fitrange'] = self.show_fitrange.IsChecked()
form_opts['show_e0'] = self.show_e0.IsChecked()
return form_opts
def onFitBaseline(self, evt=None):
opts = self.read_form()
cmd = """{gname:s}.ydat = 1.0*{gname:s}.{array_name:s}
pre_edge_baseline(energy={gname:s}.energy, norm={gname:s}.ydat, group={gname:s},
form='{baseline_form:s}', with_line=True,
elo={elo:.3f}, ehi={ehi:.3f}, emin={emin:.3f}, emax={emax:.3f})
"""
self.larch_eval(cmd.format(**opts))
dgroup = self.controller.get_group()
ppeaks = dgroup.prepeaks
dgroup.centroid_msg = "%.4f +/- %.4f eV" % (ppeaks.centroid,
ppeaks.delta_centroid)
self.msg_centroid.SetLabel(dgroup.centroid_msg)
if 'loren_' not in self.fit_components:
self.addModel(model='Lorentzian', prefix='loren_', isbkg=True)
if 'line_' not in self.fit_components:
self.addModel(model='Linear', prefix='line_', isbkg=True)
for prefix in ('loren_', 'line_'):
cmp = self.fit_components[prefix]
# cmp.bkgbox.SetValue(1)
self.fill_model_params(prefix, dgroup.prepeaks.fit_details.params)
self.fill_form(dgroup)
self.fitmodel_btn.Enable()
# self.fitallmodel_btn.Enable()
i1, i2 = self.get_xranges(dgroup.energy)
dgroup.yfit = dgroup.xfit = 0.0*dgroup.energy[i1:i2]
self.plot_choice.SetStringSelection(PLOT_BASELINE)
self.onPlot()
def fill_model_params(self, prefix, params):
comp = self.fit_components[prefix]
parwids = comp.parwids
for pname, par in params.items():
pname = prefix + pname
if pname in parwids:
wids = parwids[pname]
if wids.minval is not None:
wids.minval.SetValue(par.min)
if wids.maxval is not None:
wids.maxval.SetValue(par.max)
wids.value.SetValue(par.value)
varstr = 'vary' if par.vary else 'fix'
if par.expr is not None:
varstr = 'constrain'
if wids.vary is not None:
wids.vary.SetStringSelection(varstr)
def onPlot(self, evt=None):
plot_choice = self.plot_choice.GetStringSelection()
opts = self.read_form()
dgroup = self.controller.get_group()
ppeaks = getattr(dgroup, 'prepeaks', None)
if ppeaks is None:
return
i1, i2 = self.get_xranges(dgroup.xdat)
# i2 = len(ppeaks.baseline) + i1
if len(dgroup.yfit) > len(ppeaks.baseline):
i2 = i1 + len(ppeaks.baseline)
# print(" Indexes: ", i1, i2, i2-i1, len(dgroup.yfit), len(ppeaks.baseline))
xdat = 1.0*dgroup.energy
ydat = 1.0*dgroup.ydat
yfit = 1.0*dgroup.ydat
baseline = 1.0*dgroup.ydat
yfit[i1:i2] = dgroup.yfit[:i2-i1]
baseline[i1:i2] = ppeaks.baseline[:i2-i1]
if opts['plot_sub_bline']:
ydat = ydat - baseline
if plot_choice in (PLOT_FIT, PLOT_RESID):
yfit = yfit - baseline
if plot_choice == PLOT_RESID:
resid = ydat - yfit
_xs = dgroup.energy[i1:i2]
xrange = max(_xs) - min(_xs)
pxmin = min(_xs) - 0.05 * xrange
pxmax = max(_xs) + 0.05 * xrange
jmin = index_of(dgroup.energy, pxmin)
jmax = index_of(dgroup.energy, pxmax) + 1
_ys = ydat[jmin:jmax]
yrange = max(_ys) - min(_ys)
pymin = min(_ys) - 0.05 * yrange
pymax = max(_ys) + 0.05 * yrange
title = ' pre-edge fit'
if plot_choice == PLOT_BASELINE:
title = ' pre-edge baseline'
if opts['plot_sub_bline']:
title = ' pre-edge peaks'
array_desc = self.array_choice.GetStringSelection()
plotopts = {'xmin': pxmin, 'xmax': pxmax,
'ymin': pymin, 'ymax': pymax,
'title': '%s: %s' % (opts['gname'], title),
'xlabel': 'Energy (eV)',
'ylabel': '%s $\mu$' % opts['array_desc'],
'label': '%s $\mu$' % opts['array_desc'],
'delay_draw': True,
'show_legend': True}
plot_extras = []
if opts['show_fitrange']:
popts = {'color': '#DDDDCC'}
emin = opts['emin']
emax = opts['emax']
imin = index_of(dgroup.energy, emin)
imax = index_of(dgroup.energy, emax)
plot_extras.append(('vline', emin, None, popts))
plot_extras.append(('vline', emax, None, popts))
if opts['show_peakrange']:
popts = {'marker': '+', 'markersize': 6}
elo = opts['elo']
ehi = opts['ehi']
ilo = index_of(dgroup.xdat, elo)
ihi = index_of(dgroup.xdat, ehi)
plot_extras.append(('marker', elo, ydat[ilo], popts))
plot_extras.append(('marker', ehi, ydat[ihi], popts))
if opts['show_centroid']:
popts = {'color': '#EECCCC'}
ecen = getattr(dgroup.prepeaks, 'centroid', -1)
if ecen > min(dgroup.energy):
plot_extras.append(('vline', ecen, None, popts))
pframe = self.controller.get_display(win=2,
stacked=(plot_choice==PLOT_RESID))
ppanel = pframe.panel
axes = ppanel.axes
plotopts.update(PLOTOPTS_1)
ppanel.plot(xdat, ydat, **plotopts)
if plot_choice == PLOT_BASELINE:
if not opts['plot_sub_bline']:
ppanel.oplot(dgroup.prepeaks.energy,
dgroup.prepeaks.baseline,
label='baseline', **PLOTOPTS_2)
elif plot_choice in (PLOT_FIT, PLOT_RESID):
ppanel.oplot(dgroup.energy, yfit,
label='fit', **PLOTOPTS_1)
if hasattr(dgroup, 'ycomps'):
ncomp = len(dgroup.ycomps)
icomp = 0
for label, ycomp in dgroup.ycomps.items():
icomp +=1
fcomp = self.fit_components[label]
# print("ycomp: ", plot_choice, label, len(ycomp), len(dgroup.xfit),
# fcomp.bkgbox.IsChecked(), opts['plot_sub_bline'], icomp, ncomp)
if not (fcomp.bkgbox.IsChecked() and opts['plot_sub_bline']):
ppanel.oplot(dgroup.xfit, ycomp, label=label,
delay_draw=(icomp!=ncomp), style='short dashed')
if plot_choice == PLOT_RESID:
_ys = resid
yrange = max(_ys) - min(_ys)
plotopts['ymin'] = min(_ys) - 0.05 * yrange
plotopts['ymax'] = max(_ys) + 0.05 * yrange
plotopts['delay_draw'] = False
plotopts['ylabel'] = 'data-fit'
plotopts['label'] = '_nolegend_'
pframe.plot(dgroup.energy, resid, panel='bot', **plotopts)
pframe.Show()
# print(" RESIDUAL PLOT margins: ")
# print(" top : ", pframe.panel.conf.margins)
# print(" bot : ", pframe.panel_bot.conf.margins)
for etype, x, y, opts in plot_extras:
if etype == 'marker':
popts = {'marker': 'o', 'markersize': 4,
'label': '_nolegend_',
'markerfacecolor': 'red',
'markeredgecolor': '#884444'}
popts.update(opts)
axes.plot([x], [y], **popts)
elif etype == 'vline':
popts = {'ymin': 0, 'ymax': 1.0, 'color': '#888888',
'label': '_nolegend_'}
popts.update(opts)
axes.axvline(x, **popts)
ppanel.canvas.draw()
def onNBChanged(self, event=None):
idx = self.mod_nb.GetSelection()
def addModel(self, event=None, model=None, prefix=None, isbkg=False):
if model is None and event is not None:
model = event.GetString()
if model is None or model.startswith('<'):
return
if prefix is None:
p = model[:5].lower()
curmodels = ["%s%i_" % (p, i+1) for i in range(1+len(self.fit_components))]
for comp in self.fit_components:
if comp in curmodels:
curmodels.remove(comp)
prefix = curmodels[0]
label = "%s(prefix='%s')" % (model, prefix)
title = "%s: %s " % (prefix[:-1], model)
title = prefix[:-1]
mclass_kws = {'prefix': prefix}
if 'step' in model.lower():
form = model.lower().replace('step', '').strip()
if form.startswith('err'): form = 'erf'
label = "Step(form='%s', prefix='%s')" % (form, prefix)
title = "%s: Step %s" % (prefix[:-1], form[:3])
mclass = lm_models.StepModel
mclass_kws['form'] = form
minst = mclass(form=form, prefix=prefix)
else:
if model in ModelFuncs:
mclass = getattr(lm_models, ModelFuncs[model])
else:
mclass = getattr(lm_models, model+'Model')
minst = mclass(prefix=prefix)
panel = GridPanel(self.mod_nb, ncols=2, nrows=5, pad=2, itemstyle=CEN)
def SLabel(label, size=(80, -1), **kws):
return SimpleText(panel, label,
size=size, style=wx.ALIGN_LEFT, **kws)
usebox = Check(panel, default=True, label='Use in Fit?', size=(100, -1))
bkgbox = Check(panel, default=False, label='Is Baseline?', size=(125, -1))
if isbkg:
bkgbox.SetValue(1)
delbtn = Button(panel, 'Delete Component', size=(125, -1),
action=partial(self.onDeleteComponent, prefix=prefix))
pick2msg = SimpleText(panel, " ", size=(125, -1))
pick2btn = Button(panel, 'Pick Values from Data', size=(150, -1),
action=partial(self.onPick2Points, prefix=prefix))
# SetTip(mname, 'Label for the model component')
SetTip(usebox, 'Use this component in fit?')
SetTip(bkgbox, 'Label this component as "background" when plotting?')
SetTip(delbtn, 'Delete this model component')
SetTip(pick2btn, 'Select X range on Plot to Guess Initial Values')
panel.Add(SLabel(label, size=(275, -1), colour='#0000AA'),
dcol=3, style=wx.ALIGN_LEFT, newrow=True)
panel.Add(usebox, dcol=1)
panel.Add(bkgbox, dcol=2, style=LCEN)
panel.Add(delbtn, dcol=1, style=wx.ALIGN_LEFT)
panel.Add(pick2btn, dcol=2, style=wx.ALIGN_LEFT, newrow=True)
panel.Add(pick2msg, dcol=2, style=wx.ALIGN_RIGHT)
# panel.Add((10, 10), newrow=True)
# panel.Add(HLine(panel, size=(150, 3)), dcol=4, style=wx.ALIGN_CENTER)
panel.Add(SLabel("Parameter "), style=wx.ALIGN_LEFT, newrow=True)
panel.AddMany((SLabel(" Value"), SLabel(" Type"), SLabel(' Bounds'),
SLabel(" Min", size=(60, -1)),
SLabel(" Max", size=(60, -1)), SLabel(" Expression")))
parwids = OrderedDict()
parnames = sorted(minst.param_names)
for a in minst._func_allargs:
pname = "%s%s" % (prefix, a)
if (pname not in parnames and
a in minst.param_hints and
a not in minst.independent_vars):
parnames.append(pname)
for pname in parnames:
sname = pname[len(prefix):]
hints = minst.param_hints.get(sname, {})
par = Parameter(name=pname, value=0, vary=True)
if 'min' in hints:
par.min = hints['min']
if 'max' in hints:
par.max = hints['max']
if 'value' in hints:
par.value = hints['value']
if 'expr' in hints:
par.expr = hints['expr']
pwids = ParameterWidgets(panel, par, name_size=100, expr_size=125,
float_size=80, prefix=prefix,
widgets=('name', 'value', 'minval',
'maxval', 'vary', 'expr'))
parwids[par.name] = pwids
panel.Add(pwids.name, newrow=True)
panel.AddMany((pwids.value, pwids.vary, pwids.bounds,
pwids.minval, pwids.maxval, pwids.expr))
for sname, hint in minst.param_hints.items():
pname = "%s%s" % (prefix, sname)
if 'expr' in hint and pname not in parnames:
par = Parameter(name=pname, value=0, expr=hint['expr'])
pwids = ParameterWidgets(panel, par, name_size=100, expr_size=225,
float_size=80, prefix=prefix,
widgets=('name', 'value', 'expr'))
parwids[par.name] = pwids
panel.Add(pwids.name, newrow=True)
panel.Add(pwids.value)
panel.Add(pwids.expr, dcol=4, style=wx.ALIGN_RIGHT)
pwids.value.Disable()
fgroup = Group(prefix=prefix, title=title, mclass=mclass,
mclass_kws=mclass_kws, usebox=usebox, panel=panel,
parwids=parwids, float_size=65, expr_size=150,
pick2_msg=pick2msg, bkgbox=bkgbox)
self.fit_components[prefix] = fgroup
panel.pack()
self.mod_nb.AddPage(panel, title, True)
sx,sy = self.GetSize()
self.SetSize((sx, sy+1))
self.SetSize((sx, sy))
def onDeleteComponent(self, evt=None, prefix=None):
fgroup = self.fit_components.get(prefix, None)
if fgroup is None:
return
for i in range(self.mod_nb.GetPageCount()):
if fgroup.title == self.mod_nb.GetPageText(i):
self.mod_nb.DeletePage(i)
for attr in dir(fgroup):
setattr(fgroup, attr, None)
self.fit_components.pop(prefix)
# sx,sy = self.GetSize()
# self.SetSize((sx, sy+1))
# self.SetSize((sx, sy))
def onPick2EraseTimer(self, evt=None):
"""erases line trace showing automated 'Pick 2' guess """
self.pick2erase_timer.Stop()
panel = self.pick2erase_panel
ntrace = panel.conf.ntrace - 1
trace = panel.conf.get_mpl_line(ntrace)
panel.conf.get_mpl_line(ntrace).set_data(np.array([]), np.array([]))
panel.conf.ntrace = ntrace
panel.draw()
def onPick2Timer(self, evt=None):
"""checks for 'Pick 2' events, and initiates 'Pick 2' guess
for a model from the selected data range
"""
try:
plotframe = self.controller.get_display()
curhist = plotframe.cursor_hist[:]
plotframe.Raise()
except:
return
if (time.time() - self.pick2_t0) > self.pick2_timeout:
msg = self.pick2_group.pick2_msg.SetLabel(" ")
plotframe.cursor_hist = []
self.pick2_timer.Stop()
return
if len(curhist) < 2:
self.pick2_group.pick2_msg.SetLabel("%i/2" % (len(curhist)))
return
self.pick2_group.pick2_msg.SetLabel("done.")
self.pick2_timer.Stop()
# guess param values
xcur = (curhist[0][0], curhist[1][0])
xmin, xmax = min(xcur), max(xcur)
dgroup = getattr(self.larch.symtable, self.controller.groupname)
x, y = dgroup.xdat, dgroup.ydat
i0 = index_of(dgroup.xdat, xmin)
i1 = index_of(dgroup.xdat, xmax)
x, y = dgroup.xdat[i0:i1+1], dgroup.ydat[i0:i1+1]
mod = self.pick2_group.mclass(prefix=self.pick2_group.prefix)
parwids = self.pick2_group.parwids
try:
guesses = mod.guess(y, x=x)
except:
return
for name, param in guesses.items():
if name in parwids:
parwids[name].value.SetValue(param.value)
dgroup._tmp = mod.eval(guesses, x=dgroup.xdat)
plotframe = self.controller.get_display()
plotframe.cursor_hist = []
plotframe.oplot(dgroup.xdat, dgroup._tmp)
self.pick2erase_panel = plotframe.panel
self.pick2erase_timer.Start(5000)
def onPick2Points(self, evt=None, prefix=None):
fgroup = self.fit_components.get(prefix, None)
if fgroup is None:
return
plotframe = self.controller.get_display()
plotframe.Raise()
plotframe.cursor_hist = []
fgroup.npts = 0
self.pick2_group = fgroup
if fgroup.pick2_msg is not None:
fgroup.pick2_msg.SetLabel("0/2")
self.pick2_t0 = time.time()
self.pick2_timer.Start(250)
def onSaveFitResult(self, event=None):
dgroup = self.controller.get_group()
deffile = dgroup.filename.replace('.', '_') + '.modl'
outfile = FileSave(self, 'Save Fit Result',
default_file=deffile,
wildcard=ModelWcards)
if outfile is not None:
try:
self.save_fit_result(dgroup.fit_history[-1], outfile)
except IOError:
print('could not write %s' % outfile)
def onLoadFitResult(self, event=None):
mfile = FileOpen(self, 'Load Fit Result',
default_file='', wildcard=ModelWcards)
if mfile is not None:
self.load_modelresult(mfile)
def save_fit_result(self, fitresult, outfile):
"""saves a customized ModelResult"""
save_modelresult(fitresult, outfile)
def load_modelresult(self, inpfile):
"""read a customized ModelResult"""
result = load_modelresult(inpfile)
for prefix in list(self.fit_components.keys()):
self.onDeleteComponent(self, prefix=prefix)
for comp in result.model.components:
isbkg = comp.prefix in result.user_options['bkg_components']
self.addModel(model=comp.func.__name__,
prefix=comp.prefix, isbkg=isbkg)
for comp in result.model.components:
parwids = self.fit_components[comp.prefix].parwids
for pname, par in result.params.items():
if pname in parwids:
wids = parwids[pname]
if wids.minval is not None:
wids.minval.SetValue(par.min)
if wids.maxval is not None:
wids.maxval.SetValue(par.max)
val = result.init_values.get(pname, par.value)
wids.value.SetValue(val)
self.fill_form(result.user_options)
return result
def onExportFitResult(self, event=None):
dgroup = self.controller.get_group()
deffile = dgroup.filename.replace('.', '_') + '_result.xdi'
wcards = 'All files (*.*)|*.*'
outfile = FileSave(self, 'Export Fit Result',
default_file=deffile, wildcard=wcards)
if outfile is not None:
i1, i2 = self.get_xranges(dgroup.xdat)
x = dgroup.xdat[i1:i2]
y = dgroup.ydat[i1:i2]
yerr = None
if hasattr(dgroup, 'yerr'):
yerr = 1.0*dgroup.yerr
if not isinstance(yerr, np.ndarray):
yerr = yerr * np.ones(len(y))
else:
yerr = yerr[i1:i2]
export_modelresult(dgroup.fit_history[-1],
filename=outfile,
datafile=dgroup.filename,
ydata=y, yerr=yerr, x=x)
def onSelPoint(self, evt=None, opt='__', relative_e0=False, win=None):
"""
get last selected point from a specified plot window
and fill in the value for the widget defined by `opt`.
by default it finds the latest cursor position from the
cursor history of the first 20 plot windows.
"""
if opt not in self.wids:
return None
_x, _y = last_cursor_pos(win=win, _larch=self.larch)
if _x is not None:
if relative_e0 and 'e0' in self.wids:
_x -= self.wids['e0'].GetValue()
self.wids[opt].SetValue(_x)
def get_xranges(self, x):
opts = self.read_form()
dgroup = self.controller.get_group()
en_eps = min(np.diff(dgroup.energy)) / 5.
i1 = index_of(x, opts['emin'] + en_eps)
i2 = index_of(x, opts['emax'] + en_eps) + 1
return i1, i2
def build_fitmodel(self):
""" use fit components to build model"""
dgroup = self.controller.get_group()
fullmodel = None
params = Parameters()
self.summary = {'components': [], 'options': {}}
peaks = []
for comp in self.fit_components.values():
_cen, _amp = None, None
if comp.usebox is not None and comp.usebox.IsChecked():
for parwids in comp.parwids.values():
params.add(parwids.param)
#print(" add param ", parwids.param)
if parwids.param.name.endswith('_center'):
_cen = parwids.param.name
elif parwids.param.name.endswith('_amplitude'):
_amp = parwids.param.name
self.summary['components'].append((comp.mclass.__name__, comp.mclass_kws))
thismodel = comp.mclass(**comp.mclass_kws)
if fullmodel is None:
fullmodel = thismodel
else:
fullmodel += thismodel
if not comp.bkgbox.IsChecked() and _cen is not None and _amp is not None:
peaks.append((_amp, _cen))
if len(peaks) > 0:
denom = '+'.join([p[0] for p in peaks])
numer = '+'.join(["%s*%s "% p for p in peaks])
params.add('fit_centroid', expr="(%s)/(%s)" %(numer, denom))
self.fit_model = fullmodel
self.fit_params = params
if dgroup is not None:
i1, i2 = self.get_xranges(dgroup.xdat)
xsel = dgroup.xdat[i1:i2]
dgroup.xfit = xsel
dgroup.yfit = self.fit_model.eval(self.fit_params, x=xsel)
dgroup.ycomps = self.fit_model.eval_components(params=self.fit_params,
x=xsel)
return dgroup
def onFitSelected(self, event=None):
dgroup = self.build_fitmodel()
opts = self.read_form()
print("fitting selected groups in progress")
def onFitModel(self, event=None):
dgroup = self.build_fitmodel()
opts = self.read_form()
i1, i2 = self.get_xranges(dgroup.xdat)
dgroup.xfit = dgroup.xdat[i1:i2]
ysel = dgroup.ydat[i1:i2]
# print('onFit Model : xrange ', i1, i2, len(dgroup.xfit), len(dgroup.yfit))
weights = np.ones(len(ysel))
if hasattr(dgroup, 'yerr'):
yerr = 1.0*dgroup.yerr
if not isinstance(yerr, np.ndarray):
yerr = yerr * np.ones(len(ysel))
else:
yerr = yerr[i1:i2]
yerr_min = 1.e-9*ysel.mean()
yerr[np.where(yerr < yerr_min)] = yerr_min
weights = 1.0/yerr
result = self.fit_model.fit(ysel, params=self.fit_params,
x=dgroup.xfit, weights=weights,
method='leastsq')
self.summary['xmin'] = dgroup.xdat[i1]
self.summary['xmax'] = dgroup.xdat[i2]
for attr in ('aic', 'bic', 'chisqr', 'redchi', 'ci_out', 'covar',
'flatchain', 'success', 'nan_policy', 'nfev', 'ndata',
'nfree', 'nvarys', 'init_values'):
self.summary[attr] = getattr(result, attr)
self.summary['params'] = result.params
dgroup.yfit = result.best_fit
dgroup.ycomps = self.fit_model.eval_components(params=result.params,
x=dgroup.xfit)
result.model_repr = self.fit_model._reprstring(long=True)
## hacks to save user options
result.user_options = opts
bkg_comps = []
for label, comp in self.fit_components.items():
if comp.bkgbox.IsChecked():
bkg_comps.append(label)
result.user_options['bkg_components'] = bkg_comps
self.autosave_modelresult(result)
if not hasattr(dgroup, 'fit_history'):
dgroup.fit_history = []
dgroup.fit_history.append(result)
self.plot_choice.SetStringSelection(PLOT_FIT)
self.onPlot()
self.parent.show_subframe('prepeak_result_frame', FitResultFrame,
datagroup=dgroup, peakframe=self)
self.parent.subframes['prepeak_result_frame'].show_fitresult()
[m.Enable(True) for m in self.parent.afterfit_menus]
def update_start_values(self, params):
"""fill parameters with best fit values"""
allparwids = {}
for comp in self.fit_components.values():
if comp.usebox is not None and comp.usebox.IsChecked():
for name, parwids in comp.parwids.items():
allparwids[name] = parwids
for pname, par in params.items():
if pname in allparwids:
allparwids[pname].value.SetValue(par.value)
def autosave_modelresult(self, result, fname=None):
"""autosave model result to user larch folder"""
xasguidir = os.path.join(site_config.usr_larchdir, 'xasgui')
if not os.path.exists(xasguidir):
try:
os.makedirs(xasguidir)
except OSError:
print("Warning: cannot create XAS GUI user folder")
return
if not HAS_MODELSAVE:
print("Warning: cannot save model results: upgrade lmfit")
return
if fname is None:
fname = 'autosave.fitresult'
fname = os.path.join(xasguidir, fname)
self.save_fit_result(result, fname)
def build_display(self):
self.mod_nb = flat_nb.FlatNotebook(self, -1, agwStyle=FNB_STYLE)
self.mod_nb.SetTabAreaColour(wx.Colour(250,250,250))
self.mod_nb.SetActiveTabColour(wx.Colour(254,254,195))
self.mod_nb.SetNonActiveTabTextColour(wx.Colour(10,10,128))
self.mod_nb.SetActiveTabTextColour(wx.Colour(128,0,0))
self.mod_nb.Bind(wx.EVT_NOTEBOOK_PAGE_CHANGED, self.onNBChanged)
pan = self.panel = GridPanel(self, ncols=4, nrows=4, pad=2, itemstyle=LCEN)
self.wids = {}
def FloatSpinWithPin(name, value, **kws):
s = wx.BoxSizer(wx.HORIZONTAL)
self.wids[name] = FloatSpin(pan, value=value, **kws)
bb = BitmapButton(pan, get_icon('pin'), size=(25, 25),
action=partial(self.onSelPoint, opt=name),
tooltip='use last point selected from plot')
s.Add(self.wids[name])
s.Add(bb)
return s
opts = dict(digits=2, increment=0.1)
ppeak_e0 = FloatSpinWithPin('ppeak_e0', value=0, **opts)
ppeak_elo = FloatSpinWithPin('ppeak_elo', value=-15, **opts)
ppeak_ehi = FloatSpinWithPin('ppeak_ehi', value=-5, **opts)
ppeak_emin = FloatSpinWithPin('ppeak_emin', value=-30, **opts)
ppeak_emax = FloatSpinWithPin('ppeak_emax', value=0, **opts)
self.fitbline_btn = Button(pan,'Fit Baseline', action=self.onFitBaseline,
size=(150, 25))
self.fitmodel_btn = Button(pan, 'Fit Model',
action=self.onFitModel, size=(150, 25))
self.fitsel_btn = Button(pan, 'Fit Selected Groups',
action=self.onFitSelected, size=(150, 25))
self.fitmodel_btn.Disable()
self.fitsel_btn.Disable()
self.array_choice = Choice(pan, size=(150, -1),
choices=list(Array_Choices.keys()))
self.array_choice.SetSelection(1)
models_peaks = Choice(pan, size=(150, -1),
choices=ModelChoices['peaks'],
action=self.addModel)
models_other = Choice(pan, size=(150, -1),
choices=ModelChoices['other'],
action=self.addModel)
self.plot_choice = Choice(pan, size=(150, -1),
choices=PlotChoices,
action=self.onPlot)
self.message = SimpleText(pan,
'first fit baseline, then add peaks to fit model.')
self.msg_centroid = SimpleText(pan, '----')
opts = dict(default=True, size=(75, -1), action=self.onPlot)
self.show_centroid = Check(pan, label='show?', **opts)
self.show_peakrange = Check(pan, label='show?', **opts)
self.show_fitrange = Check(pan, label='show?', **opts)
self.show_e0 = Check(pan, label='show?', **opts)
opts = dict(default=False, size=(200, -1), action=self.onPlot)
self.plot_sub_bline = Check(pan, label='Subtract Baseline?', **opts)
def add_text(text, dcol=1, newrow=True):
pan.Add(SimpleText(pan, text), dcol=dcol, newrow=newrow)
titleopts = dict(font=Font(12), colour='#AA0000')
pan.Add(SimpleText(pan, ' Pre-edge Peak Fitting', **titleopts), dcol=5)
add_text(' Run Fit:', newrow=False)
add_text('Array to fit: ')
pan.Add(self.array_choice, dcol=3)
pan.Add((10, 10))
pan.Add(self.fitbline_btn)
add_text('E0: ')
pan.Add(ppeak_e0)
pan.Add((10, 10), dcol=2)
pan.Add(self.show_e0)
pan.Add(self.fitmodel_btn)
add_text('Fit Energy Range: ')
pan.Add(ppeak_emin)
add_text(' : ', newrow=False)
pan.Add(ppeak_emax)
pan.Add(self.show_fitrange)
pan.Add(self.fitsel_btn)
t = SimpleText(pan, 'Pre-edge Peak Range: ')
t.SetToolTip('Range used as mask for background')
pan.Add(t, newrow=True)
pan.Add(ppeak_elo)
add_text(' : ', newrow=False)
pan.Add(ppeak_ehi)
pan.Add(self.show_peakrange)
add_text( 'Peak Centroid: ')
pan.Add(self.msg_centroid, dcol=3)
pan.Add(self.show_centroid, dcol=1)
# plot buttons
ts = wx.BoxSizer(wx.HORIZONTAL)
ts.Add(self.plot_choice)
ts.Add(self.plot_sub_bline)
pan.Add(SimpleText(pan, 'Plot: '), newrow=True)
pan.Add(ts, dcol=7)
# add model
ts = wx.BoxSizer(wx.HORIZONTAL)
ts.Add(models_peaks)
ts.Add(models_other)
pan.Add(SimpleText(pan, 'Add Component: '), newrow=True)
pan.Add(ts, dcol=7)
pan.Add(SimpleText(pan, 'Messages: '), newrow=True)
pan.Add(self.message, dcol=7)
pan.pack()
sizer = wx.BoxSizer(wx.VERTICAL)
sizer.Add((5,5), 0, LCEN, 3)
sizer.Add(HLine(self, size=(550, 2)), 0, LCEN, 3)
sizer.Add(pan, 0, LCEN, 3)
sizer.Add((5,5), 0, LCEN, 3)
sizer.Add(HLine(self, size=(550, 2)), 0, LCEN, 3)
sizer.Add((5,5), 0, LCEN, 3)
sizer.Add(self.mod_nb, 1, LCEN|wx.GROW, 10)
pack(self, sizer)
def build_display(self):
titleopts = dict(font=Font(12), colour='#AA0000')
xas = self.panel
self.wids = {}
self.plotone_op = Choice(xas,
choices=list(PlotOne_Choices.keys()),
action=self.onPlotOne,
size=(175, -1))
self.plotsel_op = Choice(xas,
choices=list(PlotSel_Choices.keys()),
action=self.onPlotSel,
size=(175, -1))
self.plotone_op.SetSelection(1)
self.plotsel_op.SetSelection(1)
plot_one = Button(xas,
'Plot This Group',
size=(150, -1),
action=self.onPlotOne)
plot_sel = Button(xas,
'Plot Selected Groups',
size=(150, -1),
action=self.onPlotSel)
opts = dict(action=self.onReprocess)
e0opts_panel = wx.Panel(xas)
self.wids['autoe0'] = Check(e0opts_panel,
default=True,
label='auto?',
**opts)
self.wids['showe0'] = Check(e0opts_panel,
default=True,
label='show?',
**opts)
sx = wx.BoxSizer(wx.HORIZONTAL)
sx.Add(self.wids['autoe0'], 0, LCEN, 4)
sx.Add(self.wids['showe0'], 0, LCEN, 4)
pack(e0opts_panel, sx)
self.wids['autostep'] = Check(xas, default=True, label='auto?', **opts)
opts['size'] = (50, -1)
self.wids['vict'] = Choice(xas, choices=('0', '1', '2', '3'), **opts)
self.wids['nnor'] = Choice(xas, choices=('0', '1', '2', '3'), **opts)
self.wids['vict'].SetSelection(1)
self.wids['nnor'].SetSelection(1)
opts.update({'size': (100, -1), 'digits': 2, 'increment': 5.0})
xas_pre1 = self.add_floatspin('pre1', value=-1000, **opts)
xas_pre2 = self.add_floatspin('pre2', value=-30, **opts)
xas_nor1 = self.add_floatspin('nor1', value=50, **opts)
xas_nor2 = self.add_floatspin('nor2', value=5000, **opts)
opts = {'digits': 2, 'increment': 0.1, 'value': 0}
xas_e0 = self.add_floatspin('e0', action=self.onSet_XASE0, **opts)
xas_step = self.add_floatspin('step',
action=self.onSet_XASStep,
with_pin=False,
**opts)
saveconf = Button(xas,
'Save as Default Settings',
size=(200, -1),
action=self.onSaveConfigBtn)
def CopyBtn(name):
return Button(xas,
'Copy',
size=(50, -1),
action=partial(self.onCopyParam, name))
add_text = self.add_text
xas.Add(SimpleText(xas, ' XAS Pre-edge subtraction and Normalization',
**titleopts),
dcol=4)
xas.Add(SimpleText(xas, 'Copy to Selected Groups?'),
style=RCEN,
dcol=3)
xas.Add(plot_sel, newrow=True)
xas.Add(self.plotsel_op, dcol=6)
xas.Add(plot_one, newrow=True)
xas.Add(self.plotone_op, dcol=4)
xas.Add((10, 10))
xas.Add(CopyBtn('plotone_op'), style=RCEN)
add_text('E0 : ')
xas.Add(xas_e0)
xas.Add(e0opts_panel, dcol=3)
xas.Add((10, 1))
xas.Add(CopyBtn('xas_e0'), style=RCEN)
add_text('Edge Step: ')
xas.Add(xas_step)
xas.Add(self.wids['autostep'], dcol=3)
xas.Add((10, 1))
xas.Add(CopyBtn('xas_step'), style=RCEN)
add_text('Pre-edge range: ')
xas.Add(xas_pre1)
add_text(' : ', newrow=False)
xas.Add(xas_pre2)
xas.Add(SimpleText(xas, 'Victoreen:'))
xas.Add(self.wids['vict'])
xas.Add(CopyBtn('xas_pre'), style=RCEN)
add_text('Normalization range: ')
xas.Add(xas_nor1)
add_text(' : ', newrow=False)
xas.Add(xas_nor2)
xas.Add(SimpleText(xas, 'Poly Order:'))
xas.Add(self.wids['nnor'])
xas.Add(CopyBtn('xas_norm'), style=RCEN)
xas.Add(saveconf, dcol=6, newrow=True)
xas.pack()
sizer = wx.BoxSizer(wx.VERTICAL)
sizer.Add((5, 5), 0, LCEN, 3)
sizer.Add(HLine(self, size=(550, 2)), 0, LCEN, 3)
sizer.Add(xas, 0, LCEN, 3)
sizer.Add((5, 5), 0, LCEN, 3)
sizer.Add(HLine(self, size=(550, 2)), 0, LCEN, 3)
pack(self, sizer)
def build_display(self):
titleopts = dict(font=Font(12), colour='#AA0000')
panel = self.panel
wids = self.wids
self.skip_process = True
wids['plotone_op'] = Choice(panel,
choices=PlotOne_Choices,
action=self.onPlotOne,
size=(175, -1))
wids['plotalt_op'] = Choice(panel,
choices=PlotAlt_Choices,
action=self.onPlotOne,
size=(175, -1))
wids['plotsel_op'] = Choice(panel,
choices=PlotSel_Choices,
action=self.onPlotSel,
size=(175, -1))
wids['plotone_op'].SetStringSelection(chik)
wids['plotsel_op'].SetStringSelection(chik)
wids['plotalt_op'].SetStringSelection(noplot)
plot_one = Button(panel,
'Plot This Group',
size=(150, -1),
action=self.onPlotOne)
plot_sel = Button(panel,
'Plot Selected Groups',
size=(150, -1),
action=self.onPlotSel)
saveconf = Button(panel,
'Save as Default Settings',
size=(200, -1),
action=self.onSaveConfigBtn)
def FloatSpinWithPin(name, value, **kws):
s = wx.BoxSizer(wx.HORIZONTAL)
self.wids[name] = FloatSpin(panel, value=value, **kws)
bb = BitmapButton(panel,
get_icon('pin'),
size=(25, 25),
action=partial(self.onSelPoint, opt=name),
tooltip='use last point selected from plot')
s.Add(wids[name])
s.Add(bb)
return s
wids['plot_voffset'] = FloatSpin(panel,
value=0,
digits=2,
increment=0.25,
action=self.onProcess)
wids['plot_kweight'] = FloatSpin(panel,
value=2,
digits=1,
increment=1,
action=self.onProcess,
min_val=0,
max_val=5)
wids['plot_kweight_alt'] = FloatSpin(panel,
value=2,
digits=1,
increment=1,
action=self.onProcess,
min_val=0,
max_val=5)
opts = dict(digits=2, increment=0.1, min_val=0, action=self.onProcess)
wids['e0'] = FloatSpin(panel, **opts)
opts['max_val'] = 5
wids['rbkg'] = FloatSpin(panel, value=1.0, **opts)
opts['max_val'] = 125
bkg_kmin = FloatSpinWithPin('bkg_kmin', value=0, **opts)
bkg_kmax = FloatSpinWithPin('bkg_kmax', value=20, **opts)
fft_kmin = FloatSpinWithPin('fft_kmin', value=0, **opts)
fft_kmax = FloatSpinWithPin('fft_kmax', value=20, **opts)
wids['fft_dk'] = FloatSpin(panel, value=3, **opts)
opts.update({'increment': 1, 'digits': 1, 'max_val': 5})
wids['bkg_kweight'] = FloatSpin(panel, value=1, **opts)
wids['fft_kweight'] = FloatSpin(panel, value=1, **opts)
opts = dict(choices=CLAMPLIST, size=(80, -1), action=self.onProcess)
wids['bkg_clamplo'] = Choice(panel, **opts)
wids['bkg_clamphi'] = Choice(panel, **opts)
wids['fft_kwindow'] = Choice(panel,
choices=list(FTWINDOWS),
action=self.onProcess,
size=(125, -1))
def add_text(text, dcol=1, newrow=True):
panel.Add(SimpleText(panel, text), dcol=dcol, newrow=newrow)
panel.Add(SimpleText(panel, ' EXAFS Processing', **titleopts), dcol=5)
panel.Add(plot_sel, newrow=True)
panel.Add(self.wids['plotsel_op'], dcol=2)
add_text('Vertical offset: ', newrow=False)
panel.Add(wids['plot_voffset'], dcol=2)
panel.Add(plot_one, newrow=True)
panel.Add(self.wids['plotone_op'], dcol=2)
add_text('Plot k weight: ', newrow=False)
panel.Add(wids['plot_kweight'])
add_text('Add Second Plot: ', newrow=True)
panel.Add(self.wids['plotalt_op'], dcol=2)
add_text('Plot2 k weight: ', newrow=False)
panel.Add(wids['plot_kweight_alt'])
panel.Add(HLine(panel, size=(500, 3)), dcol=6, newrow=True)
panel.Add(SimpleText(panel, ' Background subtraction', **titleopts),
dcol=2,
newrow=True)
panel.Add(Button(panel,
'Copy To Selected Groups',
size=(175, -1),
action=partial(self.onCopyParam, 'bkg')),
dcol=2)
add_text('R bkg: ')
panel.Add(wids['rbkg'])
add_text('E0: ', newrow=False)
panel.Add(wids['e0'])
add_text('k min: ')
panel.Add(bkg_kmin)
add_text('k max:', newrow=False)
panel.Add(bkg_kmax)
add_text('kweight: ', newrow=True)
panel.Add(wids['bkg_kweight'], dcol=1)
add_text('Clamps Low E: ', newrow=True)
panel.Add(wids['bkg_clamplo'])
add_text('high E: ', newrow=False)
panel.Add(wids['bkg_clamphi'])
panel.Add(HLine(panel, size=(500, 3)), dcol=6, newrow=True)
panel.Add(SimpleText(panel, ' Fourier transform', **titleopts),
dcol=2,
newrow=True)
panel.Add(Button(panel,
'Copy to Selected Groups',
size=(175, -1),
action=partial(self.onCopyParam, 'fft')),
dcol=2)
panel.Add(SimpleText(panel, 'k min: '), newrow=True)
panel.Add(fft_kmin)
panel.Add(SimpleText(panel, 'k max:'))
panel.Add(fft_kmax)
panel.Add(SimpleText(panel, 'k weight : '), newrow=True)
panel.Add(wids['fft_kweight'])
panel.Add(SimpleText(panel, 'K window : '), newrow=True)
panel.Add(wids['fft_kwindow'])
panel.Add(SimpleText(panel, ' dk : '))
panel.Add(wids['fft_dk'])
panel.Add(HLine(panel, size=(500, 3)), dcol=6, newrow=True)
panel.Add(saveconf, dcol=4, newrow=True)
panel.pack()
sizer = wx.BoxSizer(wx.VERTICAL)
sizer.Add(panel, 1, LCEN, 3)
pack(self, sizer)
self.skip_process = False
def build_display(self):
panel = self.panel
wids = self.wids
self.skip_process = True
wids['fitspace'] = Choice(panel,
choices=FitSpace_Choices,
size=(175, -1))
wids['fitspace'].SetStringSelection(norm)
wids['plotchoice'] = Choice(panel,
choices=Plot_Choices,
size=(175, -1),
action=self.onPlot)
wids['plotchoice'].SetSelection(0)
add_text = self.add_text
opts = dict(digits=2, increment=1.0)
w_xmin = self.add_floatspin('xmin', value=defaults['xmin'], **opts)
w_xmax = self.add_floatspin('xmax', value=defaults['xmax'], **opts)
w_wmin = self.add_floatspin('weight_min',
digits=4,
value=defaults['weight_min'],
increment=0.001,
with_pin=False,
min_val=0,
max_val=0.5,
action=self.onSet_WeightMin)
autoset_fs_increment(self.wids['weight_min'], defaults['weight_min'])
self.wids['weight_auto'] = Check(panel, default=True, label='auto?')
w_mcomps = self.add_floatspin('max_components',
digits=0,
value=defaults['max_components'],
increment=1,
with_pin=False,
min_val=0)
b_build_model = Button(panel,
'Build Model With Selected Groups',
size=(275, -1),
action=self.onBuildPCAModel)
wids['fit_group'] = Button(panel,
'Test Current Group with Model',
size=(225, -1),
action=self.onFitGroup)
wids['save_model'] = Button(panel,
'Save PCA Model',
size=(150, -1),
action=self.onSavePCAModel)
wids['load_model'] = Button(panel,
'Load PCA Model',
size=(150, -1),
action=self.onLoadPCAModel)
wids['fit_group'].Disable()
wids['load_model'].Disable()
wids['save_model'].Disable()
collabels = [
' Variance ', ' IND value ', ' IND / IND_Best', ' Factor Weight'
]
colsizes = [100, 100, 100, 200]
coltypes = ['float:12,6', 'float:12,6', 'float:12,5', 'string']
coldefs = [0.0, 0.0, 1.0, '0.0']
wids['table'] = DataTableGrid(panel,
nrows=MAX_ROWS,
collabels=collabels,
datatypes=coltypes,
defaults=coldefs,
colsizes=colsizes,
rowlabelsize=60)
wids['table'].SetMinSize((625, 175))
wids['table'].EnableEditing(False)
#
# sview = self.wids['stats'] = dv.DataViewListCtrl(panel, style=DVSTYLE)
# sview.AppendTextColumn(' Component', width=75)
# sview.AppendTextColumn(' Weight', width=100)
# sview.AppendTextColumn(' Significant?', width=75)
#
# for col in range(sview.ColumnCount):
# this = sview.Columns[col]
# align = wx.ALIGN_LEFT if col == 0 else wx.ALIGN_RIGHT
# this.Sortable = False
# this.Alignment = this.Renderer.Alignment = align
# sview.SetMinSize((275, 250))
wids['status'] = SimpleText(panel, ' ')
rfont = self.GetFont()
rfont.SetPointSize(rfont.GetPointSize() + 1)
wids['fit_chi2'] = SimpleText(panel, '0.000', font=rfont)
wids['fit_dscale'] = SimpleText(panel, '1.000', font=rfont)
panel.Add(SimpleText(panel, ' Principal Component Analysis',
**self.titleopts),
dcol=4)
add_text('Array to Use: ', newrow=True)
panel.Add(wids['fitspace'], dcol=2)
panel.Add(wids['load_model'], dcol=2)
add_text('Plot : ', newrow=True)
panel.Add(wids['plotchoice'], dcol=2)
panel.Add(wids['save_model'], dcol=2)
add_text('Fit Energy Range: ')
panel.Add(w_xmin)
add_text(' : ', newrow=False)
panel.Add(w_xmax)
#panel.Add(wids['show_fitrange'])
add_text('Min Weight: ')
panel.Add(w_wmin)
panel.Add(wids['weight_auto'])
add_text('Max Components:', dcol=1, newrow=True)
panel.Add(w_mcomps)
panel.Add(Button(panel,
'Copy To Selected Groups',
size=(200, -1),
action=partial(self.onCopyParam, 'pca')),
dcol=2)
add_text('Status: ')
panel.Add(wids['status'], dcol=3)
panel.Add(HLine(panel, size=(550, 2)), dcol=5, newrow=True)
panel.Add(b_build_model, dcol=5, newrow=True)
panel.Add(wids['table'], dcol=5, newrow=True)
panel.Add(wids['fit_group'], dcol=3, newrow=True)
add_text('chi-square: ')
panel.Add(wids['fit_chi2'])
add_text('scale factor: ')
panel.Add(wids['fit_dscale'])
#
# ## add weights report: slightly tricky layout
# ## working with GridBagSizer under gridpanel...
# icol = panel.icol - 2
# irow = panel.irow
# pstyle, ppad = panel.itemstyle, panel.pad
#
# panel.sizer.Add(SimpleText(panel, 'data scalefactor = '),
# (irow+1, icol), (1, 1), pstyle, ppad)
# panel.sizer.Add(wids['fit_dscale'],
# (irow+1, icol+1), (1, 1), pstyle, ppad)
# irow +=1
# for i in range(1, NWTS+1):
# wids['fit_wt%d' % i] = SimpleText(panel, '--', font=rfont)
# panel.sizer.Add(SimpleText(panel, 'weight_%d ='%i, font=rfont),
# (irow+i, icol), (1, 1), pstyle, ppad)
# panel.sizer.Add(wids['fit_wt%d'%i],
# (irow+i, icol+1), (1, 1), pstyle, ppad)
#
panel.pack()
sizer = wx.BoxSizer(wx.VERTICAL)
sizer.Add((10, 10), 0, LCEN, 3)
sizer.Add(panel, 1, LCEN, 3)
pack(self, sizer)
self.skip_process = False
class XASNormPanel(TaskPanel):
"""XAS normalization Panel"""
def __init__(self, parent, controller=None, **kws):
TaskPanel.__init__(self,
parent,
controller,
configname='xasnorm_config',
config=defaults,
**kws)
def build_display(self):
titleopts = dict(font=Font(12), colour='#AA0000')
xas = self.panel
self.wids = {}
self.plotone_op = Choice(xas,
choices=list(PlotOne_Choices.keys()),
action=self.onPlotOne,
size=(175, -1))
self.plotsel_op = Choice(xas,
choices=list(PlotSel_Choices.keys()),
action=self.onPlotSel,
size=(175, -1))
self.plotone_op.SetSelection(1)
self.plotsel_op.SetSelection(1)
plot_one = Button(xas,
'Plot This Group',
size=(150, -1),
action=self.onPlotOne)
plot_sel = Button(xas,
'Plot Selected Groups',
size=(150, -1),
action=self.onPlotSel)
opts = dict(action=self.onReprocess)
e0opts_panel = wx.Panel(xas)
self.wids['autoe0'] = Check(e0opts_panel,
default=True,
label='auto?',
**opts)
self.wids['showe0'] = Check(e0opts_panel,
default=True,
label='show?',
**opts)
sx = wx.BoxSizer(wx.HORIZONTAL)
sx.Add(self.wids['autoe0'], 0, LCEN, 4)
sx.Add(self.wids['showe0'], 0, LCEN, 4)
pack(e0opts_panel, sx)
self.wids['autostep'] = Check(xas, default=True, label='auto?', **opts)
opts['size'] = (50, -1)
self.wids['vict'] = Choice(xas, choices=('0', '1', '2', '3'), **opts)
self.wids['nnor'] = Choice(xas, choices=('0', '1', '2', '3'), **opts)
self.wids['vict'].SetSelection(1)
self.wids['nnor'].SetSelection(1)
opts.update({'size': (100, -1), 'digits': 2, 'increment': 5.0})
xas_pre1 = self.add_floatspin('pre1', value=-1000, **opts)
xas_pre2 = self.add_floatspin('pre2', value=-30, **opts)
xas_nor1 = self.add_floatspin('nor1', value=50, **opts)
xas_nor2 = self.add_floatspin('nor2', value=5000, **opts)
opts = {'digits': 2, 'increment': 0.1, 'value': 0}
xas_e0 = self.add_floatspin('e0', action=self.onSet_XASE0, **opts)
xas_step = self.add_floatspin('step',
action=self.onSet_XASStep,
with_pin=False,
**opts)
saveconf = Button(xas,
'Save as Default Settings',
size=(200, -1),
action=self.onSaveConfigBtn)
def CopyBtn(name):
return Button(xas,
'Copy',
size=(50, -1),
action=partial(self.onCopyParam, name))
add_text = self.add_text
xas.Add(SimpleText(xas, ' XAS Pre-edge subtraction and Normalization',
**titleopts),
dcol=4)
xas.Add(SimpleText(xas, 'Copy to Selected Groups?'),
style=RCEN,
dcol=3)
xas.Add(plot_sel, newrow=True)
xas.Add(self.plotsel_op, dcol=6)
xas.Add(plot_one, newrow=True)
xas.Add(self.plotone_op, dcol=4)
xas.Add((10, 10))
xas.Add(CopyBtn('plotone_op'), style=RCEN)
add_text('E0 : ')
xas.Add(xas_e0)
xas.Add(e0opts_panel, dcol=3)
xas.Add((10, 1))
xas.Add(CopyBtn('xas_e0'), style=RCEN)
add_text('Edge Step: ')
xas.Add(xas_step)
xas.Add(self.wids['autostep'], dcol=3)
xas.Add((10, 1))
xas.Add(CopyBtn('xas_step'), style=RCEN)
add_text('Pre-edge range: ')
xas.Add(xas_pre1)
add_text(' : ', newrow=False)
xas.Add(xas_pre2)
xas.Add(SimpleText(xas, 'Victoreen:'))
xas.Add(self.wids['vict'])
xas.Add(CopyBtn('xas_pre'), style=RCEN)
add_text('Normalization range: ')
xas.Add(xas_nor1)
add_text(' : ', newrow=False)
xas.Add(xas_nor2)
xas.Add(SimpleText(xas, 'Poly Order:'))
xas.Add(self.wids['nnor'])
xas.Add(CopyBtn('xas_norm'), style=RCEN)
xas.Add(saveconf, dcol=6, newrow=True)
xas.pack()
sizer = wx.BoxSizer(wx.VERTICAL)
sizer.Add((5, 5), 0, LCEN, 3)
sizer.Add(HLine(self, size=(550, 2)), 0, LCEN, 3)
sizer.Add(xas, 0, LCEN, 3)
sizer.Add((5, 5), 0, LCEN, 3)
sizer.Add(HLine(self, size=(550, 2)), 0, LCEN, 3)
pack(self, sizer)
def get_config(self, dgroup=None):
"""custom get_config to possibly inherit from Athena settings"""
if dgroup is None:
dgroup = self.controller.get_group()
if hasattr(dgroup, self.configname):
conf = getattr(dgroup, self.configname)
else:
conf = self.get_defaultconfig()
if hasattr(dgroup, 'bkg_params'): # from Athena
conf['e0'] = getattr(dgroup.bkg_params, 'e0', conf['e0'])
conf['pre1'] = getattr(dgroup.bkg_params, 'pre1', conf['pre1'])
conf['pre2'] = getattr(dgroup.bkg_params, 'pre2', conf['pre2'])
conf['norm1'] = getattr(dgroup.bkg_params, 'nor1',
conf['norm1'])
conf['norm2'] = getattr(dgroup.bkg_params, 'nor2',
conf['norm2'])
conf['nnorm'] = getattr(dgroup.bkg_params, 'nnor',
conf['nnorm'])
conf['nvict'] = getattr(dgroup.bkg_params, 'nvic',
conf['nvict'])
conf['autostep'] = (float(
getattr(dgroup.bkg_params, 'fixstep', 0.0)) < 0.5)
setattr(dgroup, self.configname, conf)
return conf
def fill_form(self, dgroup):
"""fill in form from a data group"""
opts = self.get_config(dgroup)
self.skip_process = True
if dgroup.datatype == 'xas':
for k in self.wids.values():
k.Enable()
self.plotone_op.SetChoices(list(PlotOne_Choices.keys()))
self.plotsel_op.SetChoices(list(PlotSel_Choices.keys()))
self.plotone_op.SetStringSelection(opts['plotone_op'])
self.plotsel_op.SetStringSelection(opts['plotsel_op'])
self.wids['e0'].SetValue(opts['e0'])
edge_step = opts.get('edge_step', None)
if edge_step is None:
edge_step = 1.0
ndigits = int(2 - round(np.log10(abs(edge_step))))
self.wids['step'].SetDigits(ndigits + 1)
self.wids['step'].SetIncrement(0.2 * 10**(-ndigits))
self.wids['step'].SetValue(edge_step)
self.wids['pre1'].SetValue(opts['pre1'])
self.wids['pre2'].SetValue(opts['pre2'])
self.wids['nor1'].SetValue(opts['norm1'])
self.wids['nor2'].SetValue(opts['norm2'])
self.wids['vict'].SetSelection(opts['nvict'])
self.wids['nnor'].SetSelection(opts['nnorm'])
self.wids['showe0'].SetValue(opts['show_e0'])
self.wids['autoe0'].SetValue(opts['auto_e0'])
self.wids['autostep'].SetValue(opts['auto_step'])
else:
self.plotone_op.SetChoices(list(PlotOne_Choices_nonxas.keys()))
self.plotsel_op.SetChoices(list(PlotSel_Choices_nonxas.keys()))
self.plotone_op.SetStringSelection('Raw Data')
self.plotsel_op.SetStringSelection('Raw Data')
for k in self.wids.values():
k.Disable()
self.skip_process = False
def read_form(self):
"read form, return dict of values"
form_opts = {}
form_opts['e0'] = self.wids['e0'].GetValue()
form_opts['edge_step'] = self.wids['step'].GetValue()
form_opts['pre1'] = self.wids['pre1'].GetValue()
form_opts['pre2'] = self.wids['pre2'].GetValue()
form_opts['norm1'] = self.wids['nor1'].GetValue()
form_opts['norm2'] = self.wids['nor2'].GetValue()
form_opts['nnorm'] = int(self.wids['nnor'].GetSelection())
form_opts['nvict'] = int(self.wids['vict'].GetSelection())
form_opts['plotone_op'] = self.plotone_op.GetStringSelection()
form_opts['plotsel_op'] = self.plotsel_op.GetStringSelection()
form_opts['show_e0'] = self.wids['showe0'].IsChecked()
form_opts['auto_e0'] = self.wids['autoe0'].IsChecked()
form_opts['auto_step'] = self.wids['autostep'].IsChecked()
return form_opts
def onPlotOne(self, evt=None):
self.plot(self.controller.get_group())
def onPlotSel(self, evt=None):
newplot = True
group_ids = self.controller.filelist.GetCheckedStrings()
if len(group_ids) < 1:
return
last_id = group_ids[-1]
yarray_name = PlotSel_Choices[self.plotsel_op.GetStringSelection()]
ylabel = getattr(plotlabels, yarray_name)
for checked in group_ids:
groupname = self.controller.file_groups[str(checked)]
dgroup = self.controller.get_group(groupname)
plot_yarrays = [(yarray_name, PLOTOPTS_1, dgroup.filename)]
if dgroup is not None:
dgroup.plot_extras = []
self.plot(dgroup,
title='',
new=newplot,
multi=True,
plot_yarrays=plot_yarrays,
show_legend=True,
with_extras=False,
delay_draw=(last_id != checked))
newplot = False
self.controller.get_display(stacked=False).panel.canvas.draw()
def onSaveConfigBtn(self, evt=None):
conf = self.get_config()
conf.update(self.read_form())
self.set_defaultconfig(conf)
def onCopyParam(self, name=None, evt=None):
conf = self.get_config()
conf.update(self.read_form())
opts = {}
name = str(name)
def copy_attrs(*args):
for a in args:
opts[a] = conf[a]
if name == 'plotone_op':
copy_attrs('plotone_op')
elif name == 'xas_e0':
copy_attrs('e0', 'show_e0', 'auto_e0')
elif name == 'xas_step':
copy_attrs('edge_step', 'auto_step')
elif name == 'xas_pre':
copy_attrs('pre1', 'pre2', 'nvict')
elif name == 'xas_norm':
copy_attrs('nnorm', 'norm1', 'norm2')
for checked in self.controller.filelist.GetCheckedStrings():
groupname = self.controller.file_groups[str(checked)]
grp = self.controller.get_group(groupname)
if grp != self.controller.group:
self.set_config(grp, opts)
self.fill_form(grp)
self.process(grp)
def onSet_XASE0(self, evt=None, value=None):
"handle setting e0"
self.wids['autoe0'].SetValue(0)
self.onReprocess()
def onSet_XASStep(self, evt=None, value=None):
"handle setting edge step"
self.wids['autostep'].SetValue(0)
self.onReprocess()
def onSelPoint(self, evt=None, opt='__', win=None):
"""
get last selected point from a specified plot window
and fill in the value for the widget defined by `opt`.
by default it finds the latest cursor position from the
cursor history of the first 20 plot windows.
"""
if opt not in self.wids:
return None
_x, _y = last_cursor_pos(win=win, _larch=self.larch)
if _x is None:
return
e0 = self.wids['e0'].GetValue()
if opt == 'e0':
self.wids['e0'].SetValue(_x)
self.wids['autoe0'].SetValue(0)
elif opt in ('pre1', 'pre2', 'nor1', 'nor2'):
self.wids[opt].SetValue(_x - e0)
self.onReprocess()
def onReprocess(self, evt=None, value=None, **kws):
"handle request reprocess"
if self.skip_process:
return
try:
dgroup = self.controller.get_group()
except TypeError:
return
self.process(dgroup=dgroup)
self.plot(dgroup)
def make_dnormde(self, dgroup):
form = dict(group=dgroup.groupname)
self.larch_eval(
"{group:s}.dnormde={group:s}.dmude/{group:s}.edge_step".format(
**form))
def process(self, dgroup=None, **kws):
""" handle process (pre-edge/normalize) of XAS data from XAS form
"""
if self.skip_process:
return
if dgroup is None:
dgroup = self.controller.get_group()
self.skip_process = True
# print("process ", dgroup, dgroup.filename)
self.get_config(dgroup)
dgroup.custom_plotopts = {}
# print("XAS norm process ", dgroup.datatype)
if dgroup.datatype != 'xas':
self.skip_process = False
dgroup.mu = dgroup.ydat * 1.0
return
en_units = getattr(dgroup, 'energy_units', None)
if en_units is None:
en_units = 'eV'
units = guess_energy_units(dgroup.energy)
if units != 'eV':
dlg = EnergyUnitsDialog(self.parent, units, dgroup.energy)
res = dlg.GetResponse()
dlg.Destroy()
if res.ok:
en_units = res.units
dgroup.xdat = dgroup.energy = res.energy
dgroup.energy_units = en_units
form = self.read_form()
e0 = form['e0']
edge_step = form['edge_step']
form['group'] = dgroup.groupname
copts = [dgroup.groupname]
if not form['auto_e0']:
if e0 < max(dgroup.energy) and e0 > min(dgroup.energy):
copts.append("e0=%.4f" % float(e0))
if not form['auto_step']:
copts.append("step=%.4f" % float(edge_step))
for attr in ('pre1', 'pre2', 'nvict', 'nnorm', 'norm1', 'norm2'):
copts.append("%s=%.2f" % (attr, form[attr]))
self.larch_eval("pre_edge(%s)" % (', '.join(copts)))
self.make_dnormde(dgroup)
if form['auto_e0']:
self.wids['e0'].SetValue(dgroup.e0) # , act=False)
if form['auto_step']:
self.wids['step'].SetValue(dgroup.edge_step) # , act=False)
self.wids['pre1'].SetValue(dgroup.pre_edge_details.pre1)
self.wids['pre2'].SetValue(dgroup.pre_edge_details.pre2)
self.wids['nor1'].SetValue(dgroup.pre_edge_details.norm1)
self.wids['nor2'].SetValue(dgroup.pre_edge_details.norm2)
conf = {}
for attr in ('e0', 'edge_step'):
conf[attr] = getattr(dgroup, attr)
for attr in ('pre1', 'pre2', 'nnorm', 'norm1', 'norm2'):
conf[attr] = getattr(dgroup.pre_edge_details, attr)
self.set_config(dgroup, conf)
self.skip_process = False
def get_plot_arrays(self, dgroup):
form = self.read_form()
lab = plotlabels.norm
if dgroup is None:
return
dgroup.plot_y2label = None
dgroup.plot_xlabel = plotlabels.energy
dgroup.plot_yarrays = [('norm', PLOTOPTS_1, lab)]
if dgroup.datatype != 'xas':
pchoice = PlotOne_Choices_nonxas[
self.plotone_op.GetStringSelection()]
dgroup.plot_xlabel = 'x'
dgroup.plot_ylabel = 'y'
dgroup.plot_yarrays = [('ydat', PLOTOPTS_1, 'ydat')]
dgroup.dmude = np.gradient(dgroup.ydat) / np.gradient(dgroup.xdat)
if pchoice == 'dmude':
dgroup.plot_ylabel = 'dy/dx'
dgroup.plot_yarrays = [('dmude', PLOTOPTS_1, 'dy/dx')]
elif pchoice == 'norm+deriv':
lab = plotlabels.norm
dgroup.plot_y2label = 'dy/dx'
dgroup.plot_yarrays = [('ydat', PLOTOPTS_1, 'y'),
('dmude', PLOTOPTS_D, 'dy/dx')]
return
pchoice = PlotOne_Choices[self.plotone_op.GetStringSelection()]
if pchoice in ('mu', 'norm', 'flat', 'dmude'):
lab = getattr(plotlabels, pchoice)
dgroup.plot_yarrays = [(pchoice, PLOTOPTS_1, lab)]
elif pchoice == 'prelines':
dgroup.plot_yarrays = [('mu', PLOTOPTS_1, plotlabels.mu),
('pre_edge', PLOTOPTS_2, 'pre edge'),
('post_edge', PLOTOPTS_2, 'post edge')]
elif pchoice == 'preedge':
dgroup.pre_edge_sub = dgroup.norm * dgroup.edge_step
dgroup.plot_yarrays = [('pre_edge_sub', PLOTOPTS_1,
r'pre-edge subtracted $\mu$')]
lab = r'pre-edge subtracted $\mu$'
elif pchoice == 'norm+deriv':
lab = plotlabels.norm
lab2 = plotlabels.dmude
dgroup.plot_yarrays = [('norm', PLOTOPTS_1, lab),
('dmude', PLOTOPTS_D, lab2)]
dgroup.plot_y2label = lab2
dgroup.plot_ylabel = lab
y4e0 = dgroup.ydat = getattr(dgroup, dgroup.plot_yarrays[0][0],
dgroup.mu)
dgroup.plot_extras = []
if form['show_e0']:
ie0 = index_of(dgroup.energy, dgroup.e0)
dgroup.plot_extras.append(('marker', dgroup.e0, y4e0[ie0], {}))
def plot(self,
dgroup,
title=None,
plot_yarrays=None,
delay_draw=False,
multi=False,
new=True,
zoom_out=True,
with_extras=True,
**kws):
self.get_plot_arrays(dgroup)
ppanel = self.controller.get_display(stacked=False).panel
viewlims = ppanel.get_viewlimits()
plotcmd = ppanel.oplot
if new:
plotcmd = ppanel.plot
groupname = dgroup.groupname
if not hasattr(dgroup, 'xdat'):
print("Cannot plot group ", groupname)
if ((getattr(dgroup, 'plot_yarrays', None) is None
or getattr(dgroup, 'energy', None) is None
or getattr(dgroup, 'mu', None) is None)):
self.process(dgroup=dgroup)
if plot_yarrays is None and hasattr(dgroup, 'plot_yarrays'):
plot_yarrays = dgroup.plot_yarrays
popts = kws
path, fname = os.path.split(dgroup.filename)
if 'label' not in popts:
popts['label'] = dgroup.plot_ylabel
zoom_out = (zoom_out or min(dgroup.xdat) >= viewlims[1]
or max(dgroup.xdat) <= viewlims[0]
or min(dgroup.ydat) >= viewlims[3]
or max(dgroup.ydat) <= viewlims[2])
if not zoom_out:
popts['xmin'] = viewlims[0]
popts['xmax'] = viewlims[1]
popts['ymin'] = viewlims[2]
popts['ymax'] = viewlims[3]
popts['xlabel'] = dgroup.plot_xlabel
popts['ylabel'] = dgroup.plot_ylabel
if getattr(dgroup, 'plot_y2label', None) is not None:
popts['y2label'] = dgroup.plot_y2label
if multi:
yarray_name = PlotSel_Choices[self.plotsel_op.GetStringSelection()]
popts['ylabel'] = getattr(plotlabels, yarray_name)
plot_extras = None
if new:
if title is None:
title = fname
plot_extras = getattr(dgroup, 'plot_extras', None)
popts['title'] = title
popts['delay_draw'] = delay_draw
if hasattr(dgroup, 'custom_plotopts'):
popts.update(dgroup.custom_plotopts)
narr = len(plot_yarrays) - 1
for i, pydat in enumerate(plot_yarrays):
yaname, yopts, yalabel = pydat
popts.update(yopts)
if yalabel is not None:
popts['label'] = yalabel
popts['delay_draw'] = delay_draw or (i != narr)
# print("plot:: ", i, popts['delay_draw'], plotcmd, popts)
if yaname == 'dnormde' and not hasattr(dgroup, yaname):
self.make_dnormde(dgroup)
plotcmd(dgroup.xdat, getattr(dgroup, yaname), **popts)
plotcmd = ppanel.oplot
if with_extras and plot_extras is not None:
axes = ppanel.axes
for etype, x, y, opts in plot_extras:
if etype == 'marker':
xpopts = {
'marker': 'o',
'markersize': 4,
'label': '_nolegend_',
'markerfacecolor': 'red',
'markeredgecolor': '#884444'
}
xpopts.update(opts)
axes.plot([x], [y], **xpopts)
elif etype == 'vline':
xpopts = {
'ymin': 0,
'ymax': 1.0,
'label': '_nolegend_',
'color': '#888888'
}
xpopts.update(opts)
axes.axvline(x, **xpopts)
if not popts['delay_draw']:
ppanel.canvas.draw()
def build_display(self):
panel = self.panel
wids = self.wids
self.skip_process = True
wids['fitspace'] = Choice(panel,
choices=FitSpace_Choices,
size=(175, -1))
wids['fitspace'].SetStringSelection(norm)
add_text = self.add_text
opts = dict(digits=2, increment=1.0, relative_e0=False) # True)
elo_wids = self.add_floatspin('elo', value=defaults['elo'], **opts)
ehi_wids = self.add_floatspin('ehi', value=defaults['ehi'], **opts)
wids['fit_group'] = Button(panel,
'Fit this Group',
size=(175, -1),
action=self.onFitOne)
wids['fit_selected'] = Button(panel,
'Fit Selected Groups',
size=(175, -1),
action=self.onFitAll)
wids['add_selected'] = Button(panel,
'Use Selected Groups as Components',
size=(300, -1),
action=self.onUseSelected)
wids['saveconf'] = Button(panel,
'Save as Default Settings',
size=(225, -1),
action=self.onSaveConfigBtn)
opts = dict(default=True, size=(75, -1), action=self.onPlotOne)
wids['show_fitrange'] = Check(panel, label='show?', **opts)
wids['sum_to_one'] = Check(panel,
label='Weights Must Sum to 1?',
default=True)
wids['all_combos'] = Check(panel,
label='Fit All Combinations?',
default=True)
max_ncomps = self.add_floatspin('max_ncomps',
value=10,
digits=0,
increment=1,
min_val=0,
max_val=20,
size=(60, -1),
with_pin=False)
panel.Add(SimpleText(panel, ' Linear Combination Analysis',
**self.titleopts),
dcol=4)
add_text('Array to Fit: ', newrow=True)
panel.Add(wids['fitspace'], dcol=3)
add_text('Fit Energy Range: ')
panel.Add(elo_wids)
add_text(' : ', newrow=False)
panel.Add(ehi_wids)
panel.Add(wids['show_fitrange'])
panel.Add(HLine(panel, size=(625, 3)), dcol=5, newrow=True)
add_text('Run Fit: ')
panel.Add(wids['fit_group'], dcol=2)
panel.Add(wids['fit_selected'], dcol=3)
add_text('Fit Options: ')
panel.Add(wids['sum_to_one'], dcol=2)
panel.Add((10, 10), dcol=1, newrow=True)
panel.Add(wids['all_combos'], dcol=2)
add_text('Max # Components: ', newrow=False)
panel.Add(max_ncomps, dcol=2)
panel.Add(HLine(panel, size=(625, 3)), dcol=5, newrow=True)
add_text('Components: ')
panel.Add(wids['add_selected'], dcol=4)
collabels = [' File /Group Name ', 'weight', 'min', 'max']
colsizes = [300, 80, 80, 80]
coltypes = ['str', 'float:12,4', 'float:12,4', 'float:12,4']
coldefs = ['', 1.0 / MAX_COMPONENTS, 0.0, 1.0]
wids['table'] = DataTableGrid(panel,
nrows=MAX_COMPONENTS,
collabels=collabels,
datatypes=coltypes,
defaults=coldefs,
colsizes=colsizes)
wids['table'].SetMinSize((625, 250))
panel.Add(wids['table'], newrow=True, dcol=6)
panel.Add(HLine(panel, size=(625, 3)), dcol=5, newrow=True)
panel.Add(wids['saveconf'], dcol=4, newrow=True)
panel.pack()
sizer = wx.BoxSizer(wx.VERTICAL)
sizer.Add((10, 10), 0, LEFT, 3)
sizer.Add(panel, 1, LEFT, 3)
pack(self, sizer)
self.skip_process = False
def build_display(self):
panel = self.panel
wids = self.wids
self.skip_process = True
wids['fitspace'] = Choice(panel,
choices=FitSpace_Choices,
size=(250, -1))
wids['fitspace'].SetStringSelection(norm)
# wids['plotchoice'] = Choice(panel, choices=Plot_Choices,
# size=(250, -1), action=self.onPlot)
wids['method'] = Choice(panel,
choices=Regress_Choices,
size=(250, -1),
action=self.onRegressMethod)
add_text = self.add_text
opts = dict(digits=2, increment=1.0)
w_xmin = self.add_floatspin('xmin', value=defaults['xmin'], **opts)
w_xmax = self.add_floatspin('xmax', value=defaults['xmax'], **opts)
wids['alpha'] = NumericCombo(panel,
make_steps(),
fmt='%.6g',
default_val=0.01,
width=100)
wids['auto_scale_pls'] = Check(panel,
default=True,
label='auto scale?')
wids['auto_alpha'] = Check(panel, default=False, label='auto alpha?')
wids['fit_intercept'] = Check(panel,
default=True,
label='fit intercept?')
wids['save_csv'] = Button(panel,
'Save CSV File',
size=(150, -1),
action=self.onSaveCSV)
wids['load_csv'] = Button(panel,
'Load CSV File',
size=(150, -1),
action=self.onLoadCSV)
wids['save_model'] = Button(panel,
'Save Model',
size=(150, -1),
action=self.onSaveModel)
wids['save_model'].Disable()
wids['load_model'] = Button(panel,
'Load Model',
size=(150, -1),
action=self.onLoadModel)
wids['train_model'] = Button(panel,
'Train Model From These Data',
size=(275, -1),
action=self.onTrainModel)
wids['fit_group'] = Button(panel,
'Predict Variable for Selected Groups',
size=(275, -1),
action=self.onPredictGroups)
wids['fit_group'].Disable()
w_cvfolds = self.add_floatspin('cv_folds',
digits=0,
with_pin=False,
value=0,
increment=1,
min_val=-1)
w_cvreps = self.add_floatspin('cv_repeats',
digits=0,
with_pin=False,
value=0,
increment=1,
min_val=-1)
w_ncomps = self.add_floatspin('ncomps',
digits=0,
with_pin=False,
value=3,
increment=1,
min_val=1)
wids['varname'] = wx.TextCtrl(panel, -1, 'valence', size=(150, -1))
wids['stat1'] = SimpleText(panel, ' - - - ')
wids['stat2'] = SimpleText(panel, ' - - - ')
collabels = [' File /Group Name ', 'External Value', 'Predicted Value']
colsizes = [300, 120, 120]
coltypes = ['str', 'float:12,4', 'float:12,4']
coldefs = ['', 0.0, 0.0]
wids['table'] = DataTableGrid(panel,
nrows=MAX_ROWS,
collabels=collabels,
datatypes=coltypes,
defaults=coldefs,
colsizes=colsizes)
wids['table'].SetMinSize((625, 175))
wids['use_selected'] = Button(panel,
'Use Selected Groups',
size=(150, -1),
action=self.onFillTable)
panel.Add(SimpleText(panel, 'Feature Regression, Model Selection',
**self.titleopts),
dcol=4)
add_text('Array to Use: ', newrow=True)
panel.Add(wids['fitspace'], dcol=4)
# add_text('Plot : ', newrow=True)
# panel.Add(wids['plotchoice'], dcol=3)
add_text('Fit Energy Range: ')
panel.Add(w_xmin)
add_text(' : ', newrow=False)
panel.Add(w_xmax, dcol=3)
add_text('Regression Method:')
panel.Add(wids['method'], dcol=4)
add_text('PLS # components: ')
panel.Add(w_ncomps)
panel.Add(wids['auto_scale_pls'], dcol=2)
add_text('Lasso Alpha: ')
panel.Add(wids['alpha'])
panel.Add(wids['auto_alpha'], dcol=2)
panel.Add(wids['fit_intercept'])
wids['alpha'].Disable()
wids['auto_alpha'].Disable()
wids['fit_intercept'].Disable()
add_text('Cross Validation: ')
add_text(' # folds, # repeats: ', newrow=False)
panel.Add(w_cvfolds, dcol=2)
panel.Add(w_cvreps)
panel.Add(HLine(panel, size=(600, 2)), dcol=6, newrow=True)
add_text('Build Model: ', newrow=True)
panel.Add(wids['use_selected'], dcol=2)
add_text('Attribute Name: ', newrow=False)
panel.Add(wids['varname'], dcol=4)
add_text('Read/Save Data: ', newrow=True)
panel.Add(wids['load_csv'], dcol=3)
panel.Add(wids['save_csv'], dcol=2)
panel.Add(wids['table'], newrow=True, dcol=5) # , drow=3)
panel.Add(HLine(panel, size=(550, 2)), dcol=5, newrow=True)
panel.Add((5, 5), newrow=True)
add_text('Train Model : ')
panel.Add(wids['train_model'], dcol=3)
panel.Add(wids['load_model'])
add_text('Use This Model : ')
panel.Add(wids['fit_group'], dcol=3)
panel.Add(wids['save_model'])
add_text('Fit Statistics : ')
panel.Add(wids['stat1'], dcol=4)
panel.Add((5, 5), newrow=True)
panel.Add(wids['stat2'], dcol=4)
panel.pack()
sizer = wx.BoxSizer(wx.VERTICAL)
sizer.Add((10, 10), 0, LEFT, 3)
sizer.Add(panel, 1, LEFT, 3)
pack(self, sizer)
self.skip_process = False
def build(self):
sizer = wx.GridBagSizer(5, 5)
sizer.SetVGap(5)
sizer.SetHGap(5)
splitter = wx.SplitterWindow(self, style=wx.SP_LIVE_UPDATE)
splitter.SetMinimumPaneSize(200)
self.datalistbox = EditableListBox(splitter,
self.ShowDataSet,
size=(250, -1))
panel = scrolled.ScrolledPanel(splitter)
self.SetMinSize((650, 600))
self.colors = GUIColors()
# title row
self.wids = wids = {}
title = SimpleText(panel,
'Linear Combination Results',
font=Font(FONTSIZE + 2),
colour=self.colors.title,
style=LEFT)
wids['plot_one'] = Button(panel,
'Plot This Fit',
size=(125, -1),
action=self.onPlotOne)
wids['plot_sel'] = Button(panel,
'Plot N Best Fits',
size=(125, -1),
action=self.onPlotSel)
wids['plot_ntitle'] = SimpleText(panel, 'N fits to plot: ')
wids['plot_nchoice'] = Choice(panel,
size=(60, -1),
choices=['%d' % i for i in range(1, 21)])
wids['plot_nchoice'].SetStringSelection('5')
wids['data_title'] = SimpleText(panel,
'<--> ',
font=Font(FONTSIZE + 2),
colour=self.colors.title,
style=LEFT)
wids['nfits_title'] = SimpleText(panel, 'showing 5 best fits')
copts = dict(size=(125, 30), default=True, action=self.onPlotOne)
# wids['show_e0'] = Check(panel, label='show E0?', **copts)
wids['show_fitrange'] = Check(panel, label='show fit range?', **copts)
irow = 0
sizer.Add(title, (irow, 0), (1, 1), LEFT)
sizer.Add(wids['data_title'], (irow, 1), (1, 2), LEFT)
irow += 1
sizer.Add(wids['nfits_title'], (irow, 0), (1, 1), LEFT)
irow += 1
self.wids['paramstitle'] = SimpleText(panel,
'[[Parameters]]',
font=Font(FONTSIZE + 2),
colour=self.colors.title,
style=LEFT)
sizer.Add(self.wids['paramstitle'], (irow, 0), (1, 1), LEFT)
pview = self.wids['params'] = dv.DataViewListCtrl(panel, style=DVSTYLE)
pview.SetMinSize((475, 200))
pview.AppendTextColumn(' Parameter ', width=230)
pview.AppendTextColumn(' Best-Fit Value', width=120)
pview.AppendTextColumn(' Standard Error ', width=120)
for col in range(3):
this = pview.Columns[col]
isort, align = True, wx.ALIGN_RIGHT
if col == 0:
align = wx.ALIGN_LEFT
this.Sortable = isort
this.Alignment = this.Renderer.Alignment = align
irow += 1
sizer.Add(self.wids['params'], (irow, 0), (4, 1), LEFT)
sizer.Add(self.wids['plot_one'], (irow, 1), (1, 2), LEFT)
sizer.Add(self.wids['plot_sel'], (irow + 1, 1), (1, 2), LEFT)
sizer.Add(self.wids['plot_ntitle'], (irow + 2, 1), (1, 1), LEFT)
sizer.Add(self.wids['plot_nchoice'], (irow + 2, 2), (1, 1), LEFT)
# sizer.Add(self.wids['show_e0'], (irow+3, 1), (1, 2), LEFT)
sizer.Add(self.wids['show_fitrange'], (irow + 3, 1), (1, 2), LEFT)
irow += 4
sizer.Add(HLine(panel, size=(675, 3)), (irow, 0), (1, 4), LEFT)
sview = self.wids['stats'] = dv.DataViewListCtrl(panel, style=DVSTYLE)
sview.Bind(dv.EVT_DATAVIEW_SELECTION_CHANGED, self.onSelectFitStat)
sview.AppendTextColumn(' Fit #', width=50)
sview.AppendTextColumn(' N_vary', width=50)
sview.AppendTextColumn(' N_eval', width=60)
sview.AppendTextColumn(' \u03c7\u00B2', width=110)
sview.AppendTextColumn(' \u03c7\u00B2_reduced', width=110)
sview.AppendTextColumn(' Akaike Info', width=110)
sview.AppendTextColumn(' Bayesian Info', width=110)
for col in range(sview.ColumnCount):
this = sview.Columns[col]
isort, align = True, wx.ALIGN_RIGHT
if col == 0:
align = wx.ALIGN_CENTER
this.Sortable = isort
this.Alignment = this.Renderer.Alignment = align
sview.SetMinSize((675, 175))
irow += 1
title = SimpleText(panel,
'[[Fit Statistics]]',
font=Font(FONTSIZE + 2),
colour=self.colors.title,
style=LEFT)
sizer.Add(title, (irow, 0), (1, 4), LEFT)
irow += 1
sizer.Add(sview, (irow, 0), (1, 4), LEFT)
irow += 1
sizer.Add(HLine(panel, size=(675, 3)), (irow, 0), (1, 4), LEFT)
irow += 1
title = SimpleText(panel,
'[[Weights]]',
font=Font(FONTSIZE + 2),
colour=self.colors.title,
style=LEFT)
sizer.Add(title, (irow, 0), (1, 4), LEFT)
self.wids['weightspanel'] = ppan = wx.Panel(panel)
p1 = SimpleText(ppan, ' < Weights > ')
os = wx.BoxSizer(wx.VERTICAL)
os.Add(p1, 1, 3)
pack(ppan, os)
ppan.SetMinSize((675, 175))
irow += 1
sizer.Add(ppan, (irow, 0), (1, 4), LEFT)
irow += 1
sizer.Add(HLine(panel, size=(675, 3)), (irow, 0), (1, 4), LEFT)
pack(panel, sizer)
panel.SetupScrolling()
splitter.SplitVertically(self.datalistbox, panel, 1)
mainsizer = wx.BoxSizer(wx.VERTICAL)
mainsizer.Add(splitter, 1, wx.GROW | wx.ALL, 5)
pack(self, mainsizer)
# self.SetSize((725, 750))
self.Show()
self.Raise()
def build_display(self):
titleopts = dict(font=Font(12), colour='#AA0000')
panel = self.panel
wids = self.wids
self.skip_process = True
wids['fitspace'] = Choice(panel,
choices=FitSpace_Choices,
size=(175, -1))
wids['fitspace'].SetStringSelection(norm)
wids['plotchoice'] = Choice(panel,
choices=Plot_Choices,
size=(175, -1),
action=self.onPlot)
wids['plotchoice'].SetSelection(1)
add_text = self.add_text
opts = dict(digits=2, increment=1.0, relative_e0=True)
e0_wids = self.add_floatspin('e0', value=0, **opts)
elo_wids = self.add_floatspin('elo', value=-20, **opts)
ehi_wids = self.add_floatspin('ehi', value=30, **opts)
wids['fit_group'] = Button(panel,
'Fit this Group',
size=(150, -1),
action=self.onFitOne)
wids['fit_selected'] = Button(panel,
'Fit Selected Groups',
size=(150, -1),
action=self.onFitAll)
wids['add_selected'] = Button(panel,
'Use Selected Groups as Components',
size=(250, -1),
action=self.onUseSelected)
wids['saveconf'] = Button(panel,
'Save as Default Settings',
size=(200, -1),
action=self.onSaveConfigBtn)
opts = dict(default=True, size=(75, -1), action=self.onPlotOne)
wids['show_e0'] = Check(panel, label='show?', **opts)
wids['show_fitrange'] = Check(panel, label='show?', **opts)
wids['sum_to_one'] = Check(panel,
label='Weights Must Sum to 1?',
default=True)
wids['all_combos'] = Check(panel,
label='Fit All Combinations?',
default=True)
panel.Add(SimpleText(panel, ' Linear Combination Analysis',
**titleopts),
dcol=5)
add_text('Run Fit', newrow=False)
add_text('Array to Fit: ', newrow=True)
panel.Add(wids['fitspace'], dcol=4)
panel.Add(wids['fit_group'])
add_text('Plot : ', newrow=True)
panel.Add(wids['plotchoice'], dcol=4)
panel.Add(wids['fit_selected'])
add_text('E0: ')
panel.Add(e0_wids, dcol=3)
panel.Add(wids['show_e0'])
add_text('Fit Energy Range: ')
panel.Add(elo_wids)
add_text(' : ', newrow=False)
panel.Add(ehi_wids)
panel.Add(wids['show_fitrange'])
panel.Add(wids['sum_to_one'], dcol=2, newrow=True)
panel.Add(wids['all_combos'], dcol=3)
panel.Add(HLine(panel, size=(400, 2)), dcol=5, newrow=True)
add_text('Components: ')
panel.Add(wids['add_selected'], dcol=4)
groupnames = [noname] + list(self.controller.file_groups.keys())
sgrid = GridPanel(panel, nrows=6)
sgrid.Add(SimpleText(sgrid, "#"))
sgrid.Add(SimpleText(sgrid, "Group"))
sgrid.Add(SimpleText(sgrid, "Weight"))
sgrid.Add(SimpleText(sgrid, "Min Weight"))
sgrid.Add(SimpleText(sgrid, "Max Weight"))
fopts = dict(minval=-10, maxval=20, precision=4, size=(60, -1))
for i in range(1, 1 + MAX_COMPONENTS):
si = ("comp", "_%2.2d" % i)
sgrid.Add(SimpleText(sgrid, "%2i" % i), newrow=True)
wids['%schoice%s' % si] = Choice(sgrid,
choices=groupnames,
size=(200, -1),
action=partial(self.onComponent,
comp=i))
wids['%sval%s' % si] = FloatCtrl(sgrid, value=0, **fopts)
wids['%smin%s' % si] = FloatCtrl(sgrid, value=0, **fopts)
wids['%smax%s' % si] = FloatCtrl(sgrid, value=1, **fopts)
for cname in ('choice', 'val', 'min', 'max'):
sgrid.Add(wids[("%%s%s%%s" % cname) % si])
sgrid.pack()
panel.Add(sgrid, dcol=5, newrow=True)
panel.Add(HLine(panel, size=(400, 2)), dcol=5, newrow=True)
panel.Add(wids['saveconf'], dcol=4, newrow=True)
panel.pack()
sizer = wx.BoxSizer(wx.VERTICAL)
sizer.Add(panel, 1, LCEN, 3)
pack(self, sizer)
self.skip_process = False
def build(self):
sizer = wx.GridBagSizer(10, 5)
sizer.SetVGap(5)
sizer.SetHGap(5)
panel = scrolled.ScrolledPanel(self)
self.SetMinSize((675, 600))
self.colors = GUIColors()
# title row
self.wids = wids = {}
title = SimpleText(panel,
'Linear Combination Results',
font=Font(12),
colour=self.colors.title,
style=LCEN)
wids['plot_one'] = Button(panel,
'Plot This Fit',
size=(150, -1),
action=self.onPlotOne)
wids['plot_sel'] = Button(panel,
'Plot top N Fits',
size=(150, -1),
action=self.onPlotSel)
wids['plot_ntitle'] = SimpleText(panel, 'N fits to plot: ')
wids['plot_nchoice'] = Choice(panel,
size=(100, -1),
choices=['%d' % i for i in range(1, 21)])
wids['plot_nchoice'].SetStringSelection('5')
wids['data_title'] = SimpleText(panel,
'<--> ',
font=Font(12),
colour=self.colors.title,
style=LCEN)
wids['nfits_title'] = SimpleText(panel, 'showing top 5 fits')
irow = 0
sizer.Add(title, (irow, 0), (1, 2), LCEN)
sizer.Add(wids['data_title'], (irow, 2), (1, 2), LCEN)
irow += 1
sizer.Add(wids['nfits_title'], (irow, 0), (1, 4), LCEN)
irow += 1
sizer.Add(self.wids['plot_one'], (irow, 0), (1, 1), LCEN)
sizer.Add(self.wids['plot_sel'], (irow, 1), (1, 1), LCEN)
sizer.Add(self.wids['plot_ntitle'], (irow, 2), (1, 1), LCEN)
sizer.Add(self.wids['plot_nchoice'], (irow, 3), (1, 1), LCEN)
irow += 1
self.wids['paramstitle'] = SimpleText(panel,
'[[Parameters]]',
font=Font(12),
colour=self.colors.title,
style=LCEN)
sizer.Add(self.wids['paramstitle'], (irow, 0), (1, 4), LCEN)
pview = self.wids['params'] = dv.DataViewListCtrl(panel, style=DVSTYLE)
pview.SetMinSize((675, 200))
pview.AppendTextColumn(' Parameter ', width=250)
pview.AppendTextColumn(' Best-Fit Value', width=150)
pview.AppendTextColumn(' Uncertainty ', width=150)
for col in range(3):
this = pview.Columns[col]
isort, align = True, wx.ALIGN_RIGHT
if col == 0:
align = wx.ALIGN_LEFT
this.Sortable = isort
this.Alignment = this.Renderer.Alignment = align
irow += 1
sizer.Add(self.wids['params'], (irow, 0), (1, 4), LCEN)
irow += 1
sizer.Add(HLine(panel, size=(675, 3)), (irow, 0), (1, 4), LCEN)
sview = self.wids['stats'] = dv.DataViewListCtrl(panel, style=DVSTYLE)
sview.Bind(dv.EVT_DATAVIEW_SELECTION_CHANGED, self.onSelectFitStat)
sview.AppendTextColumn(' Fit #', width=50)
sview.AppendTextColumn(' N_Vary', width=60)
sview.AppendTextColumn(' N_Data', width=60)
sview.AppendTextColumn(' N_eval', width=60)
sview.AppendTextColumn(u' \u03c7\u00B2', width=100)
sview.AppendTextColumn(u' \u03c7\u00B2_reduced', width=100)
sview.AppendTextColumn(' Akaike Info', width=100)
sview.AppendTextColumn(' Bayesian Info', width=100)
for col in range(8):
this = sview.Columns[col]
isort, align = True, wx.ALIGN_RIGHT
if col == 0:
align = wx.ALIGN_CENTER
this.Sortable = isort
this.Alignment = this.Renderer.Alignment = align
sview.SetMinSize((675, 175))
irow += 1
title = SimpleText(panel,
'[[Fit Statistics]]',
font=Font(12),
colour=self.colors.title,
style=LCEN)
sizer.Add(title, (irow, 0), (1, 4), LCEN)
irow += 1
sizer.Add(sview, (irow, 0), (1, 4), LCEN)
irow += 1
sizer.Add(HLine(panel, size=(675, 3)), (irow, 0), (1, 4), LCEN)
irow += 1
title = SimpleText(panel,
'[[Weights]]',
font=Font(12),
colour=self.colors.title,
style=LCEN)
sizer.Add(title, (irow, 0), (1, 4), LCEN)
self.wids['weightspanel'] = ppan = wx.Panel(self)
p1 = SimpleText(ppan, ' < Weights > ')
os = wx.BoxSizer(wx.VERTICAL)
os.Add(p1, 1, 3)
pack(ppan, os)
ppan.SetMinSize((675, 175))
irow += 1
sizer.Add(ppan, (irow, 0), (1, 4), LCEN)
irow += 1
sizer.Add(HLine(panel, size=(675, 3)), (irow, 0), (1, 4), LCEN)
pack(panel, sizer)
panel.SetupScrolling()
mainsizer = wx.BoxSizer(wx.VERTICAL)
mainsizer.Add(panel, 1, wx.GROW | wx.ALL, 1)
pack(self, mainsizer)
self.Show()
self.Raise()
def build_display(self):
self.mod_nb = flatnotebook(self, {})
pan = self.panel = GridPanel(self, ncols=4, nrows=4, pad=2, itemstyle=LEFT)
self.wids = {}
fsopts = dict(digits=2, increment=0.1, with_pin=True)
ppeak_e0 = self.add_floatspin('ppeak_e0', value=0, **fsopts)
ppeak_elo = self.add_floatspin('ppeak_elo', value=-15, **fsopts)
ppeak_ehi = self.add_floatspin('ppeak_ehi', value=-5, **fsopts)
ppeak_emin = self.add_floatspin('ppeak_emin', value=-30, **fsopts)
ppeak_emax = self.add_floatspin('ppeak_emax', value=0, **fsopts)
self.fitbline_btn = Button(pan,'Fit Baseline', action=self.onFitBaseline,
size=(125, -1))
self.plotmodel_btn = Button(pan, 'Plot Model',
action=self.onPlotModel, size=(125, -1))
self.fitmodel_btn = Button(pan, 'Fit Model',
action=self.onFitModel, size=(125, -1))
self.loadmodel_btn = Button(pan, 'Load Model',
action=self.onLoadFitResult, size=(125, -1))
self.fitmodel_btn.Disable()
self.array_choice = Choice(pan, size=(175, -1),
choices=list(Array_Choices.keys()))
self.array_choice.SetSelection(1)
models_peaks = Choice(pan, size=(150, -1),
choices=ModelChoices['peaks'],
action=self.addModel)
models_other = Choice(pan, size=(150, -1),
choices=ModelChoices['other'],
action=self.addModel)
self.models_peaks = models_peaks
self.models_other = models_other
self.message = SimpleText(pan,
'first fit baseline, then add peaks to fit model.')
self.msg_centroid = SimpleText(pan, '----')
opts = dict(default=True, size=(75, -1), action=self.onPlot)
self.show_centroid = Check(pan, label='show?', **opts)
self.show_peakrange = Check(pan, label='show?', **opts)
self.show_fitrange = Check(pan, label='show?', **opts)
self.show_e0 = Check(pan, label='show?', **opts)
opts = dict(default=False, size=(200, -1), action=self.onPlot)
def add_text(text, dcol=1, newrow=True):
pan.Add(SimpleText(pan, text), dcol=dcol, newrow=newrow)
pan.Add(SimpleText(pan, ' Pre-edge Peak Fitting',
**self.titleopts), dcol=5)
add_text(' Run Fit:', newrow=False)
add_text('Array to fit: ')
pan.Add(self.array_choice, dcol=3)
pan.Add((10, 10))
pan.Add(self.fitbline_btn)
add_text('E0: ')
pan.Add(ppeak_e0)
pan.Add((10, 10), dcol=2)
pan.Add(self.show_e0)
pan.Add(self.plotmodel_btn)
add_text('Fit Energy Range: ')
pan.Add(ppeak_emin)
add_text(' : ', newrow=False)
pan.Add(ppeak_emax)
pan.Add(self.show_fitrange)
pan.Add(self.fitmodel_btn)
t = SimpleText(pan, 'Pre-edge Peak Range: ')
SetTip(t, 'Range used as mask for background')
pan.Add(t, newrow=True)
pan.Add(ppeak_elo)
add_text(' : ', newrow=False)
pan.Add(ppeak_ehi)
pan.Add(self.show_peakrange)
# pan.Add(self.fitsel_btn)
add_text( 'Peak Centroid: ')
pan.Add(self.msg_centroid, dcol=3)
pan.Add(self.show_centroid, dcol=1)
pan.Add(self.loadmodel_btn)
# add model
ts = wx.BoxSizer(wx.HORIZONTAL)
ts.Add(models_peaks)
ts.Add(models_other)
pan.Add(SimpleText(pan, 'Add Component: '), newrow=True)
pan.Add(ts, dcol=7)
pan.Add(SimpleText(pan, 'Messages: '), newrow=True)
pan.Add(self.message, dcol=7)
pan.pack()
sizer = wx.BoxSizer(wx.VERTICAL)
sizer.Add((10, 10), 0, LEFT, 3)
sizer.Add(pan, 0, LEFT, 3)
sizer.Add((10, 10), 0, LEFT, 3)
sizer.Add(HLine(self, size=(550, 2)), 0, LEFT, 3)
sizer.Add((10, 10), 0, LEFT, 3)
sizer.Add(self.mod_nb, 1, LEFT|wx.GROW, 10)
pack(self, sizer)
class PrePeakPanel(TaskPanel):
def __init__(self, parent=None, controller=None, **kws):
TaskPanel.__init__(self, parent, controller,
configname='prepeaks_config',
config=defaults, **kws)
self.fit_components = OrderedDict()
self.user_added_params = None
self.pick2_timer = wx.Timer(self)
self.pick2_group = None
self.Bind(wx.EVT_TIMER, self.onPick2Timer, self.pick2_timer)
self.pick2_t0 = 0.
self.pick2_timeout = 15.
self.pick2erase_timer = wx.Timer(self)
self.pick2erase_panel = None
self.Bind(wx.EVT_TIMER, self.onPick2EraseTimer, self.pick2erase_timer)
def onPanelExposed(self, **kws):
# called when notebook is selected
try:
fname = self.controller.filelist.GetStringSelection()
gname = self.controller.file_groups[fname]
dgroup = self.controller.get_group(gname)
self.fill_form(dgroup)
except:
pass # print(" Cannot Fill prepeak panel from group ")
def build_display(self):
self.mod_nb = flatnotebook(self, {})
pan = self.panel = GridPanel(self, ncols=4, nrows=4, pad=2, itemstyle=LEFT)
self.wids = {}
fsopts = dict(digits=2, increment=0.1, with_pin=True)
ppeak_e0 = self.add_floatspin('ppeak_e0', value=0, **fsopts)
ppeak_elo = self.add_floatspin('ppeak_elo', value=-15, **fsopts)
ppeak_ehi = self.add_floatspin('ppeak_ehi', value=-5, **fsopts)
ppeak_emin = self.add_floatspin('ppeak_emin', value=-30, **fsopts)
ppeak_emax = self.add_floatspin('ppeak_emax', value=0, **fsopts)
self.fitbline_btn = Button(pan,'Fit Baseline', action=self.onFitBaseline,
size=(125, -1))
self.plotmodel_btn = Button(pan, 'Plot Model',
action=self.onPlotModel, size=(125, -1))
self.fitmodel_btn = Button(pan, 'Fit Model',
action=self.onFitModel, size=(125, -1))
self.loadmodel_btn = Button(pan, 'Load Model',
action=self.onLoadFitResult, size=(125, -1))
self.fitmodel_btn.Disable()
self.array_choice = Choice(pan, size=(175, -1),
choices=list(Array_Choices.keys()))
self.array_choice.SetSelection(1)
models_peaks = Choice(pan, size=(150, -1),
choices=ModelChoices['peaks'],
action=self.addModel)
models_other = Choice(pan, size=(150, -1),
choices=ModelChoices['other'],
action=self.addModel)
self.models_peaks = models_peaks
self.models_other = models_other
self.message = SimpleText(pan,
'first fit baseline, then add peaks to fit model.')
self.msg_centroid = SimpleText(pan, '----')
opts = dict(default=True, size=(75, -1), action=self.onPlot)
self.show_centroid = Check(pan, label='show?', **opts)
self.show_peakrange = Check(pan, label='show?', **opts)
self.show_fitrange = Check(pan, label='show?', **opts)
self.show_e0 = Check(pan, label='show?', **opts)
opts = dict(default=False, size=(200, -1), action=self.onPlot)
def add_text(text, dcol=1, newrow=True):
pan.Add(SimpleText(pan, text), dcol=dcol, newrow=newrow)
pan.Add(SimpleText(pan, ' Pre-edge Peak Fitting',
**self.titleopts), dcol=5)
add_text(' Run Fit:', newrow=False)
add_text('Array to fit: ')
pan.Add(self.array_choice, dcol=3)
pan.Add((10, 10))
pan.Add(self.fitbline_btn)
add_text('E0: ')
pan.Add(ppeak_e0)
pan.Add((10, 10), dcol=2)
pan.Add(self.show_e0)
pan.Add(self.plotmodel_btn)
add_text('Fit Energy Range: ')
pan.Add(ppeak_emin)
add_text(' : ', newrow=False)
pan.Add(ppeak_emax)
pan.Add(self.show_fitrange)
pan.Add(self.fitmodel_btn)
t = SimpleText(pan, 'Pre-edge Peak Range: ')
SetTip(t, 'Range used as mask for background')
pan.Add(t, newrow=True)
pan.Add(ppeak_elo)
add_text(' : ', newrow=False)
pan.Add(ppeak_ehi)
pan.Add(self.show_peakrange)
# pan.Add(self.fitsel_btn)
add_text( 'Peak Centroid: ')
pan.Add(self.msg_centroid, dcol=3)
pan.Add(self.show_centroid, dcol=1)
pan.Add(self.loadmodel_btn)
# add model
ts = wx.BoxSizer(wx.HORIZONTAL)
ts.Add(models_peaks)
ts.Add(models_other)
pan.Add(SimpleText(pan, 'Add Component: '), newrow=True)
pan.Add(ts, dcol=7)
pan.Add(SimpleText(pan, 'Messages: '), newrow=True)
pan.Add(self.message, dcol=7)
pan.pack()
sizer = wx.BoxSizer(wx.VERTICAL)
sizer.Add((10, 10), 0, LEFT, 3)
sizer.Add(pan, 0, LEFT, 3)
sizer.Add((10, 10), 0, LEFT, 3)
sizer.Add(HLine(self, size=(550, 2)), 0, LEFT, 3)
sizer.Add((10, 10), 0, LEFT, 3)
sizer.Add(self.mod_nb, 1, LEFT|wx.GROW, 10)
pack(self, sizer)
def get_config(self, dgroup=None):
"""get processing configuration for a group"""
if dgroup is None:
dgroup = self.controller.get_group()
conf = getattr(dgroup, 'prepeak_config', {})
if 'e0' not in conf:
conf = defaults
conf['e0'] = getattr(dgroup, 'e0', -1)
dgroup.prepeak_config = conf
if not hasattr(dgroup, 'prepeaks'):
dgroup.prepeaks = Group()
return conf
def fill_form(self, dat):
if isinstance(dat, Group):
self.wids['ppeak_e0'].SetValue(dat.e0)
if hasattr(dat, 'prepeaks'):
self.wids['ppeak_emin'].SetValue(dat.prepeaks.emin)
self.wids['ppeak_emax'].SetValue(dat.prepeaks.emax)
self.wids['ppeak_elo'].SetValue(dat.prepeaks.elo)
self.wids['ppeak_ehi'].SetValue(dat.prepeaks.ehi)
elif isinstance(dat, dict):
self.wids['ppeak_e0'].SetValue(dat['e0'])
self.wids['ppeak_emin'].SetValue(dat['emin'])
self.wids['ppeak_emax'].SetValue(dat['emax'])
self.wids['ppeak_elo'].SetValue(dat['elo'])
self.wids['ppeak_ehi'].SetValue(dat['ehi'])
self.array_choice.SetStringSelection(dat['array_desc'])
self.show_e0.Enable(dat['show_e0'])
self.show_centroid.Enable(dat['show_centroid'])
self.show_fitrange.Enable(dat['show_fitrange'])
self.show_peakrange.Enable(dat['show_peakrange'])
def read_form(self):
"read for, returning dict of values"
dgroup = self.controller.get_group()
array_desc = self.array_choice.GetStringSelection()
form_opts = {'gname': dgroup.groupname,
'filename': dgroup.filename,
'array_desc': array_desc.lower(),
'array_name': Array_Choices[array_desc],
'baseline_form': 'lorentzian',
'bkg_components': []}
form_opts['e0'] = self.wids['ppeak_e0'].GetValue()
form_opts['emin'] = self.wids['ppeak_emin'].GetValue()
form_opts['emax'] = self.wids['ppeak_emax'].GetValue()
form_opts['elo'] = self.wids['ppeak_elo'].GetValue()
form_opts['ehi'] = self.wids['ppeak_ehi'].GetValue()
form_opts['plot_sub_bline'] = False # self.plot_sub_bline.IsChecked()
form_opts['show_centroid'] = self.show_centroid.IsChecked()
form_opts['show_peakrange'] = self.show_peakrange.IsChecked()
form_opts['show_fitrange'] = self.show_fitrange.IsChecked()
form_opts['show_e0'] = self.show_e0.IsChecked()
return form_opts
def onFitBaseline(self, evt=None):
opts = self.read_form()
cmd = """{gname:s}.ydat = 1.0*{gname:s}.{array_name:s}
pre_edge_baseline(energy={gname:s}.energy, norm={gname:s}.ydat, group={gname:s}, form='{baseline_form:s}',
with_line=True, elo={elo:.3f}, ehi={ehi:.3f}, emin={emin:.3f}, emax={emax:.3f})"""
self.larch_eval(cmd.format(**opts))
dgroup = self.controller.get_group()
ppeaks = dgroup.prepeaks
dgroup.centroid_msg = "%.4f +/- %.4f eV" % (ppeaks.centroid,
ppeaks.delta_centroid)
self.msg_centroid.SetLabel(dgroup.centroid_msg)
if 'bpeak_' not in self.fit_components:
self.addModel(model='Lorentzian', prefix='bpeak_', isbkg=True)
if 'bline_' not in self.fit_components:
self.addModel(model='Linear', prefix='bline_', isbkg=True)
for prefix in ('bpeak_', 'bline_'):
cmp = self.fit_components[prefix]
# cmp.bkgbox.SetValue(1)
self.fill_model_params(prefix, dgroup.prepeaks.fit_details.params)
self.fill_form(dgroup)
self.fitmodel_btn.Enable()
# self.fitallmodel_btn.Enable()
i1, i2 = self.get_xranges(dgroup.energy)
dgroup.yfit = dgroup.xfit = 0.0*dgroup.energy[i1:i2]
# self.plot_choice.SetStringSelection(PLOT_BASELINE)
self.onPlot(baseline_only=True)
def fill_model_params(self, prefix, params):
comp = self.fit_components[prefix]
parwids = comp.parwids
for pname, par in params.items():
pname = prefix + pname
if pname in parwids:
wids = parwids[pname]
if wids.minval is not None:
wids.minval.SetValue(par.min)
if wids.maxval is not None:
wids.maxval.SetValue(par.max)
wids.value.SetValue(par.value)
varstr = 'vary' if par.vary else 'fix'
if par.expr is not None:
varstr = 'constrain'
if wids.vary is not None:
wids.vary.SetStringSelection(varstr)
def onPlotModel(self, evt=None):
dgroup = self.controller.get_group()
g = self.build_fitmodel(dgroup)
self.onPlot(show_init=True)
def onPlot(self, evt=None, baseline_only=False, show_init=False):
opts = self.read_form()
dgroup = self.controller.get_group()
opts['group'] = opts['gname']
self.larch_eval(COMMANDS['prepeaks_setup'].format(**opts))
cmd = "plot_prepeaks_fit"
args = ['{gname}']
if baseline_only:
cmd = "plot_prepeaks_baseline"
else:
args.append("show_init=%s" % (show_init))
cmd = "%s(%s)" % (cmd, ', '.join(args))
self.larch_eval(cmd.format(**opts))
def addModel(self, event=None, model=None, prefix=None, isbkg=False):
if model is None and event is not None:
model = event.GetString()
if model is None or model.startswith('<'):
return
self.models_peaks.SetSelection(0)
self.models_other.SetSelection(0)
if prefix is None:
p = model[:5].lower()
curmodels = ["%s%i_" % (p, i+1) for i in range(1+len(self.fit_components))]
for comp in self.fit_components:
if comp in curmodels:
curmodels.remove(comp)
prefix = curmodels[0]
label = "%s(prefix='%s')" % (model, prefix)
title = "%s: %s " % (prefix[:-1], model)
title = prefix[:-1]
mclass_kws = {'prefix': prefix}
if 'step' in model.lower():
form = model.lower().replace('step', '').strip()
if form.startswith('err'):
form = 'erf'
label = "Step(form='%s', prefix='%s')" % (form, prefix)
title = "%s: Step %s" % (prefix[:-1], form[:3])
mclass = lm_models.StepModel
mclass_kws['form'] = form
minst = mclass(form=form, prefix=prefix)
else:
if model in ModelFuncs:
mclass = getattr(lm_models, ModelFuncs[model])
else:
mclass = getattr(lm_models, model+'Model')
minst = mclass(prefix=prefix)
panel = GridPanel(self.mod_nb, ncols=2, nrows=5, pad=1, itemstyle=CEN)
def SLabel(label, size=(80, -1), **kws):
return SimpleText(panel, label,
size=size, style=wx.ALIGN_LEFT, **kws)
usebox = Check(panel, default=True, label='Use in Fit?', size=(100, -1))
bkgbox = Check(panel, default=False, label='Is Baseline?', size=(125, -1))
if isbkg:
bkgbox.SetValue(1)
delbtn = Button(panel, 'Delete This Component', size=(200, -1),
action=partial(self.onDeleteComponent, prefix=prefix))
pick2msg = SimpleText(panel, " ", size=(125, -1))
pick2btn = Button(panel, 'Pick Values from Plotted Data', size=(200, -1),
action=partial(self.onPick2Points, prefix=prefix))
# SetTip(mname, 'Label for the model component')
SetTip(usebox, 'Use this component in fit?')
SetTip(bkgbox, 'Label this component as "background" when plotting?')
SetTip(delbtn, 'Delete this model component')
SetTip(pick2btn, 'Select X range on Plot to Guess Initial Values')
panel.Add(SLabel(label, size=(275, -1), colour='#0000AA'),
dcol=4, style=wx.ALIGN_LEFT, newrow=True)
panel.Add(usebox, dcol=2)
panel.Add(bkgbox, dcol=1, style=RIGHT)
panel.Add(pick2btn, dcol=2, style=wx.ALIGN_LEFT, newrow=True)
panel.Add(pick2msg, dcol=3, style=wx.ALIGN_RIGHT)
panel.Add(delbtn, dcol=2, style=wx.ALIGN_RIGHT)
# panel.Add(HLine(panel, size=(150, 3)), dcol=4, style=wx.ALIGN_CENTER)
panel.Add(SLabel("Parameter "), style=wx.ALIGN_LEFT, newrow=True)
panel.AddMany((SLabel(" Value"), SLabel(" Type"), SLabel(' Bounds'),
SLabel(" Min", size=(60, -1)),
SLabel(" Max", size=(60, -1)), SLabel(" Expression")))
parwids = OrderedDict()
parnames = sorted(minst.param_names)
for a in minst._func_allargs:
pname = "%s%s" % (prefix, a)
if (pname not in parnames and
a in minst.param_hints and
a not in minst.independent_vars):
parnames.append(pname)
for pname in parnames:
sname = pname[len(prefix):]
hints = minst.param_hints.get(sname, {})
par = Parameter(name=pname, value=0, vary=True)
if 'min' in hints:
par.min = hints['min']
if 'max' in hints:
par.max = hints['max']
if 'value' in hints:
par.value = hints['value']
if 'expr' in hints:
par.expr = hints['expr']
pwids = ParameterWidgets(panel, par, name_size=100, expr_size=150,
float_size=80, prefix=prefix,
widgets=('name', 'value', 'minval',
'maxval', 'vary', 'expr'))
parwids[par.name] = pwids
panel.Add(pwids.name, newrow=True)
panel.AddMany((pwids.value, pwids.vary, pwids.bounds,
pwids.minval, pwids.maxval, pwids.expr))
for sname, hint in minst.param_hints.items():
pname = "%s%s" % (prefix, sname)
if 'expr' in hint and pname not in parnames:
par = Parameter(name=pname, value=0, expr=hint['expr'])
pwids = ParameterWidgets(panel, par, name_size=100, expr_size=400,
float_size=80, prefix=prefix,
widgets=('name', 'value', 'expr'))
parwids[par.name] = pwids
panel.Add(pwids.name, newrow=True)
panel.Add(pwids.value)
panel.Add(pwids.expr, dcol=5, style=wx.ALIGN_RIGHT)
pwids.value.Disable()
fgroup = Group(prefix=prefix, title=title, mclass=mclass,
mclass_kws=mclass_kws, usebox=usebox, panel=panel,
parwids=parwids, float_size=65, expr_size=150,
pick2_msg=pick2msg, bkgbox=bkgbox)
self.fit_components[prefix] = fgroup
panel.pack()
self.mod_nb.AddPage(panel, title, True)
sx,sy = self.GetSize()
self.SetSize((sx, sy+1))
self.SetSize((sx, sy))
self.fitmodel_btn.Enable()
def onDeleteComponent(self, evt=None, prefix=None):
fgroup = self.fit_components.get(prefix, None)
if fgroup is None:
return
for i in range(self.mod_nb.GetPageCount()):
if fgroup.title == self.mod_nb.GetPageText(i):
self.mod_nb.DeletePage(i)
for attr in dir(fgroup):
setattr(fgroup, attr, None)
self.fit_components.pop(prefix)
if len(self.fit_components) < 1:
self.fitmodel_btn.Disable()
# sx,sy = self.GetSize()
# self.SetSize((sx, sy+1))
# self.SetSize((sx, sy))
def onPick2EraseTimer(self, evt=None):
"""erases line trace showing automated 'Pick 2' guess """
self.pick2erase_timer.Stop()
panel = self.pick2erase_panel
ntrace = panel.conf.ntrace - 1
trace = panel.conf.get_mpl_line(ntrace)
panel.conf.get_mpl_line(ntrace).set_data(np.array([]), np.array([]))
panel.conf.ntrace = ntrace
panel.draw()
def onPick2Timer(self, evt=None):
"""checks for 'Pick 2' events, and initiates 'Pick 2' guess
for a model from the selected data range
"""
try:
plotframe = self.controller.get_display(win=1)
curhist = plotframe.cursor_hist[:]
plotframe.Raise()
except:
return
if (time.time() - self.pick2_t0) > self.pick2_timeout:
msg = self.pick2_group.pick2_msg.SetLabel(" ")
plotframe.cursor_hist = []
self.pick2_timer.Stop()
return
if len(curhist) < 2:
self.pick2_group.pick2_msg.SetLabel("%i/2" % (len(curhist)))
return
self.pick2_group.pick2_msg.SetLabel("done.")
self.pick2_timer.Stop()
# guess param values
xcur = (curhist[0][0], curhist[1][0])
xmin, xmax = min(xcur), max(xcur)
dgroup = getattr(self.larch.symtable, self.controller.groupname)
x, y = dgroup.xdat, dgroup.ydat
i0 = index_of(dgroup.xdat, xmin)
i1 = index_of(dgroup.xdat, xmax)
x, y = dgroup.xdat[i0:i1+1], dgroup.ydat[i0:i1+1]
mod = self.pick2_group.mclass(prefix=self.pick2_group.prefix)
parwids = self.pick2_group.parwids
try:
guesses = mod.guess(y, x=x)
except:
return
for name, param in guesses.items():
if name in parwids:
parwids[name].value.SetValue(param.value)
dgroup._tmp = mod.eval(guesses, x=dgroup.xdat)
plotframe = self.controller.get_display(win=1)
plotframe.cursor_hist = []
plotframe.oplot(dgroup.xdat, dgroup._tmp)
self.pick2erase_panel = plotframe.panel
self.pick2erase_timer.Start(5000)
def onPick2Points(self, evt=None, prefix=None):
fgroup = self.fit_components.get(prefix, None)
if fgroup is None:
return
plotframe = self.controller.get_display(win=1)
plotframe.Raise()
plotframe.cursor_hist = []
fgroup.npts = 0
self.pick2_group = fgroup
if fgroup.pick2_msg is not None:
fgroup.pick2_msg.SetLabel("0/2")
self.pick2_t0 = time.time()
self.pick2_timer.Start(250)
def onLoadFitResult(self, event=None):
dlg = wx.FileDialog(self, message="Load Saved File Model",
wildcard=ModelWcards, style=wx.FD_OPEN)
rfile = None
if dlg.ShowModal() == wx.ID_OK:
rfile = dlg.GetPath()
dlg.Destroy()
if rfile is None:
return
self.larch_eval("# peakmodel = lm_load_modelresult('%s')" %rfile)
result = load_modelresult(str(rfile))
for prefix in list(self.fit_components.keys()):
self.onDeleteComponent(self, prefix=prefix)
for comp in result.model.components:
isbkg = comp.prefix in result.user_options['bkg_components']
self.addModel(model=comp.func.__name__,
prefix=comp.prefix, isbkg=isbkg)
for comp in result.model.components:
parwids = self.fit_components[comp.prefix].parwids
for pname, par in result.params.items():
if pname in parwids:
wids = parwids[pname]
if wids.minval is not None:
wids.minval.SetValue(par.min)
if wids.maxval is not None:
wids.maxval.SetValue(par.max)
val = result.init_values.get(pname, par.value)
wids.value.SetValue(val)
self.fill_form(result.user_options)
def onSelPoint(self, evt=None, opt='__', relative_e0=False, win=None):
"""
get last selected point from a specified plot window
and fill in the value for the widget defined by `opt`.
by default it finds the latest cursor position from the
cursor history of the first 20 plot windows.
"""
if opt not in self.wids:
return None
_x, _y = last_cursor_pos(win=win, _larch=self.larch)
if _x is not None:
if relative_e0 and 'e0' in self.wids:
_x -= self.wids['e0'].GetValue()
self.wids[opt].SetValue(_x)
def get_xranges(self, x):
opts = self.read_form()
dgroup = self.controller.get_group()
en_eps = min(np.diff(dgroup.energy)) / 5.
i1 = index_of(x, opts['emin'] + en_eps)
i2 = index_of(x, opts['emax'] + en_eps) + 1
return i1, i2
def build_fitmodel(self, dgroup):
""" use fit components to build model"""
# self.summary = {'components': [], 'options': {}}
peaks = []
cmds = ["## set up pre-edge peak parameters", "peakpars = Parameters()"]
modcmds = ["## define pre-edge peak model"]
modop = " ="
opts = self.read_form()
opts['group'] = opts['gname']
self.larch_eval(COMMANDS['prepeaks_setup'].format(**opts))
for comp in self.fit_components.values():
_cen, _amp = None, None
if comp.usebox is not None and comp.usebox.IsChecked():
for parwids in comp.parwids.values():
this = parwids.param
pargs = ["'%s'" % this.name, 'value=%f' % (this.value),
'min=%f' % (this.min), 'max=%f' % (this.max)]
if this.expr is not None:
pargs.append("expr='%s'" % (this.expr))
elif not this.vary:
pargs.pop()
pargs.pop()
pargs.append("vary=False")
cmds.append("peakpars.add(%s)" % (', '.join(pargs)))
if this.name.endswith('_center'):
_cen = this.name
elif parwids.param.name.endswith('_amplitude'):
_amp = this.name
compargs = ["%s='%s'" % (k,v) for k,v in comp.mclass_kws.items()]
modcmds.append("peakmodel %s %s(%s)" % (modop, comp.mclass.__name__,
', '.join(compargs)))
modop = "+="
if not comp.bkgbox.IsChecked() and _cen is not None and _amp is not None:
peaks.append((_amp, _cen))
if len(peaks) > 0:
denom = '+'.join([p[0] for p in peaks])
numer = '+'.join(["%s*%s "% p for p in peaks])
cmds.append("peakpars.add('fit_centroid', expr='(%s)/(%s)')" % (numer, denom))
cmds.extend(modcmds)
cmds.append(COMMANDS['prepfit'].format(group=dgroup.groupname,
user_opts=repr(opts)))
self.larch_eval("\n".join(cmds))
def onFitSelected(self, event=None):
dgroup = self.controller.get_group()
self.build_fitmodel(dgroup)
def onFitModel(self, event=None):
dgroup = self.controller.get_group()
if dgroup is None:
return
self.build_fitmodel(dgroup)
opts = self.read_form()
dgroup = self.controller.get_group()
opts['group'] = opts['gname']
self.larch_eval(COMMANDS['prepeaks_setup'].format(**opts))
ppeaks = dgroup.prepeaks
# add bkg_component to saved user options
bkg_comps = []
for label, comp in self.fit_components.items():
if comp.bkgbox.IsChecked():
bkg_comps.append(label)
opts['bkg_components'] = bkg_comps
imin, imax = self.get_xranges(dgroup.xdat)
cmds = ["## do peak fit: "]
yerr_type = 'set_yerr_const'
yerr = getattr(dgroup, 'yerr', None)
if yerr is None:
if hasattr(dgroup, 'norm_std'):
cmds.append("{group}.yerr = {group}.norm_std")
yerr_type = 'set_yerr_array'
elif hasattr(dgroup, 'mu_std'):
cmds.append("{group}.yerr = {group}.mu_std/(1.e-15+{group}.edge_step)")
yerr_type = 'set_yerr_array'
else:
cmds.append("{group}.yerr = 1")
elif isinstance(dgroup.yerr, np.ndarray):
yerr_type = 'set_yerr_array'
cmds.extend([COMMANDS[yerr_type], COMMANDS['dofit']])
cmd = '\n'.join(cmds)
self.larch_eval(cmd.format(group=dgroup.groupname,
imin=imin, imax=imax,
user_opts=repr(opts)))
self.autosave_modelresult(self.larch_get("peakresult"))
self.onPlot()
self.show_subframe('prepeak_result_frame', FitResultFrame,
datagroup=dgroup, peakframe=self)
self.subframes['prepeak_result_frame'].show_results()
def update_start_values(self, params):
"""fill parameters with best fit values"""
allparwids = {}
for comp in self.fit_components.values():
if comp.usebox is not None and comp.usebox.IsChecked():
for name, parwids in comp.parwids.items():
allparwids[name] = parwids
for pname, par in params.items():
if pname in allparwids:
allparwids[pname].value.SetValue(par.value)
def autosave_modelresult(self, result, fname=None):
"""autosave model result to user larch folder"""
confdir = os.path.join(site_config.usr_larchdir, 'xas_viewer')
if not os.path.exists(confdir):
try:
os.makedirs(confdir)
except OSError:
print("Warning: cannot create XAS_Viewer user folder")
return
if not HAS_MODELSAVE:
print("Warning: cannot save model results: upgrade lmfit")
return
if fname is None:
fname = 'autosave.fitmodel'
save_modelresult(result, os.path.join(confdir, fname))