Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

stationary estimator and smapler #2

Open
wants to merge 5 commits into
base: master
Choose a base branch
from
Open

stationary estimator and smapler #2

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
dcad614
set estimator and sampler to stationary=True
ChayaSt Mar 29, 2016
6fcb175
set estimator and sampler to stationary=True
ChayaSt Mar 29, 2016
8cfbc4d
set estimator to stationary=True
ChayaSt Mar 30, 2016
40db5f1
generated new statistics table
ChayaSt Apr 6, 2016
7c5724d
added versioneer.py
ChayaSt Sep 27, 2016
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions examples/p5ab-hairpin/generate-figure.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -33,7 +33,7 @@ def run(nstates, nsamples):

# Initialize MLHMM.
print "Initializing MLHMM with "+str(nstates)+" states."
estimator = bhmm.MLHMM(O, nstates)
estimator = bhmm.MLHMM(O, nstates,stationary=True)

# Plot initial guess.
plots.plot_state_assignments(estimator.hmm, None, O[0], time_units=time_units, obs_label=obs_label, tau=tau,
Expand All @@ -49,7 +49,7 @@ def run(nstates, nsamples):

# Initialize BHMM, using MLHMM model as initial model.
print "Initializing BHMM and running with "+str(nsamples)+" samples."
sampler = bhmm.BHMM(O, nstates, initial_model=mle)
sampler = bhmm.BHMM(O, nstates, initial_model=mle, stationary=True)

# Sample models.
bhmm_models = sampler.sample(nsamples=nsamples, save_hidden_state_trajectory=False)
Expand Down
4 changes: 2 additions & 2 deletions examples/p5ab-hairpin/hairpin-example.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,7 +34,7 @@ def run(nstates, nsamples):

# Initialize MLHMM.
print "Initializing MLHMM with "+str(nstates)+" states."
estimator = bhmm.MLHMM(O, nstates)
estimator = bhmm.MLHMM(O, nstates, stationary=True)

# Plot initial guess.
plots.plot_state_assignments(estimator.hmm, None, O[0], time_units=time_units, obs_label=obs_label, tau=tau,
Expand All @@ -50,7 +50,7 @@ def run(nstates, nsamples):

# Initialize BHMM, using MLHMM model as initial model.
print "Initializing BHMM and running with "+str(nsamples)+" samples."
sampler = bhmm.BHMM(O, nstates, initial_model=mle)
sampler = bhmm.BHMM(O, nstates, initial_model=mle, stationary=True)

# Sample models.
bhmm_models = sampler.sample(nsamples=nsamples, save_hidden_state_trajectory=False)
Expand Down
2 changes: 1 addition & 1 deletion examples/rnase-h-d10a/generate-figure.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -30,7 +30,7 @@ def run(nstates, nsamples):

# Initialize MLHMM.
print "Initializing MLHMM with "+str(nstates)+" states."
estimator = bhmm.MLHMM(O, nstates)
estimator = bhmm.MLHMM(O, nstates, stationary=True)

# Plot initial guess.
plots.plot_state_assignments(estimator.hmm, None, O[0], time_units=time_units, obs_label=obs_label, tau=tau,
Expand Down
6 changes: 4 additions & 2 deletions examples/synthetic-three-state-model/generate-figure.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -64,7 +64,7 @@ def analyze_data(O, nstates, nsamples=1000, nobservations=None):

# Generate MLHMM.
print "Generating MLHMM..."
estimator = bhmm.MLHMM(O, nstates)
estimator = bhmm.MLHMM(O, nstates, stationary=True)

print "Initial guess:"
print str(estimator.hmm.output_model)
Expand All @@ -89,7 +89,7 @@ def analyze_data(O, nstates, nsamples=1000, nobservations=None):

# Initialize BHMM with MLHMM model.
print "Sampling models from BHMM..."
sampler = bhmm.BHMM(O, nstates, initial_model=mle)
sampler = bhmm.BHMM(O, nstates, initial_model=mle, stationary=True)
bhmm_models = sampler.sample(nsamples=nsamples, save_hidden_state_trajectory=False)

# Generate a sample saving a hidden state trajectory.
Expand Down Expand Up @@ -164,6 +164,8 @@ def generate_latex_table(true_hmm, sampled_hmm_list, conf=0.95, dt=1, time_unit=
table += '\t\tEquilibrium probability '
table += '\t\t& $\pi_{%d}$ & $%0.3f$' % (i+1, true_hmm.stationary_distribution[i])
for sampled_hmm in sampled_hmm_list:
print sampled_hmm
print 'trying to find distribution mean'
p = sampled_hmm.stationary_distribution_mean
p_lo, p_hi = sampled_hmm.stationary_distribution_conf
table += ' & $%0.3f_{\:%0.3f}^{\:%0.3f}$ ' % (p[i], p_lo[i], p_hi[i])
Expand Down
6 changes: 3 additions & 3 deletions examples/synthetic-three-state-model/synthetic-example.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,7 @@
import argparse

import bhmm
from bhmm.util import testsystems
import bhmm.util.testsystems as testsystems
from bhmm.util.analysis import generate_latex_table


Expand Down Expand Up @@ -36,7 +36,7 @@ def run(nstates, nsamples):

# Generate MLHMM.
print "Generating MLHMM..."
estimator = bhmm.MLHMM(O, nstates)
estimator = bhmm.MLHMM(O, nstates, stationary=True)

print "Initial guess:"
print str(estimator.hmm.output_model)
Expand All @@ -61,7 +61,7 @@ def run(nstates, nsamples):

# Initialize BHMM with MLHMM model.
print "Sampling models from BHMM..."
sampler = bhmm.BHMM(O, nstates, initial_model=mle)
sampler = bhmm.BHMM(O, nstates, initial_model=mle, stationary=True)
bhmm_models = sampler.sample(nsamples=nsamples, save_hidden_state_trajectory=False)

# Generate a sample saving a hidden state trajectory.
Expand Down
74 changes: 37 additions & 37 deletions examples/synthetic-three-state-model/synthetic-three-state-model-bhmm-statistics.tex
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,43 +6,43 @@
\hline
& & & \multicolumn{3}{c}{\bf Estimated Model Parameters} \\ \cline{4-6}
\multicolumn{2}{l}{\bf Property} & \bf True Value & \bf 1 000 observations & \bf 10 000 observations & \bf 100 000 observations\\ \hline
Equilibrium probability & $\pi_{1}$ & $0.385$ & $0.678_{\:0.367}^{\:0.889}$ & $0.533_{\:0.442}^{\:0.629}$ & $0.517_{\:0.486}^{\:0.549}$ \\
& $\pi_{2}$ & $0.094$ & $0.246_{\:0.069}^{\:0.546}$ & $0.351_{\:0.264}^{\:0.434}$ & $0.381_{\:0.353}^{\:0.410}$ \\
& $\pi_{3}$ & $0.521$ & $0.077_{\:0.029}^{\:0.156}$ & $0.116_{\:0.091}^{\:0.144}$ & $0.101_{\:0.093}^{\:0.110}$ \\
\hline
Transition probability ($\Delta t = $1.0 ms) & $T_{11}$ & $0.980$ & $0.991_{\:0.981}^{\:0.997}$ & $0.989_{\:0.986}^{\:0.992}$ & $0.989_{\:0.988}^{\:0.990}$ \\
& $T_{12}$ & $0.015$ & $0.001_{\:0.000}^{\:0.006}$ & $0.003_{\:0.002}^{\:0.005}$ & $0.004_{\:0.003}^{\:0.004}$ \\
& $T_{13}$ & $0.005$ & $0.009_{\:0.003}^{\:0.018}$ & $0.008_{\:0.006}^{\:0.010}$ & $0.007_{\:0.006}^{\:0.008}$ \\
& $T_{21}$ & $0.063$ & $0.003_{\:0.000}^{\:0.017}$ & $0.005_{\:0.003}^{\:0.008}$ & $0.005_{\:0.004}^{\:0.005}$ \\
& $T_{22}$ & $0.900$ & $0.967_{\:0.932}^{\:0.988}$ & $0.976_{\:0.970}^{\:0.982}$ & $0.978_{\:0.977}^{\:0.980}$ \\
& $T_{23}$ & $0.037$ & $0.030_{\:0.010}^{\:0.068}$ & $0.019_{\:0.014}^{\:0.025}$ & $0.017_{\:0.015}^{\:0.018}$ \\
& $T_{31}$ & $0.003$ & $0.079_{\:0.029}^{\:0.146}$ & $0.036_{\:0.026}^{\:0.047}$ & $0.036_{\:0.033}^{\:0.040}$ \\
& $T_{32}$ & $0.007$ & $0.093_{\:0.028}^{\:0.193}$ & $0.057_{\:0.043}^{\:0.072}$ & $0.063_{\:0.058}^{\:0.068}$ \\
& $T_{33}$ & $0.990$ & $0.828_{\:0.711}^{\:0.920}$ & $0.907_{\:0.887}^{\:0.925}$ & $0.901_{\:0.894}^{\:0.907}$ \\
\hline
\hline
Transition rate (ms$^{-1}$) & $k_{12}$ & $0.015$ & $0.001_{\:0.000}^{\:0.006}$ & $0.003_{\:0.002}^{\:0.005}$ & $0.004_{\:0.003}^{\:0.004}$ \\
& $k_{13}$ & $0.005$ & $0.009_{\:0.003}^{\:0.018}$ & $0.008_{\:0.006}^{\:0.010}$ & $0.007_{\:0.006}^{\:0.008}$ \\
& $k_{21}$ & $0.063$ & $0.003_{\:0.000}^{\:0.017}$ & $0.005_{\:0.003}^{\:0.008}$ & $0.005_{\:0.004}^{\:0.005}$ \\
& $k_{23}$ & $0.037$ & $0.030_{\:0.010}^{\:0.068}$ & $0.019_{\:0.014}^{\:0.025}$ & $0.017_{\:0.015}^{\:0.018}$ \\
& $k_{31}$ & $0.003$ & $0.079_{\:0.029}^{\:0.146}$ & $0.036_{\:0.026}^{\:0.047}$ & $0.036_{\:0.033}^{\:0.040}$ \\
& $k_{32}$ & $0.007$ & $0.093_{\:0.028}^{\:0.193}$ & $0.057_{\:0.043}^{\:0.072}$ & $0.063_{\:0.058}^{\:0.068}$ \\
\hline
State mean lifetime (ms) & $t_{1}$ & $49.498$ & $128.116_{\:55.334}^{\:309.568}$ & $91.902_{\:70.769}^{\:120.147}$ & $93.429_{\:86.014}^{\:101.962}$ \\
& $t_{2}$ & $9.491$ & $36.448_{\:15.042}^{\:92.972}$ & $42.545_{\:33.175}^{\:54.544}$ & $45.889_{\:42.634}^{\:49.415}$ \\
& $t_{3}$ & $99.499$ & $5.986_{\:3.065}^{\:13.200}$ & $10.396_{\:8.396}^{\:12.904}$ & $9.555_{\:8.969}^{\:10.219}$ \\
\hline
Relaxation time (ms) & $\tau_{1}$ & $56.704$ & $54.467_{\:26.737}^{\:109.929}$ & $51.201_{\:41.887}^{\:62.042}$ & $53.387_{\:50.301}^{\:56.513}$ \\
& $\tau_{2}$ & $8.377$ & $5.191_{\:2.656}^{\:10.780}$ & $8.855_{\:7.255}^{\:10.815}$ & $8.335_{\:7.857}^{\:8.866}$ \\
\hline
\hline
State force mean (pN) & $\mu_{1}$ & $3.000$ & $5.603_{\:5.588}^{\:5.619}$ & $5.600_{\:5.595}^{\:5.606}$ & $5.599_{\:5.597}^{\:5.600}$ \\
& $\mu_{2}$ & $4.700$ & $2.950_{\:2.818}^{\:3.085}$ & $3.003_{\:2.970}^{\:3.039}$ & $2.996_{\:2.986}^{\:3.007}$ \\
& $\mu_{3}$ & $5.600$ & $4.706_{\:4.598}^{\:4.822}$ & $4.712_{\:4.694}^{\:4.731}$ & $4.700_{\:4.694}^{\:4.706}$ \\
\hline
State force std dev (pN) & $s_{1}$ & $1.000$ & $0.198_{\:0.188}^{\:0.209}$ & $0.198_{\:0.195}^{\:0.202}$ & $0.200_{\:0.198}^{\:0.201}$ \\
& $s_{2}$ & $0.300$ & $1.011_{\:0.918}^{\:1.114}$ & $0.994_{\:0.970}^{\:1.018}$ & $0.998_{\:0.991}^{\:1.005}$ \\
& $s_{3}$ & $0.200$ & $0.417_{\:0.290}^{\:0.524}$ & $0.291_{\:0.277}^{\:0.305}$ & $0.298_{\:0.294}^{\:0.303}$ \\
Equilibrium probability & $\pi_{1}$ & $0.385$ & $0.609_{\:0.247}^{\:0.896}$ & $0.576_{\:0.479}^{\:0.672}$ & $0.527_{\:0.495}^{\:0.559}$ \\
& $\pi_{2}$ & $0.094$ & $0.124_{\:0.039}^{\:0.270}$ & $0.323_{\:0.240}^{\:0.416}$ & $0.093_{\:0.085}^{\:0.101}$ \\
& $\pi_{3}$ & $0.521$ & $0.267_{\:0.047}^{\:0.613}$ & $0.101_{\:0.076}^{\:0.132}$ & $0.379_{\:0.351}^{\:0.409}$ \\
\hline
Transition probability ($\Delta t = $1.0 ms) & $T_{11}$ & $0.980$ & $0.993_{\:0.985}^{\:0.998}$ & $0.990_{\:0.987}^{\:0.993}$ & $0.990_{\:0.989}^{\:0.991}$ \\
& $T_{12}$ & $0.015$ & $0.006_{\:0.002}^{\:0.014}$ & $0.003_{\:0.002}^{\:0.004}$ & $0.006_{\:0.006}^{\:0.007}$ \\
& $T_{13}$ & $0.005$ & $0.001_{\:0.000}^{\:0.004}$ & $0.007_{\:0.005}^{\:0.009}$ & $0.003_{\:0.003}^{\:0.004}$ \\
& $T_{21}$ & $0.063$ & $0.032_{\:0.010}^{\:0.074}$ & $0.005_{\:0.003}^{\:0.008}$ & $0.036_{\:0.033}^{\:0.040}$ \\
& $T_{22}$ & $0.900$ & $0.933_{\:0.880}^{\:0.970}$ & $0.977_{\:0.971}^{\:0.982}$ & $0.901_{\:0.894}^{\:0.908}$ \\
& $T_{23}$ & $0.037$ & $0.035_{\:0.009}^{\:0.084}$ & $0.018_{\:0.014}^{\:0.023}$ & $0.063_{\:0.058}^{\:0.068}$ \\
& $T_{31}$ & $0.003$ & $0.002_{\:0.000}^{\:0.011}$ & $0.039_{\:0.029}^{\:0.051}$ & $0.005_{\:0.004}^{\:0.005}$ \\
& $T_{32}$ & $0.007$ & $0.017_{\:0.005}^{\:0.040}$ & $0.058_{\:0.045}^{\:0.075}$ & $0.015_{\:0.014}^{\:0.017}$ \\
& $T_{33}$ & $0.990$ & $0.981_{\:0.956}^{\:0.994}$ & $0.903_{\:0.882}^{\:0.921}$ & $0.980_{\:0.978}^{\:0.981}$ \\
\hline
\hline
Transition rate (ms$^{-1}$) & $k_{12}$ & $0.015$ & $0.006_{\:0.002}^{\:0.014}$ & $0.003_{\:0.002}^{\:0.004}$ & $0.006_{\:0.006}^{\:0.007}$ \\
& $k_{13}$ & $0.005$ & $0.001_{\:0.000}^{\:0.004}$ & $0.007_{\:0.005}^{\:0.009}$ & $0.003_{\:0.003}^{\:0.004}$ \\
& $k_{21}$ & $0.063$ & $0.032_{\:0.010}^{\:0.074}$ & $0.005_{\:0.003}^{\:0.008}$ & $0.036_{\:0.033}^{\:0.040}$ \\
& $k_{23}$ & $0.037$ & $0.035_{\:0.009}^{\:0.084}$ & $0.018_{\:0.014}^{\:0.023}$ & $0.063_{\:0.058}^{\:0.068}$ \\
& $k_{31}$ & $0.003$ & $0.002_{\:0.000}^{\:0.011}$ & $0.039_{\:0.029}^{\:0.051}$ & $0.005_{\:0.004}^{\:0.005}$ \\
& $k_{32}$ & $0.007$ & $0.017_{\:0.005}^{\:0.040}$ & $0.058_{\:0.045}^{\:0.075}$ & $0.015_{\:0.014}^{\:0.017}$ \\
\hline
State mean lifetime (ms) & $t_{1}$ & $49.498$ & $185.195_{\:68.381}^{\:556.289}$ & $103.364_{\:78.952}^{\:136.436}$ & $101.562_{\:92.914}^{\:111.068}$ \\
& $t_{2}$ & $9.491$ & $16.424_{\:8.007}^{\:36.775}$ & $42.709_{\:33.876}^{\:54.814}$ & $9.574_{\:8.919}^{\:10.348}$ \\
& $t_{3}$ & $99.499$ & $65.388_{\:23.763}^{\:168.941}$ & $9.914_{\:8.011}^{\:12.252}$ & $49.227_{\:45.544}^{\:53.111}$ \\
\hline
Relaxation time (ms) & $\tau_{1}$ & $56.704$ & $84.338_{\:40.682}^{\:174.791}$ & $52.374_{\:43.244}^{\:63.413}$ & $56.871_{\:53.479}^{\:60.398}$ \\
& $\tau_{2}$ & $8.377$ & $13.226_{\:6.719}^{\:25.925}$ & $8.564_{\:7.041}^{\:10.344}$ & $8.443_{\:7.908}^{\:9.080}$ \\
\hline
\hline
State force mean (pN) & $\mu_{1}$ & $3.000$ & $5.596_{\:5.581}^{\:5.611}$ & $5.599_{\:5.594}^{\:5.604}$ & $5.600_{\:5.599}^{\:5.602}$ \\
& $\mu_{2}$ & $4.700$ & $4.659_{\:4.604}^{\:4.715}$ & $3.005_{\:2.969}^{\:3.041}$ & $4.699_{\:4.692}^{\:4.705}$ \\
& $\mu_{3}$ & $5.600$ & $2.967_{\:2.842}^{\:3.087}$ & $4.702_{\:4.682}^{\:4.721}$ & $2.999_{\:2.988}^{\:3.008}$ \\
\hline
State force std dev (pN) & $s_{1}$ & $1.000$ & $0.192_{\:0.181}^{\:0.204}$ & $0.200_{\:0.196}^{\:0.204}$ & $0.200_{\:0.199}^{\:0.202}$ \\
& $s_{2}$ & $0.300$ & $0.320_{\:0.284}^{\:0.362}$ & $1.009_{\:0.985}^{\:1.033}$ & $0.302_{\:0.297}^{\:0.306}$ \\
& $s_{3}$ & $0.200$ & $1.074_{\:0.990}^{\:1.164}$ & $0.307_{\:0.292}^{\:0.323}$ & $0.997_{\:0.990}^{\:1.005}$ \\
\hline
\hline
\end{tabular*}
Expand Down
Loading