def _eval_block(iblock, start, stop, nstates, total_fluxes, cond_fluxes, rates,
mcbs_alpha, mcbs_nsets, mcbs_acalpha):
results = [[], [], []]
# results are target fluxes, conditional fluxes, rates
for istate in xrange(nstates):
ci_res = mcbs_ci_correl(total_fluxes[:, istate],
estimator=numpy.mean,
alpha=mcbs_alpha,
n_sets=mcbs_nsets,
autocorrel_alpha=mcbs_acalpha,
subsample=numpy.mean)
results[0].append((iblock, istate, (start, stop) + ci_res))
for jstate in xrange(nstates):
if istate == jstate: continue
ci_res = mcbs_ci_correl(cond_fluxes[:, istate, jstate],
estimator=numpy.mean,
alpha=mcbs_alpha,
n_sets=mcbs_nsets,
autocorrel_alpha=mcbs_acalpha,
subsample=numpy.mean)
results[1].append((iblock, istate, jstate, (start, stop) + ci_res))
ci_res = mcbs_ci_correl(rates[:, istate, jstate],
estimator=numpy.mean,
alpha=mcbs_alpha,
n_sets=mcbs_nsets,
autocorrel_alpha=mcbs_acalpha,
subsample=numpy.mean)
results[2].append((iblock, istate, jstate, (start, stop) + ci_res))
return results
def _2D_eval_block(iblock, start, stop, nstates, data_input, name, mcbs_alpha,
mcbs_nsets, mcbs_acalpha, do_correl, mcbs_enable,
estimator_kwargs):
# As our reweighting estimator is a weird function, we can't use the general mclib block.
results = []
for istate in range(nstates):
for jstate in range(nstates):
if istate == jstate: continue
estimator_kwargs.update(
dict(istate=istate, jstate=jstate, nstates=nstates))
dataset = {
'indices':
np.array(list(range(start - 1, stop - 1)), dtype=np.uint16)
}
ci_res = mcbs_ci_correl(dataset,
estimator=reweight_for_c,
alpha=mcbs_alpha,
n_sets=mcbs_nsets,
autocorrel_alpha=mcbs_acalpha,
subsample=(lambda x: x[0]),
do_correl=do_correl,
mcbs_enable=mcbs_enable,
estimator_kwargs=estimator_kwargs)
results.append(
(name, iblock, istate, jstate, (start, stop) + ci_res))
return results
def _rate_eval_block(iblock, start, stop, nstates, data_input, name,
mcbs_alpha, mcbs_nsets, mcbs_acalpha, do_correl,
mcbs_enable):
# Our rate estimator is a little more complex, so we've defined a custom evaluation block for it,
# instead of just using the block evalutors that we've imported.
results = []
for istate in range(nstates):
for jstate in range(nstates):
if istate == jstate: continue
kwargs = {'istate': istate, 'jstate': jstate}
# Why are we sending in the total population dataset, instead of a sliced one?
# It's a requirement of our estimator; we need to pull from any given i to j state in order to properly normalize
# and avoid i to j rate constants which are affected by a third state k.
# That is, we need the populations for both i and j, and it's easier to just send in the entire dataset.
dataset = {
'dataset': data_input['dataset'][:, istate, jstate],
'pops': data_input['pops']
}
ci_res = mcbs_ci_correl(dataset,
estimator=sequence_macro_flux_to_rate,
alpha=mcbs_alpha,
n_sets=mcbs_nsets,
autocorrel_alpha=mcbs_acalpha,
subsample=numpy.mean,
do_correl=do_correl,
mcbs_enable=mcbs_enable,
estimator_kwargs=kwargs)
results.append(
(name, iblock, istate, jstate, (start, stop) + ci_res))
return results
def _1D_eval_block(iblock, start, stop, nstates, data_input, name, mcbs_alpha,
mcbs_nsets, mcbs_acalpha, do_correl, mcbs_enable,
estimator_kwargs):
# As our reweighting estimator is a weird function, we can't use the general mclib block.
results = []
for istate in range(nstates):
# A little hack to make our estimator play nice, as jstate must be there.
# For 1D datasets (state probabilities, etc), the argument isn't used in our estimator,
# and so any variable which has the proper type is fine.
estimator_kwargs.update(
dict(istate=istate, jstate=istate, nstates=nstates))
dataset = {
'indices': np.array(list(range(start - 1, stop - 1)),
dtype=np.uint16)
}
ci_res = mcbs_ci_correl(dataset,
estimator=reweight_for_c,
alpha=mcbs_alpha,
n_sets=mcbs_nsets,
autocorrel_alpha=mcbs_acalpha,
subsample=(lambda x: x[0]),
do_correl=do_correl,
mcbs_enable=mcbs_enable,
estimator_kwargs=estimator_kwargs)
results.append((name, iblock, istate, (start, stop) + ci_res))
return results
def calc_evol_flux(self):
westpa.rc.pstatus('Calculating cumulative evolution of flux confidence intervals every {} iteration(s)'
.format(self.evol_step))
for itarget, (target_label, target_fluxdata) in enumerate(self.fluxdata.iteritems()):
fluxes = target_fluxdata['flux']
target_group = self.target_groups[target_label]
iter_start = target_group['n_iter'][0]
iter_stop = target_group['n_iter'][-1]
iter_count = iter_stop - iter_start
n_blocks = iter_count // self.evol_step
if iter_count % self.evol_step > 0: n_blocks += 1
cis = numpy.empty((n_blocks,), dtype=ci_dtype)
for iblock in xrange(n_blocks):
block_iter_stop = min(iter_start + (iblock+1)*self.evol_step, iter_stop)
istop = min((iblock+1)*self.evol_step, len(target_fluxdata['flux']))
fluxes = target_fluxdata['flux'][:istop]
avg, ci_lb, ci_ub, correl_len = mclib.mcbs_ci_correl(fluxes, numpy.mean, self.alpha, self.n_sets,
autocorrel_alpha = self.autocorrel_alpha,
subsample=numpy.mean)
cis[iblock]['iter_start'] = iter_start
cis[iblock]['iter_stop'] = block_iter_stop
cis[iblock]['expected'], cis[iblock]['ci_lbound'], cis[iblock]['ci_ubound'] = avg, ci_lb, ci_ub
cis[iblock]['corr_len'] = correl_len
del fluxes
cis_ds = target_group.create_dataset('flux_evolution', data=cis)
cis_ds.attrs['iter_step'] = self.evol_step
cis_ds.attrs['mcbs_alpha'] = self.alpha
cis_ds.attrs['mcbs_autocorrel_alpha'] = self.autocorrel_alpha
cis_ds.attrs['mcbs_n_sets'] = self.n_sets
def _eval_block(iblock, istate, start, stop, state_pops, mcbs_alpha,
mcbs_nsets, mcbs_acalpha):
ci_res = mcbs_ci_correl(state_pops,
estimator=numpy.mean,
alpha=mcbs_alpha,
n_sets=mcbs_nsets,
autocorrel_alpha=mcbs_acalpha,
subsample=numpy.mean)
return (iblock, istate, (start, stop) + ci_res)
def _eval_block(iblock, start, stop, nstates, total_fluxes, cond_fluxes, rates, mcbs_alpha, mcbs_nsets, mcbs_acalpha):
results = [[],[],[]]
# results are target fluxes, conditional fluxes, rates
for istate in xrange(nstates):
ci_res = mcbs_ci_correl(total_fluxes[:,istate],estimator=numpy.mean,
alpha=mcbs_alpha,n_sets=mcbs_nsets,autocorrel_alpha=mcbs_acalpha,
subsample=numpy.mean)
results[0].append((iblock,istate,(start,stop)+ci_res))
for jstate in xrange(nstates):
if istate == jstate: continue
ci_res = mcbs_ci_correl(cond_fluxes[:,istate,jstate],estimator=numpy.mean,
alpha=mcbs_alpha,n_sets=mcbs_nsets,autocorrel_alpha=mcbs_acalpha,
subsample=numpy.mean)
results[1].append((iblock, istate, jstate, (start,stop) + ci_res))
ci_res = mcbs_ci_correl(rates[:,istate,jstate],estimator=numpy.mean,
alpha=mcbs_alpha,n_sets=mcbs_nsets,autocorrel_alpha=mcbs_acalpha,
subsample=numpy.mean)
results[2].append((iblock, istate, jstate, (start,stop) + ci_res))
return results
def calc_evol_flux(self):
westpa.rc.pstatus(
'Calculating cumulative evolution of flux confidence intervals every {} iteration(s)'
.format(self.evol_step))
for itarget, (target_label,
target_fluxdata) in enumerate(self.fluxdata.items()):
fluxes = target_fluxdata['flux']
target_group = self.target_groups[target_label]
iter_start = target_group['n_iter'][0]
iter_stop = target_group['n_iter'][-1]
iter_count = iter_stop - iter_start
n_blocks = iter_count // self.evol_step
if iter_count % self.evol_step > 0: n_blocks += 1
cis = numpy.empty((n_blocks, ), dtype=ci_dtype)
for iblock in range(n_blocks):
block_iter_stop = min(
iter_start + (iblock + 1) * self.evol_step, iter_stop)
istop = min((iblock + 1) * self.evol_step,
len(target_fluxdata['flux']))
fluxes = target_fluxdata['flux'][:istop]
#avg, ci_lb, ci_ub, correl_len = mclib.mcbs_ci_correl(fluxes, numpy.mean, self.alpha, self.n_sets,
# autocorrel_alpha = self.autocorrel_alpha,
# subsample=numpy.mean)
avg, ci_lb, ci_ub, sterr, correl_len = mclib.mcbs_ci_correl(
{'dataset': fluxes},
estimator=(lambda stride, dataset: numpy.mean(dataset)),
alpha=self.alpha,
n_sets=self.n_sets,
autocorrel_alpha=self.autocorrel_alpha,
subsample=numpy.mean,
do_correl=self.do_correl,
mcbs_enable=self.mcbs_enable)
cis[iblock]['iter_start'] = iter_start
cis[iblock]['iter_stop'] = block_iter_stop
cis[iblock]['expected'], cis[iblock]['ci_lbound'], cis[iblock][
'ci_ubound'] = avg, ci_lb, ci_ub
cis[iblock]['corr_len'] = correl_len
cis[iblock]['sterr'] = sterr
del fluxes
cis_ds = target_group.create_dataset('flux_evolution', data=cis)
cis_ds.attrs['iter_step'] = self.evol_step
cis_ds.attrs['mcbs_alpha'] = self.alpha
cis_ds.attrs['mcbs_autocorrel_alpha'] = self.autocorrel_alpha
cis_ds.attrs['mcbs_n_sets'] = self.n_sets
def _2D_eval_block(iblock, start, stop, nstates, data_input, name, mcbs_alpha, mcbs_nsets, mcbs_acalpha, do_correl, mcbs_enable, estimator_kwargs):
# As our reweighting estimator is a weird function, we can't use the general mclib block.
results = []
for istate in xrange(nstates):
for jstate in xrange(nstates):
if istate == jstate: continue
estimator_kwargs.update(dict(istate=istate, jstate=jstate, nstates=nstates))
dataset = { 'indices' : np.array(range(start-1, stop-1), dtype=np.uint16) }
ci_res = mcbs_ci_correl(dataset,estimator=reweight_for_c,
alpha=mcbs_alpha,n_sets=mcbs_nsets,autocorrel_alpha=mcbs_acalpha,
subsample=(lambda x: x[0]), do_correl=do_correl, mcbs_enable=mcbs_enable, estimator_kwargs=estimator_kwargs)
results.append((name, iblock, istate, jstate, (start,stop) + ci_res))
return results
def _1D_eval_block(iblock, start, stop, nstates, data_input, name, mcbs_alpha, mcbs_nsets, mcbs_acalpha, do_correl, mcbs_enable, estimator_kwargs):
# As our reweighting estimator is a weird function, we can't use the general mclib block.
results = []
for istate in xrange(nstates):
# A little hack to make our estimator play nice, as jstate must be there.
# For 1D datasets (state probabilities, etc), the argument isn't used in our estimator,
# and so any variable which has the proper type is fine.
estimator_kwargs.update(dict(istate=istate, jstate=istate, nstates=nstates))
dataset = { 'indices' : np.array(range(start-1, stop-1), dtype=np.uint16) }
ci_res = mcbs_ci_correl(dataset,estimator=reweight_for_c,
alpha=mcbs_alpha,n_sets=mcbs_nsets,autocorrel_alpha=mcbs_acalpha,
subsample=(lambda x: x[0]), do_correl=do_correl, mcbs_enable=mcbs_enable, estimator_kwargs=estimator_kwargs)
results.append((name, iblock, istate, (start,stop) + ci_res))
return results
def _rate_eval_block(iblock, start, stop, nstates, data_input, name, mcbs_alpha, mcbs_nsets, mcbs_acalpha, do_correl, mcbs_enable):
# Our rate estimator is a little more complex, so we've defined a custom evaluation block for it,
# instead of just using the block evalutors that we've imported.
results = []
for istate in xrange(nstates):
for jstate in xrange(nstates):
if istate == jstate: continue
kwargs = { 'istate' : istate, 'jstate': jstate }
# Why are we sending in the total population dataset, instead of a sliced one?
# It's a requirement of our estimator; we need to pull from any given i to j state in order to properly normalize
# and avoid i to j rate constants which are affected by a third state k.
# That is, we need the populations for both i and j, and it's easier to just send in the entire dataset.
dataset = {'dataset': data_input['dataset'][:, istate, jstate], 'pops': data_input['pops'] }
ci_res = mcbs_ci_correl(dataset,estimator=sequence_macro_flux_to_rate,
alpha=mcbs_alpha,n_sets=mcbs_nsets,autocorrel_alpha=mcbs_acalpha,
subsample=numpy.mean, do_correl=do_correl, mcbs_enable=mcbs_enable, estimator_kwargs=kwargs)
results.append((name, iblock, istate, jstate, (start,stop) + ci_res))
return results
def go(self):
pi = self.progress.indicator
with pi:
pi.new_operation('Initializing')
self.open_files()
nstates = self.assignments_file.attrs['nstates']
nbins = self.assignments_file.attrs['nbins']
state_labels = self.assignments_file['state_labels'][...]
assert nstates == len(state_labels)
start_iter, stop_iter, step_iter = self.iter_range.iter_start, self.iter_range.iter_stop, self.iter_range.iter_step
pi.new_operation('Reading data')
cond_fluxes = h5io.IterBlockedDataset(self.kinetics_file['conditional_fluxes'])
cond_fluxes.cache_data()
total_fluxes = h5io.IterBlockedDataset(self.kinetics_file['total_fluxes'])
pops = h5io.IterBlockedDataset(self.assignments_file['labeled_populations'])
pops.cache_data()
pops.data = pops.data.sum(axis=2)
rates = h5io.IterBlockedDataset.empty_like(cond_fluxes)
rates.data = sequence_macro_flux_to_rate(cond_fluxes.data, pops.data[:nstates,:nbins])
avg_total_fluxes = numpy.zeros((nstates,), dtype=ci_dtype)
avg_conditional_fluxes = numpy.zeros((nstates,nstates), dtype=ci_dtype)
avg_rates = numpy.zeros((nstates,nstates), dtype=ci_dtype)
# Calculate overall average rates
pi.new_operation('Averaging overall fluxes into states', nstates)
for istate in xrange(nstates):
ci_res = mcbs_ci_correl(total_fluxes.iter_slice(start_iter,stop_iter)[:,istate],estimator=numpy.mean,
alpha=self.mcbs_alpha,n_sets=self.mcbs_nsets,autocorrel_alpha=self.mcbs_acalpha,
subsample=numpy.mean)
avg_total_fluxes[istate] = (start_iter, stop_iter) + ci_res
pi.progress += 1
pi.new_operation('Averaging state-to-state fluxes and rates', nstates*(nstates-1))
for istate in xrange(nstates):
for jstate in xrange(nstates):
if istate == jstate: continue
flux_ci_res = mcbs_ci_correl(cond_fluxes.iter_slice(start_iter,stop_iter)[:,istate,jstate],estimator=numpy.mean,
alpha=self.mcbs_alpha,n_sets=self.mcbs_nsets,autocorrel_alpha=self.mcbs_acalpha,
subsample=numpy.mean)
rate_ci_res = mcbs_ci_correl(rates.iter_slice(start_iter,stop_iter)[:,istate,jstate],estimator=numpy.mean,
alpha=self.mcbs_alpha,n_sets=self.mcbs_nsets,autocorrel_alpha=self.mcbs_acalpha,
subsample=numpy.mean)
avg_conditional_fluxes[istate, jstate] = (start_iter, stop_iter) + flux_ci_res
avg_rates[istate, jstate] = (start_iter, stop_iter) + rate_ci_res
pi.progress += 1
pi.new_operation('Saving averages')
self.output_file['avg_rates'] = avg_rates
self.output_file['avg_conditional_fluxes'] = avg_conditional_fluxes
self.output_file['avg_total_fluxes'] = avg_total_fluxes
for ds in ('avg_rates', 'avg_conditional_fluxes', 'avg_total_fluxes'):
self.stamp_mcbs_info(self.output_file[ds])
self.output_file['state_labels'] = state_labels
maxlabellen = max(map(len,state_labels))
pi.clear()
print('fluxes into macrostates:')
for istate in xrange(nstates):
print('{:{maxlabellen}s}: mean={:21.15e} CI=({:21.15e}, {:21.15e}) * tau^-1'
.format(state_labels[istate],
avg_total_fluxes['expected'][istate],
avg_total_fluxes['ci_lbound'][istate],
avg_total_fluxes['ci_ubound'][istate],
maxlabellen=maxlabellen))
print('\nfluxes from state to state:')
for istate in xrange(nstates):
for jstate in xrange(nstates):
if istate == jstate: continue
print('{:{maxlabellen}s} -> {:{maxlabellen}s}: mean={:21.15e} CI=({:21.15e}, {:21.15e}) * tau^-1'
.format(state_labels[istate], state_labels[jstate],
avg_conditional_fluxes['expected'][istate,jstate],
avg_conditional_fluxes['ci_lbound'][istate,jstate],
avg_conditional_fluxes['ci_ubound'][istate,jstate],
maxlabellen=maxlabellen))
print('\nrates from state to state:')
for istate in xrange(nstates):
for jstate in xrange(nstates):
if istate == jstate: continue
print('{:{maxlabellen}s} -> {:{maxlabellen}s}: mean={:21.15e} CI=({:21.15e}, {:21.15e}) * tau^-1'
.format(state_labels[istate], state_labels[jstate],
avg_rates['expected'][istate,jstate],
avg_rates['ci_lbound'][istate,jstate],
avg_rates['ci_ubound'][istate,jstate],
maxlabellen=maxlabellen))
# skip evolution if not requested
if self.evolution_mode == 'none' or not step_iter: return
start_pts = range(start_iter, stop_iter, step_iter)
target_evol = numpy.zeros((len(start_pts), nstates), dtype=ci_dtype)
flux_evol = numpy.zeros((len(start_pts), nstates, nstates), dtype=ci_dtype)
rate_evol = numpy.zeros((len(start_pts), nstates, nstates), dtype=ci_dtype)
pi.new_operation('Calculating flux/rate evolution', len(start_pts))
futures = []
for iblock, start in enumerate(start_pts):
stop = min(start+step_iter, stop_iter)
if self.evolution_mode == 'cumulative':
windowsize = int(self.evol_window_frac * (stop - start_iter))
block_start = max(start_iter, stop - windowsize)
else: # self.evolution_mode == 'blocked'
block_start = start
future = self.work_manager.submit(_eval_block, kwargs=dict(iblock=iblock, start=block_start, stop=stop,
nstates=nstates,
total_fluxes=total_fluxes.iter_slice(block_start,stop),
cond_fluxes = cond_fluxes.iter_slice(block_start,stop),
rates=rates.iter_slice(block_start,stop),
mcbs_alpha=self.mcbs_alpha, mcbs_nsets=self.mcbs_nsets,
mcbs_acalpha=self.mcbs_acalpha))
futures.append(future)
for future in self.work_manager.as_completed(futures):
pi.progress += 1
target_results, condflux_results, rate_results = future.get_result(discard=True)
for result in target_results:
iblock,istate,ci_result = result
target_evol[iblock,istate] = ci_result
for result in condflux_results:
iblock,istate,jstate,ci_result = result
flux_evol[iblock,istate, jstate] = ci_result
for result in rate_results:
iblock, istate, jstate, ci_result = result
rate_evol[iblock, istate, jstate] = ci_result
df_ds = self.output_file.create_dataset('conditional_flux_evolution', data=flux_evol, shuffle=True, compression=9)
tf_ds = self.output_file.create_dataset('target_flux_evolution', data=target_evol, shuffle=True, compression=9)
rate_ds = self.output_file.create_dataset('rate_evolution', data=rate_evol, shuffle=True, compression=9)
for ds in (df_ds, tf_ds, rate_ds):
self.stamp_mcbs_info(ds)
def calc_store_flux_data(self):
westpa.rc.pstatus(
'Calculating mean flux and confidence intervals for iterations [{},{})'
.format(self.iter_range.iter_start, self.iter_range.iter_stop))
fluxdata = extract_fluxes(self.iter_range.iter_start,
self.iter_range.iter_stop, self.data_reader)
# Create a group to store data in
output_group = h5io.create_hdf5_group(self.output_h5file,
'target_flux',
replace=False,
creating_program=self.prog)
self.output_group = output_group
output_group.attrs['version_code'] = self.output_format_version
self.iter_range.record_data_iter_range(output_group)
n_targets = len(fluxdata)
index = numpy.empty((len(fluxdata), ), dtype=target_index_dtype)
avg_fluxdata = numpy.empty((n_targets, ), dtype=ci_dtype)
for itarget, (target_label,
target_fluxdata) in enumerate(fluxdata.items()):
# Create group and index entry
index[itarget]['target_label'] = str(target_label)
target_group = output_group.create_group(
'target_{}'.format(itarget))
self.target_groups[target_label] = target_group
# Store per-iteration values
target_group['n_iter'] = target_fluxdata['n_iter']
target_group['count'] = target_fluxdata['count']
target_group['flux'] = target_fluxdata['flux']
h5io.label_axes(target_group['flux'], ['n_iter'], units=['tau^-1'])
# Calculate flux autocorrelation
fluxes = target_fluxdata['flux']
mean_flux = fluxes.mean()
fmm = fluxes - mean_flux
acorr = fftconvolve(fmm, fmm[::-1])
acorr = acorr[len(acorr) // 2:]
acorr /= acorr[0]
acorr_ds = target_group.create_dataset('flux_autocorrel',
data=acorr)
h5io.label_axes(acorr_ds, ['lag'], ['tau'])
# Calculate overall averages and CIs
#avg, lb_ci, ub_ci, correl_len = mclib.mcbs_ci_correl(fluxes, numpy.mean, self.alpha, self.n_sets,
# autocorrel_alpha=self.autocorrel_alpha, subsample=numpy.mean)
avg, lb_ci, ub_ci, sterr, correl_len = mclib.mcbs_ci_correl(
{'dataset': fluxes},
estimator=(lambda stride, dataset: numpy.mean(dataset)),
alpha=self.alpha,
n_sets=self.n_sets,
autocorrel_alpha=self.autocorrel_alpha,
subsample=numpy.mean,
do_correl=self.do_correl,
mcbs_enable=self.mcbs_enable)
avg_fluxdata[itarget] = (self.iter_range.iter_start,
self.iter_range.iter_stop, avg, lb_ci,
ub_ci, sterr, correl_len)
westpa.rc.pstatus('target {!r}:'.format(target_label))
westpa.rc.pstatus(
' correlation length = {} tau'.format(correl_len))
westpa.rc.pstatus(
' mean flux and CI = {:e} ({:e},{:e}) tau^(-1)'.format(
avg, lb_ci, ub_ci))
index[itarget]['mean_flux'] = avg
index[itarget]['mean_flux_ci_lb'] = lb_ci
index[itarget]['mean_flux_ci_ub'] = ub_ci
index[itarget]['mean_flux_correl_len'] = correl_len
# Write index and summary
index_ds = output_group.create_dataset('index', data=index)
index_ds.attrs['mcbs_alpha'] = self.alpha
index_ds.attrs['mcbs_autocorrel_alpha'] = self.autocorrel_alpha
index_ds.attrs['mcbs_n_sets'] = self.n_sets
self.fluxdata = fluxdata
self.output_h5file['avg_flux'] = avg_fluxdata
def _eval_block(iblock, istate, start, stop, state_pops, mcbs_alpha, mcbs_nsets, mcbs_acalpha):
ci_res = mcbs_ci_correl(state_pops,estimator=numpy.mean,alpha=mcbs_alpha,n_sets=mcbs_nsets,
autocorrel_alpha=mcbs_acalpha,subsample=numpy.mean)
return (iblock,istate,(start,stop)+ci_res)
def go(self):
pi = self.progress.indicator
with pi:
pi.new_operation('Initializing')
self.open_files()
nstates = self.assignments_file.attrs['nstates']
nbins = self.assignments_file.attrs['nbins']
state_labels = self.assignments_file['state_labels'][...]
assert nstates == len(state_labels)
start_iter, stop_iter, step_iter = self.iter_range.iter_start, self.iter_range.iter_stop, self.iter_range.iter_step
pi.new_operation('Reading data')
cond_fluxes = h5io.IterBlockedDataset(
self.kinetics_file['conditional_fluxes'])
cond_fluxes.cache_data()
total_fluxes = h5io.IterBlockedDataset(
self.kinetics_file['total_fluxes'])
pops = h5io.IterBlockedDataset(
self.assignments_file['labeled_populations'])
pops.cache_data()
pops.data = pops.data.sum(axis=2)
rates = h5io.IterBlockedDataset.empty_like(cond_fluxes)
rates.data = sequence_macro_flux_to_rate(
cond_fluxes.data, pops.data[:nstates, :nbins])
avg_total_fluxes = numpy.zeros((nstates, ), dtype=ci_dtype)
avg_conditional_fluxes = numpy.zeros((nstates, nstates),
dtype=ci_dtype)
avg_rates = numpy.zeros((nstates, nstates), dtype=ci_dtype)
# Calculate overall average rates
pi.new_operation('Averaging overall fluxes into states', nstates)
for istate in xrange(nstates):
ci_res = mcbs_ci_correl(total_fluxes.iter_slice(
start_iter, stop_iter)[:, istate],
estimator=numpy.mean,
alpha=self.mcbs_alpha,
n_sets=self.mcbs_nsets,
autocorrel_alpha=self.mcbs_acalpha,
subsample=numpy.mean)
avg_total_fluxes[istate] = (start_iter, stop_iter) + ci_res
pi.progress += 1
pi.new_operation('Averaging state-to-state fluxes and rates',
nstates * (nstates - 1))
for istate in xrange(nstates):
for jstate in xrange(nstates):
if istate == jstate: continue
flux_ci_res = mcbs_ci_correl(
cond_fluxes.iter_slice(start_iter,
stop_iter)[:, istate, jstate],
estimator=numpy.mean,
alpha=self.mcbs_alpha,
n_sets=self.mcbs_nsets,
autocorrel_alpha=self.mcbs_acalpha,
subsample=numpy.mean)
rate_ci_res = mcbs_ci_correl(
rates.iter_slice(start_iter, stop_iter)[:, istate,
jstate],
estimator=numpy.mean,
alpha=self.mcbs_alpha,
n_sets=self.mcbs_nsets,
autocorrel_alpha=self.mcbs_acalpha,
subsample=numpy.mean)
avg_conditional_fluxes[istate,
jstate] = (start_iter,
stop_iter) + flux_ci_res
avg_rates[istate,
jstate] = (start_iter, stop_iter) + rate_ci_res
pi.progress += 1
pi.new_operation('Saving averages')
self.output_file['avg_rates'] = avg_rates
self.output_file['avg_conditional_fluxes'] = avg_conditional_fluxes
self.output_file['avg_total_fluxes'] = avg_total_fluxes
for ds in ('avg_rates', 'avg_conditional_fluxes',
'avg_total_fluxes'):
self.stamp_mcbs_info(self.output_file[ds])
self.output_file['state_labels'] = state_labels
maxlabellen = max(map(len, state_labels))
pi.clear()
print('fluxes into macrostates:')
for istate in xrange(nstates):
print(
'{:{maxlabellen}s}: mean={:21.15e} CI=({:21.15e}, {:21.15e}) * tau^-1'
.format(state_labels[istate],
avg_total_fluxes['expected'][istate],
avg_total_fluxes['ci_lbound'][istate],
avg_total_fluxes['ci_ubound'][istate],
maxlabellen=maxlabellen))
print('\nfluxes from state to state:')
for istate in xrange(nstates):
for jstate in xrange(nstates):
if istate == jstate: continue
print(
'{:{maxlabellen}s} -> {:{maxlabellen}s}: mean={:21.15e} CI=({:21.15e}, {:21.15e}) * tau^-1'
.format(state_labels[istate],
state_labels[jstate],
avg_conditional_fluxes['expected'][istate,
jstate],
avg_conditional_fluxes['ci_lbound'][istate,
jstate],
avg_conditional_fluxes['ci_ubound'][istate,
jstate],
maxlabellen=maxlabellen))
print('\nrates from state to state:')
for istate in xrange(nstates):
for jstate in xrange(nstates):
if istate == jstate: continue
print(
'{:{maxlabellen}s} -> {:{maxlabellen}s}: mean={:21.15e} CI=({:21.15e}, {:21.15e}) * tau^-1'
.format(state_labels[istate],
state_labels[jstate],
avg_rates['expected'][istate, jstate],
avg_rates['ci_lbound'][istate, jstate],
avg_rates['ci_ubound'][istate, jstate],
maxlabellen=maxlabellen))
# skip evolution if not requested
if self.evolution_mode == 'none' or not step_iter: return
start_pts = range(start_iter, stop_iter, step_iter)
target_evol = numpy.zeros((len(start_pts), nstates),
dtype=ci_dtype)
flux_evol = numpy.zeros((len(start_pts), nstates, nstates),
dtype=ci_dtype)
rate_evol = numpy.zeros((len(start_pts), nstates, nstates),
dtype=ci_dtype)
pi.new_operation('Calculating flux/rate evolution', len(start_pts))
futures = []
for iblock, start in enumerate(start_pts):
stop = min(start + step_iter, stop_iter)
if self.evolution_mode == 'cumulative':
windowsize = int(self.evol_window_frac *
(stop - start_iter))
block_start = max(start_iter, stop - windowsize)
else: # self.evolution_mode == 'blocked'
block_start = start
future = self.work_manager.submit(
_eval_block,
kwargs=dict(iblock=iblock,
start=block_start,
stop=stop,
nstates=nstates,
total_fluxes=total_fluxes.iter_slice(
block_start, stop),
cond_fluxes=cond_fluxes.iter_slice(
block_start, stop),
rates=rates.iter_slice(block_start, stop),
mcbs_alpha=self.mcbs_alpha,
mcbs_nsets=self.mcbs_nsets,
mcbs_acalpha=self.mcbs_acalpha))
futures.append(future)
for future in self.work_manager.as_completed(futures):
pi.progress += 1
target_results, condflux_results, rate_results = future.get_result(
discard=True)
for result in target_results:
iblock, istate, ci_result = result
target_evol[iblock, istate] = ci_result
for result in condflux_results:
iblock, istate, jstate, ci_result = result
flux_evol[iblock, istate, jstate] = ci_result
for result in rate_results:
iblock, istate, jstate, ci_result = result
rate_evol[iblock, istate, jstate] = ci_result
df_ds = self.output_file.create_dataset(
'conditional_flux_evolution',
data=flux_evol,
shuffle=True,
compression=9)
tf_ds = self.output_file.create_dataset('target_flux_evolution',
data=target_evol,
shuffle=True,
compression=9)
rate_ds = self.output_file.create_dataset('rate_evolution',
data=rate_evol,
shuffle=True,
compression=9)
for ds in (df_ds, tf_ds, rate_ds):
self.stamp_mcbs_info(ds)
numpy.mean,
0.05,
n_sets=n_sets,
sort=numpy.msort)
n_uncorrel_width = uncorrel_ub - uncorrel_lb
n_correl_width = correl_ub - correl_lb
print(' uncorrelated: {} ({},{})'.format(uncorrel_mean, uncorrel_lb,
uncorrel_ub))
print(' correlated: {} ({},{})'.format(correl_mean, correl_lb,
correl_ub))
print(' width ratio c/u: {}'.format((n_correl_width / n_uncorrel_width)))
print('blocked MCBS:')
#uncorrel_mean, uncorrel_lb, uncorrel_ub = mcbs_ci(ds_uncorrel[::k_uncorrel+1], numpy.mean, 0.05, n_sets=1000, sort=numpy.msort)
#correl_mean, correl_lb, correl_ub = mcbs_ci(ds_correl[::k_correl+1], numpy.mean, 0.05, n_sets=1000, sort=numpy.msort)
uncorrel_mean, uncorrel_lb, uncorrel_ub, k_ = mcbs_ci_correl(
ds_uncorrel, numpy.mean, 0.05, n_sets=n_sets, subsample=numpy.mean)
correl_mean, correl_lb, correl_ub, k_ = mcbs_ci_correl(ds_correl,
numpy.mean,
0.05,
n_sets=n_sets,
subsample=numpy.mean)
b_uncorrel_width = uncorrel_ub - uncorrel_lb
b_correl_width = correl_ub - correl_lb
print(' uncorrelated: {} ({},{})'.format(uncorrel_mean, uncorrel_lb,
uncorrel_ub))
print(' correlated: {} ({},{})'.format(correl_mean, correl_lb,
correl_ub))
print(' width ratio c/u: {}'.format((b_correl_width / b_uncorrel_width)))
print('width ratio blocked/naive:')
print(' uncorrelated: {}'.format(b_uncorrel_width / n_uncorrel_width))
def calc_store_flux_data(self):
westpa.rc.pstatus('Calculating mean flux and confidence intervals for iterations [{},{})'
.format(self.iter_range.iter_start, self.iter_range.iter_stop))
fluxdata = extract_fluxes(self.iter_range.iter_start, self.iter_range.iter_stop, self.data_reader)
# Create a group to store data in
output_group = h5io.create_hdf5_group(self.output_h5file, 'target_flux', replace=False, creating_program=self.prog)
self.output_group = output_group
output_group.attrs['version_code'] = self.output_format_version
self.iter_range.record_data_iter_range(output_group)
n_targets = len(fluxdata)
index = numpy.empty((len(fluxdata),), dtype=target_index_dtype)
avg_fluxdata = numpy.empty((n_targets,), dtype=ci_dtype)
for itarget, (target_label, target_fluxdata) in enumerate(fluxdata.iteritems()):
# Create group and index entry
index[itarget]['target_label'] = str(target_label)
target_group = output_group.create_group('target_{}'.format(itarget))
self.target_groups[target_label] = target_group
# Store per-iteration values
target_group['n_iter'] = target_fluxdata['n_iter']
target_group['count'] = target_fluxdata['count']
target_group['flux'] = target_fluxdata['flux']
h5io.label_axes(target_group['flux'], ['n_iter'], units=['tau^-1'])
# Calculate flux autocorrelation
fluxes = target_fluxdata['flux']
mean_flux = fluxes.mean()
fmm = fluxes - mean_flux
acorr = fftconvolve(fmm,fmm[::-1])
acorr = acorr[len(acorr)//2:]
acorr /= acorr[0]
acorr_ds = target_group.create_dataset('flux_autocorrel', data=acorr)
h5io.label_axes(acorr_ds, ['lag'], ['tau'])
# Calculate overall averages and CIs
#avg, lb_ci, ub_ci, correl_len = mclib.mcbs_ci_correl(fluxes, numpy.mean, self.alpha, self.n_sets,
# autocorrel_alpha=self.autocorrel_alpha, subsample=numpy.mean)
avg, lb_ci, ub_ci, sterr, correl_len = mclib.mcbs_ci_correl({'dataset': fluxes}, estimator=(lambda stride, dataset: numpy.mean(dataset)), alpha=self.alpha, n_sets=self.n_sets,
autocorrel_alpha=self.autocorrel_alpha, subsample=numpy.mean, do_correl=self.do_correl, mcbs_enable=self.mcbs_enable )
avg_fluxdata[itarget] = (self.iter_range.iter_start, self.iter_range.iter_stop, avg, lb_ci, ub_ci, sterr, correl_len)
westpa.rc.pstatus('target {!r}:'.format(target_label))
westpa.rc.pstatus(' correlation length = {} tau'.format(correl_len))
westpa.rc.pstatus(' mean flux and CI = {:e} ({:e},{:e}) tau^(-1)'.format(avg,lb_ci,ub_ci))
index[itarget]['mean_flux'] = avg
index[itarget]['mean_flux_ci_lb'] = lb_ci
index[itarget]['mean_flux_ci_ub'] = ub_ci
index[itarget]['mean_flux_correl_len'] = correl_len
# Write index and summary
index_ds = output_group.create_dataset('index', data=index)
index_ds.attrs['mcbs_alpha'] = self.alpha
index_ds.attrs['mcbs_autocorrel_alpha'] = self.autocorrel_alpha
index_ds.attrs['mcbs_n_sets'] = self.n_sets
self.fluxdata = fluxdata
self.output_h5file['avg_flux'] = avg_fluxdata