Save and load results as HDF5 files

# install if not done yet
# %pip install pypesto --quiet

import tempfile

import matplotlib.pyplot as plt
import numpy as np
import scipy as sp

import pypesto
import pypesto.optimize as optimize
import pypesto.profile as profile
import pypesto.sample as sample
import pypesto.store as store

%matplotlib inline

In this notebook, we will demonstrate how to save and (re-)load optimization results, profile results and sampling results to an .hdf5 file. The use case of this notebook is to generate visualizations from reloaded result objects.

Define the objective and problem

objective = pypesto.Objective(
    fun=sp.optimize.rosen,
    grad=sp.optimize.rosen_der,
    hess=sp.optimize.rosen_hess,
)

dim_full = 20
lb = -5 * np.ones((dim_full, 1))
ub = 5 * np.ones((dim_full, 1))

problem = pypesto.Problem(objective=objective, lb=lb, ub=ub)

Fill result object with profile, sample, optimization

# create optimizers
optimizer = optimize.ScipyOptimizer()

# set number of starts
n_starts = 10

# Optimization
result = pypesto.optimize.minimize(
    problem=problem, optimizer=optimizer, n_starts=n_starts, filename=None
)
# Profiling
result = profile.parameter_profile(
    problem=problem, result=result, optimizer=optimizer, filename=None
)
# Sampling
sampler = sample.AdaptiveMetropolisSampler()
result = sample.sample(
    problem=problem,
    sampler=sampler,
    n_samples=100000,
    result=result,
    filename=None,
)

  0%|          | 0/10 [00:00<?, ?it/s]Executing task 0.
Final fval=0.0000, time=0.0355s, n_fval=154.
Executing task 1.
Final fval=0.0000, time=0.0317s, n_fval=145.
Executing task 2.
Final fval=0.0000, time=0.0326s, n_fval=161.
 30%|███       | 3/10 [00:00<00:00, 28.87it/s]Executing task 3.
Final fval=0.0000, time=0.0275s, n_fval=144.
Executing task 4.
Final fval=0.0000, time=0.0295s, n_fval=154.
Executing task 5.
Final fval=0.0000, time=0.0258s, n_fval=133.
Executing task 6.
Final fval=0.0000, time=0.0281s, n_fval=140.
 70%|███████   | 7/10 [00:00<00:00, 32.19it/s]Executing task 7.
Final fval=3.9866, time=0.0256s, n_fval=118.
Executing task 8.
Final fval=3.9866, time=0.0297s, n_fval=150.
Executing task 9.
Final fval=3.9866, time=0.0281s, n_fval=114.
100%|██████████| 10/10 [00:00<00:00, 32.28it/s]
  0%|          | 0/20 [00:00<?, ?it/s]Executing task 0.
Final fval=0.0784, time=0.0043s, n_fval=19.
Final fval=0.2225, time=0.0019s, n_fval=7.
Final fval=0.4391, time=0.0017s, n_fval=7.
Final fval=0.7634, time=0.0014s, n_fval=3.
Final fval=1.2500, time=0.0014s, n_fval=5.
Final fval=1.9800, time=0.0013s, n_fval=5.
Final fval=0.1040, time=0.0013s, n_fval=4.
Final fval=0.2610, time=0.0013s, n_fval=4.
Final fval=0.4966, time=0.0015s, n_fval=5.
Final fval=0.8497, time=0.0016s, n_fval=4.
Final fval=1.3791, time=0.0014s, n_fval=5.
Final fval=2.1730, time=0.0012s, n_fval=3.
Executing task 1.
Final fval=0.0625, time=0.0040s, n_fval=18.
Final fval=0.1990, time=0.0017s, n_fval=7.
Final fval=0.4038, time=0.0022s, n_fval=7.
Final fval=0.7105, time=0.0014s, n_fval=4.
Final fval=1.1706, time=0.0013s, n_fval=5.
Final fval=1.8608, time=0.0013s, n_fval=5.
Final fval=2.8962, time=0.0018s, n_fval=7.
Final fval=0.1049, time=0.0014s, n_fval=5.
Final fval=0.2623, time=0.0013s, n_fval=4.
Final fval=0.4986, time=0.0013s, n_fval=5.
Final fval=0.8527, time=0.0012s, n_fval=3.
Final fval=1.3836, time=0.0018s, n_fval=6.
Final fval=2.1797, time=0.0015s, n_fval=5.
 10%|█         | 2/20 [00:00<00:01, 11.41it/s]Executing task 2.
Final fval=0.0625, time=0.0041s, n_fval=18.
Final fval=0.1990, time=0.0017s, n_fval=7.
Final fval=0.4038, time=0.0022s, n_fval=8.
Final fval=0.7105, time=0.0020s, n_fval=6.
Final fval=1.1707, time=0.0012s, n_fval=4.
Final fval=1.8609, time=0.0018s, n_fval=4.
Final fval=2.8964, time=0.0016s, n_fval=5.
Final fval=0.1049, time=0.0036s, n_fval=7.
Final fval=0.2623, time=0.0011s, n_fval=3.
Final fval=0.4986, time=0.0018s, n_fval=6.
Final fval=0.8527, time=0.0014s, n_fval=4.
Final fval=1.3836, time=0.0017s, n_fval=5.
Final fval=2.1797, time=0.0018s, n_fval=5.
Executing task 3.
Final fval=0.0625, time=0.0062s, n_fval=20.
Final fval=0.1990, time=0.0022s, n_fval=8.
Final fval=0.4038, time=0.0027s, n_fval=7.
Final fval=0.7105, time=0.0011s, n_fval=3.
Final fval=1.1707, time=0.0019s, n_fval=6.
Final fval=1.8609, time=0.0018s, n_fval=7.
Final fval=2.8964, time=0.0015s, n_fval=4.
Final fval=0.1049, time=0.0017s, n_fval=5.
Final fval=0.2623, time=0.0011s, n_fval=3.
Final fval=0.4986, time=0.0011s, n_fval=4.
Final fval=0.8527, time=0.0016s, n_fval=6.
Final fval=1.3836, time=0.0012s, n_fval=4.
Final fval=2.1797, time=0.0020s, n_fval=4.
 20%|██        | 4/20 [00:00<00:01,  9.79it/s]Executing task 4.
Final fval=0.0625, time=0.0051s, n_fval=19.
Final fval=0.1990, time=0.0025s, n_fval=8.
Final fval=0.4038, time=0.0024s, n_fval=8.
Final fval=0.7105, time=0.0012s, n_fval=4.
Final fval=1.1707, time=0.0041s, n_fval=17.
Final fval=1.8609, time=0.0052s, n_fval=19.
Final fval=2.8964, time=0.0013s, n_fval=4.
Final fval=0.1049, time=0.0065s, n_fval=18.
Final fval=0.2623, time=0.0055s, n_fval=17.
Final fval=0.4986, time=0.0014s, n_fval=4.
Final fval=0.8527, time=0.0052s, n_fval=18.
Final fval=1.3836, time=0.0044s, n_fval=16.
Final fval=2.1797, time=0.0040s, n_fval=18.
 25%|██▌       | 5/20 [00:00<00:01,  9.11it/s]Executing task 5.
Final fval=0.0625, time=0.0048s, n_fval=18.
Final fval=0.1990, time=0.0067s, n_fval=20.
Final fval=0.4038, time=0.0029s, n_fval=8.
Final fval=0.7106, time=0.0051s, n_fval=15.
Final fval=1.1707, time=0.0046s, n_fval=18.
Final fval=1.8609, time=0.0024s, n_fval=6.
Final fval=2.8964, time=0.0092s, n_fval=18.
Final fval=0.1049, time=0.0072s, n_fval=16.
Final fval=0.2623, time=0.0074s, n_fval=16.
Final fval=0.4986, time=0.0024s, n_fval=6.
Final fval=0.8526, time=0.0059s, n_fval=18.
Final fval=1.3835, time=0.0053s, n_fval=17.
Final fval=2.1796, time=0.0013s, n_fval=4.
 30%|███       | 6/20 [00:00<00:01,  8.01it/s]Executing task 6.
Final fval=0.0625, time=0.0141s, n_fval=31.
Final fval=0.1990, time=0.0029s, n_fval=8.
Final fval=0.4038, time=0.0019s, n_fval=8.
Final fval=0.7106, time=0.0012s, n_fval=3.
Final fval=1.1707, time=0.0024s, n_fval=4.
Final fval=1.8609, time=0.0029s, n_fval=4.
Final fval=2.8964, time=0.0023s, n_fval=4.
Final fval=0.1049, time=0.0016s, n_fval=4.
Final fval=0.2623, time=0.0019s, n_fval=4.
Final fval=0.4986, time=0.0012s, n_fval=3.
Final fval=0.8527, time=0.0013s, n_fval=5.
Final fval=1.3836, time=0.0013s, n_fval=4.
Final fval=2.1797, time=0.0017s, n_fval=3.
 35%|███▌      | 7/20 [00:00<00:01,  7.55it/s]Executing task 7.
Final fval=0.0625, time=0.0110s, n_fval=29.
Final fval=0.1990, time=0.0033s, n_fval=8.
Final fval=0.4038, time=0.0034s, n_fval=8.
Final fval=0.7105, time=0.0019s, n_fval=3.
Final fval=1.1707, time=0.0014s, n_fval=4.
Final fval=1.8609, time=0.0016s, n_fval=5.
Final fval=2.8964, time=0.0014s, n_fval=4.
Final fval=0.1049, time=0.0020s, n_fval=4.
Final fval=0.2623, time=0.0029s, n_fval=3.
Final fval=0.4986, time=0.0019s, n_fval=5.
Final fval=0.8527, time=0.0018s, n_fval=4.
Final fval=1.3836, time=0.0023s, n_fval=5.
Final fval=2.1797, time=0.0027s, n_fval=4.
 40%|████      | 8/20 [00:00<00:01,  7.45it/s]Executing task 8.
Final fval=0.0625, time=0.0093s, n_fval=28.
Final fval=0.1990, time=0.0031s, n_fval=8.
Final fval=0.4038, time=0.0024s, n_fval=8.
Final fval=0.7105, time=0.0014s, n_fval=3.
Final fval=1.1706, time=0.0035s, n_fval=7.
Final fval=1.8608, time=0.0026s, n_fval=6.
Final fval=2.8963, time=0.0025s, n_fval=4.
Final fval=0.1049, time=0.0026s, n_fval=7.
Final fval=0.2623, time=0.0011s, n_fval=3.
Final fval=0.4986, time=0.0023s, n_fval=4.
Final fval=0.8527, time=0.0018s, n_fval=4.
Final fval=1.3836, time=0.0016s, n_fval=5.
Final fval=2.1797, time=0.0022s, n_fval=5.
 45%|████▌     | 9/20 [00:01<00:01,  7.46it/s]Executing task 9.
Final fval=0.0624, time=0.0066s, n_fval=29.
Final fval=0.1990, time=0.0036s, n_fval=8.
Final fval=0.4037, time=0.0031s, n_fval=8.
Final fval=0.7104, time=0.0021s, n_fval=3.
Final fval=1.1704, time=0.0050s, n_fval=17.
Final fval=1.8605, time=0.0045s, n_fval=18.
Final fval=2.8958, time=0.0023s, n_fval=4.
Final fval=0.1049, time=0.0068s, n_fval=20.
Final fval=0.2623, time=0.0020s, n_fval=6.
Final fval=0.4986, time=0.0050s, n_fval=18.
Final fval=0.8526, time=0.0055s, n_fval=20.
Final fval=1.3835, time=0.0016s, n_fval=6.
Final fval=2.1797, time=0.0053s, n_fval=16.
 50%|█████     | 10/20 [00:01<00:01,  7.28it/s]Executing task 10.
Final fval=0.0623, time=0.0074s, n_fval=33.
Final fval=0.1987, time=0.0024s, n_fval=9.
Final fval=0.4033, time=0.0024s, n_fval=9.
Final fval=0.7099, time=0.0043s, n_fval=16.
Final fval=1.1696, time=0.0067s, n_fval=20.
Final fval=1.8594, time=0.0017s, n_fval=7.
Final fval=2.8941, time=0.0046s, n_fval=20.
Final fval=0.1049, time=0.0077s, n_fval=22.
Final fval=0.2622, time=0.0050s, n_fval=20.
Final fval=0.4985, time=0.0019s, n_fval=8.
Final fval=0.8525, time=0.0050s, n_fval=19.
Final fval=1.3834, time=0.0047s, n_fval=21.
Final fval=2.1794, time=0.0037s, n_fval=17.
 55%|█████▌    | 11/20 [00:01<00:01,  7.19it/s]Executing task 11.
Final fval=0.0616, time=0.0104s, n_fval=30.
Final fval=0.1977, time=0.0048s, n_fval=11.
Final fval=0.4016, time=0.0075s, n_fval=25.
Final fval=0.7073, time=0.0072s, n_fval=21.
Final fval=1.1658, time=0.0060s, n_fval=20.
Final fval=1.8536, time=0.0069s, n_fval=21.
Final fval=2.8855, time=0.0055s, n_fval=18.
Final fval=0.1048, time=0.0052s, n_fval=21.
Final fval=0.2621, time=0.0019s, n_fval=6.
Final fval=0.4984, time=0.0044s, n_fval=16.
Final fval=0.8523, time=0.0015s, n_fval=5.
Final fval=1.3830, time=0.0029s, n_fval=8.
Final fval=2.1788, time=0.0050s, n_fval=19.
 60%|██████    | 12/20 [00:01<00:01,  6.71it/s]Executing task 12.
Final fval=0.0591, time=0.0065s, n_fval=29.
Final fval=0.1924, time=0.0066s, n_fval=24.
Final fval=0.3924, time=0.0071s, n_fval=23.
Final fval=0.6933, time=0.0027s, n_fval=9.
Final fval=1.1439, time=0.0077s, n_fval=22.
Final fval=1.8179, time=0.0028s, n_fval=10.
Final fval=2.7805, time=0.0090s, n_fval=28.
Final fval=0.1021, time=0.0037s, n_fval=10.
Final fval=0.2572, time=0.0033s, n_fval=10.
Final fval=0.4897, time=0.0063s, n_fval=22.
Final fval=0.8391, time=0.0046s, n_fval=10.
Final fval=1.3631, time=0.0039s, n_fval=9.
Final fval=2.1490, time=0.0033s, n_fval=9.
 65%|██████▌   | 13/20 [00:01<00:01,  6.50it/s]Executing task 13.
Final fval=0.0507, time=0.0125s, n_fval=29.
Final fval=0.1662, time=0.0088s, n_fval=25.
Final fval=0.3382, time=0.0071s, n_fval=24.
Final fval=0.6094, time=0.0072s, n_fval=19.
Final fval=0.9915, time=0.0062s, n_fval=18.
Final fval=1.5142, time=0.0054s, n_fval=20.
Final fval=1.9433, time=0.0101s, n_fval=25.
Final fval=0.0824, time=0.0095s, n_fval=23.
Final fval=0.2219, time=0.0108s, n_fval=24.
Final fval=0.4289, time=0.0060s, n_fval=25.
Final fval=0.7446, time=0.0027s, n_fval=10.
Final fval=1.2200, time=0.0049s, n_fval=23.
Final fval=1.9333, time=0.0021s, n_fval=9.
 70%|███████   | 14/20 [00:01<00:00,  6.11it/s]Executing task 14.
Final fval=0.0322, time=0.0061s, n_fval=30.
Final fval=0.1082, time=0.0050s, n_fval=24.
Final fval=0.2237, time=0.0045s, n_fval=23.
Final fval=0.4400, time=0.0037s, n_fval=18.
Final fval=0.7550, time=0.0048s, n_fval=22.
Final fval=1.1701, time=0.0042s, n_fval=21.
Final fval=1.6074, time=0.0054s, n_fval=24.
Final fval=1.9770, time=0.0046s, n_fval=21.
Final fval=0.0623, time=0.0051s, n_fval=24.
Final fval=0.1820, time=0.0044s, n_fval=22.
Final fval=0.3765, time=0.0054s, n_fval=24.
Final fval=0.6670, time=0.0049s, n_fval=23.
Final fval=1.1029, time=0.0026s, n_fval=10.
Final fval=1.7575, time=0.0051s, n_fval=22.
Final fval=2.7386, time=0.0023s, n_fval=10.
 75%|███████▌  | 15/20 [00:02<00:00,  6.34it/s]Executing task 15.
Final fval=0.0134, time=0.0051s, n_fval=26.
Final fval=0.0651, time=0.0051s, n_fval=22.
Final fval=0.1459, time=0.0047s, n_fval=22.
Final fval=0.3111, time=0.0093s, n_fval=18.
Final fval=0.5169, time=0.0056s, n_fval=17.
Final fval=0.8692, time=0.0059s, n_fval=18.
Final fval=1.2729, time=0.0096s, n_fval=18.
Final fval=1.9125, time=0.0065s, n_fval=19.
Final fval=2.3210, time=0.0065s, n_fval=18.
Final fval=0.1016, time=0.0079s, n_fval=21.
Final fval=0.2553, time=0.0070s, n_fval=23.
Final fval=0.4858, time=0.0051s, n_fval=21.
Final fval=0.8322, time=0.0077s, n_fval=23.
Final fval=1.3513, time=0.0025s, n_fval=9.
Final fval=2.1303, time=0.0046s, n_fval=22.
 80%|████████  | 16/20 [00:02<00:00,  5.61it/s]Executing task 16.
Final fval=0.0041, time=0.0064s, n_fval=25.
Final fval=0.0457, time=0.0104s, n_fval=22.
Final fval=0.1096, time=0.0078s, n_fval=22.
Final fval=0.2522, time=0.0067s, n_fval=17.
Final fval=0.4291, time=0.0056s, n_fval=17.
Final fval=0.7320, time=0.0052s, n_fval=15.
Final fval=1.0960, time=0.0059s, n_fval=16.
Final fval=1.6605, time=0.0037s, n_fval=15.
Final fval=2.1834, time=0.0040s, n_fval=17.
Final fval=0.0918, time=0.0063s, n_fval=21.
Final fval=0.2416, time=0.0058s, n_fval=18.
Final fval=0.4671, time=0.0042s, n_fval=19.
Final fval=0.8054, time=0.0030s, n_fval=6.
Final fval=1.3123, time=0.0087s, n_fval=18.
Final fval=1.9996, time=0.0096s, n_fval=19.
 85%|████████▌ | 17/20 [00:02<00:00,  5.04it/s]Executing task 17.
Final fval=0.0011, time=0.0061s, n_fval=23.
Final fval=0.0381, time=0.0086s, n_fval=21.
Final fval=0.0960, time=0.0050s, n_fval=21.
Final fval=0.2393, time=0.0047s, n_fval=18.
Final fval=0.4307, time=0.0045s, n_fval=18.
Final fval=0.7496, time=0.0056s, n_fval=15.
Final fval=1.2156, time=0.0036s, n_fval=15.
Final fval=1.9255, time=0.0038s, n_fval=16.
Final fval=2.9569, time=0.0056s, n_fval=22.
Final fval=0.1025, time=0.0055s, n_fval=21.
Final fval=0.2587, time=0.0057s, n_fval=19.
Final fval=0.4932, time=0.0070s, n_fval=17.
Final fval=0.8445, time=0.0042s, n_fval=17.
Final fval=1.3711, time=0.0054s, n_fval=18.
Final fval=1.8504, time=0.0043s, n_fval=20.
Final fval=1.9674, time=0.0021s, n_fval=7.
 90%|█████████ | 18/20 [00:02<00:00,  4.74it/s]Executing task 18.
Final fval=0.0003, time=0.0047s, n_fval=20.
Final fval=0.0368, time=0.0044s, n_fval=17.
Final fval=0.0957, time=0.0042s, n_fval=17.
Final fval=0.2481, time=0.0049s, n_fval=15.
Final fval=0.4731, time=0.0046s, n_fval=16.
Final fval=0.8122, time=0.0035s, n_fval=15.
Final fval=1.3100, time=0.0040s, n_fval=18.
Final fval=2.0483, time=0.0045s, n_fval=20.
Final fval=0.0413, time=0.0045s, n_fval=18.
Final fval=0.1357, time=0.0051s, n_fval=17.
Final fval=0.2998, time=0.0039s, n_fval=16.
Final fval=0.5519, time=0.0045s, n_fval=17.
Final fval=0.9310, time=0.0048s, n_fval=16.
Final fval=1.5000, time=0.0038s, n_fval=14.
Final fval=2.0084, time=0.0032s, n_fval=12.
 95%|█████████▌| 19/20 [00:02<00:00,  4.96it/s]Executing task 19.
Final fval=0.0004, time=0.0050s, n_fval=20.
Final fval=0.0437, time=0.0056s, n_fval=20.
Final fval=0.1140, time=0.0050s, n_fval=19.
Final fval=0.2623, time=0.0046s, n_fval=17.
Final fval=0.4680, time=0.0046s, n_fval=19.
Final fval=0.7918, time=0.0050s, n_fval=22.
Final fval=1.2672, time=0.0051s, n_fval=22.
Final fval=1.9574, time=0.0057s, n_fval=22.
Final fval=0.0027, time=0.0051s, n_fval=20.
Final fval=0.0143, time=0.0040s, n_fval=17.
Final fval=0.0364, time=0.0049s, n_fval=19.
Final fval=0.0790, time=0.0049s, n_fval=18.
Final fval=0.1642, time=0.0040s, n_fval=18.
Final fval=0.2922, time=0.0041s, n_fval=18.
Final fval=0.4880, time=0.0043s, n_fval=17.
Final fval=0.7726, time=0.0037s, n_fval=17.
Final fval=1.1180, time=0.0042s, n_fval=17.
Final fval=1.5483, time=0.0040s, n_fval=17.
Final fval=1.8360, time=0.0038s, n_fval=15.
100%|██████████| 20/20 [00:03<00:00,  6.28it/s]
100%|██████████| 100000/100000 [00:34<00:00, 2910.44it/s]
Elapsed time: 34.077659

Plot results

We now want to plot the results (before saving).

import pypesto.visualize

# plot waterfalls
pypesto.visualize.waterfall(result, size=(15, 6))

<AxesSubplot:title={'center':'Waterfall plot'}, xlabel='Ordered optimizer run', ylabel='Offsetted function value (relative to best start)'>

# plot profiles
pypesto.visualize.profiles(result)

[<AxesSubplot:xlabel='x0', ylabel='Log-posterior ratio'>,
 <AxesSubplot:xlabel='x1'>,
 <AxesSubplot:xlabel='x2'>,
 <AxesSubplot:xlabel='x3'>,
 <AxesSubplot:xlabel='x4'>,
 <AxesSubplot:xlabel='x5', ylabel='Log-posterior ratio'>,
 <AxesSubplot:xlabel='x6'>,
 <AxesSubplot:xlabel='x7'>,
 <AxesSubplot:xlabel='x8'>,
 <AxesSubplot:xlabel='x9'>,
 <AxesSubplot:xlabel='x10', ylabel='Log-posterior ratio'>,
 <AxesSubplot:xlabel='x11'>,
 <AxesSubplot:xlabel='x12'>,
 <AxesSubplot:xlabel='x13'>,
 <AxesSubplot:xlabel='x14'>,
 <AxesSubplot:xlabel='x15', ylabel='Log-posterior ratio'>,
 <AxesSubplot:xlabel='x16'>,
 <AxesSubplot:xlabel='x17'>,
 <AxesSubplot:xlabel='x18'>,
 <AxesSubplot:xlabel='x19'>]

# plot samples
pypesto.visualize.sampling_fval_traces(result)

<AxesSubplot:xlabel='iteration index', ylabel='log-posterior'>

Save result object in HDF5 File

# create temporary file
fn = tempfile.mktemp(".hdf5")

# write result with write_result function.
# Choose which parts of the result object to save with
# corresponding booleans.
store.write_result(
    result=result,
    filename=fn,
    problem=True,
    optimize=True,
    profile=True,
    sample=True,
)

Reload results

# Read result
result2 = store.read_result(fn, problem=True)

WARNING: You are loading a problem.
This problem is not to be used without a separately created objective.

Plot (reloaded) results

# plot waterfalls
pypesto.visualize.waterfall(result2, size=(15, 6))

<AxesSubplot:title={'center':'Waterfall plot'}, xlabel='Ordered optimizer run', ylabel='Offsetted function value (relative to best start)'>

# plot profiles
pypesto.visualize.profiles(result2)

[<AxesSubplot:xlabel='x0', ylabel='Log-posterior ratio'>,
 <AxesSubplot:xlabel='x1'>,
 <AxesSubplot:xlabel='x2'>,
 <AxesSubplot:xlabel='x3'>,
 <AxesSubplot:xlabel='x4'>,
 <AxesSubplot:xlabel='x5', ylabel='Log-posterior ratio'>,
 <AxesSubplot:xlabel='x6'>,
 <AxesSubplot:xlabel='x7'>,
 <AxesSubplot:xlabel='x8'>,
 <AxesSubplot:xlabel='x9'>,
 <AxesSubplot:xlabel='x10', ylabel='Log-posterior ratio'>,
 <AxesSubplot:xlabel='x11'>,
 <AxesSubplot:xlabel='x12'>,
 <AxesSubplot:xlabel='x13'>,
 <AxesSubplot:xlabel='x14'>,
 <AxesSubplot:xlabel='x15', ylabel='Log-posterior ratio'>,
 <AxesSubplot:xlabel='x16'>,
 <AxesSubplot:xlabel='x17'>,
 <AxesSubplot:xlabel='x18'>,
 <AxesSubplot:xlabel='x19'>]

# plot samples
pypesto.visualize.sampling_fval_traces(result2)

<AxesSubplot:xlabel='iteration index', ylabel='log-posterior'>

For the saving of optimization history, we refer to store.ipynb.