def __init__(self, get_filenames, n_merge_bins=None, plot_bins_step=4,
spread_data=False, neg_shift=True):
Struct.__init__(self, get_filenames=get_filenames,
n_merge_bins=n_merge_bins,
plot_bins_step=plot_bins_step,
spread_data=spread_data, neg_shift=neg_shift)
self.current = Struct(filenames=None, dir_base=None, base_names=None)
self.source_data = Struct(data=None, fdata=None, bins=None)
def __init__(self, model_class, n_components, parameters, constraints,
solver_conf, output_dir):
Struct.__init__(self,
model_class=model_class,
n_components=n_components,
parameters=parameters,
constraints=constraints,
solver_conf=solver_conf,
output_dir=output_dir)
def __init__(self, model_class, n_components, parameters, constraints, solver_conf, output_dir):
Struct.__init__(
self,
model_class=model_class,
n_components=n_components,
parameters=parameters,
constraints=constraints,
solver_conf=solver_conf,
output_dir=output_dir,
)
def __init__(self,
get_filenames,
n_merge_bins=None,
plot_bins_step=4,
spread_data=False,
neg_shift=True):
Struct.__init__(self,
get_filenames=get_filenames,
n_merge_bins=n_merge_bins,
plot_bins_step=plot_bins_step,
spread_data=spread_data,
neg_shift=neg_shift)
self.current = Struct(filenames=None, dir_base=None, base_names=None)
self.source_data = Struct(data=None, fdata=None, bins=None)
def __call__(self):
for filenames in self.get_filenames:
dir_base, base_names = io.split_dir_base(filenames)
print '======================================================'
print 'directory base:', dir_base
data = io.load_data(filenames)
if self.n_merge_bins is not None:
data = tr.merge_bins(data, self.n_merge_bins)
print 'angles range:', data[:, 0].min(), data[:, 0].max()
data = tr.fix_increasing(tr.fix_range(tr.transform_2pi(data)))
print 'transformed angles range:', data[0, 0], data[-1, 0]
print 'data range:', data[:, 1].min(), data[:, 1].max()
# Simulate the "random process" the histogram was done from.
counts = tr.get_counts_from_lengths(data[:, 1])
fdata = tr.spread_by_counts(data[:, 0],
counts,
trivial=self.spread_data == False)
print 'simulated counts range:', counts.min(), counts.max()
ddata = np.sort(
tr.transform_pi_deg(data[:, 0], neg_shift=self.neg_shift))
dd = ddata[1] - ddata[0]
all_bins = np.r_[ddata - 1e-8, ddata[-1] + dd]
bins = all_bins[::self.plot_bins_step]
self.current = Struct(filenames=filenames,
dir_base=dir_base,
base_names=base_names)
self.source_data = Struct(counts=counts,
data=data,
fdata=fdata,
bins=bins)
yield self.source_data
def parse_row(self, row):
"""
Parse single hitogram row.
"""
n_components = int(row[12])
off = 13 + 3 * n_components
params = np.array(map(float, row[13:off])).reshape((n_components, 3))
item = Struct(
dir_base=row[0],
converged=int(row[1]),
fit_criteria=map(float, row[2:5]),
chisquare=map(float, row[5:12]),
n_components=n_components,
params=params,
base_names=[ii.strip() for ii in row[off + 1].split(',')])
assert (len(item.base_names) == int(row[off]))
return item
def write_row(self, dir_base, base_names, chisquare, params, converged,
fit_criteria):
n_components = params.shape[0]
item = Struct(dir_base=dir_base,
base_names=base_names,
chisquare=chisquare,
params=params,
converged=converged,
n_components=n_components,
fit_criteria=fit_criteria)
self.items.append(item)
fd = open(self.log_name, 'a')
writer = csv.writer(fd)
writer.writerow([dir_base, int(converged)] + fit_criteria + chisquare +
[n_components] + params.ravel().tolist() +
[len(base_names), ', '.join(base_names)])
fd.close()
def write_row(self, dir_base, base_names, area_angles):
x0, xm, x1, area1, area2 = area_angles
item = Struct(dir_base=dir_base,
base_names=base_names,
x0=x0,
xm=xm,
x1=x1,
area1=area1,
area2=area2)
self.items.append(item)
fd = open(self.log_name, 'a')
writer = csv.writer(fd)
writer.writerow(
[dir_base] +
[x0, 0.5 * (x0 + xm), xm, 0.5 * (xm + x1), x1, area1, area2] +
[len(base_names), ', '.join(base_names)])
fd.close()
def make_summary(logs):
"""
Make a summary table of all results.
The best parameter set according to each criterion is selected.
Parameters
----------
logs : list of CSVLog
A log with fitting results for each parameter set.
Returns
-------
summary : dict
The summary dictionary with data directory base names as keys.
"""
dir_bases = [item.dir_base for item in logs[0].items]
summary = {}
for ii, dir_base in enumerate(dir_bases):
# Results for dir_base for all parameter sets.
items = [log.items[ii] for log in logs]
criteria = [item.fit_criteria for item in items]
pset_ids = np.array(criteria).argsort(axis=0)
converged = np.take([item.converged for item in items], pset_ids)
probs = np.take([item.params[:, 2] for item in items],
pset_ids,
axis=0)
summary[dir_base] = Struct(
sorted=zip(pset_ids, converged, probs),
criteria=criteria,
pset_ids=pset_ids,
chisquare=[item.chisquare for item in items])
return summary
def __init__(self, psets):
Struct.__init__(self, psets=psets)
def __init__(self, psets):
Struct.__init__(self, psets=psets)
