Note
Go to the end to download the full example code.
Mixture of logit
Example of the use of different algorithms to estimate the model.
- author:
Michel Bierlaire, EPFL
- date:
Sun Apr 9 17:38:34 2023
import itertools
import pandas as pd
from biogeme.parameters import Parameters
from biogeme.tools import format_timedelta
import biogeme.biogeme_logging as blog
import biogeme.biogeme as bio
from biogeme import models
from biogeme.exceptions import BiogemeError
from biogeme.expressions import Beta, bioDraws, log, MonteCarlo
See the data processing script: Data preparation for Swissmetro.
from swissmetro_data import (
database,
CHOICE,
SM_AV,
CAR_AV_SP,
TRAIN_AV_SP,
TRAIN_TT_SCALED,
TRAIN_COST_SCALED,
SM_TT_SCALED,
SM_COST_SCALED,
CAR_TT_SCALED,
CAR_CO_SCALED,
)
logger = blog.get_screen_logger(level=blog.INFO)
logger.info('Example b05normal_mixture_all_algos.py')
Example b05normal_mixture_all_algos.py
Parameters to be estimated
ASC_CAR = Beta('ASC_CAR', 0, None, None, 0)
ASC_TRAIN = Beta('ASC_TRAIN', 0, None, None, 0)
ASC_SM = Beta('ASC_SM', 0, None, None, 1)
B_COST = Beta('B_COST', 0, None, None, 0)
Define a random parameter, normally distributed, designed to be used for Monte-Carlo simulation.
B_TIME = Beta('B_TIME', 0, None, None, 0)
It is advised not to use 0 as starting value for the following parameter.
B_TIME_S = Beta('B_TIME_S', 1, None, None, 0)
B_TIME_RND = B_TIME + B_TIME_S * bioDraws('b_time_rnd', 'NORMAL')
Definition of the utility functions.
V1 = ASC_TRAIN + B_TIME_RND * TRAIN_TT_SCALED + B_COST * TRAIN_COST_SCALED
V2 = ASC_SM + B_TIME_RND * SM_TT_SCALED + B_COST * SM_COST_SCALED
V3 = ASC_CAR + B_TIME_RND * CAR_TT_SCALED + B_COST * CAR_CO_SCALED
Associate utility functions with the numbering of alternatives
V = {1: V1, 2: V2, 3: V3}
Associate the availability conditions with the alternatives
av = {1: TRAIN_AV_SP, 2: SM_AV, 3: CAR_AV_SP}
Conditional to b_time_rnd, we have a logit model (called the kernel)
prob = models.logit(V, av, CHOICE)
We integrate over b_time_rnd using Monte-Carlo
logprob = log(MonteCarlo(prob))
Options for the optimization algorithm
The conjugate gradient iteration can be constrained to stay feasible, or not.
infeasible_cg_values = [True, False]
The radius of the first trust region is tested with three different values.
initial_radius_values = [0.1, 1.0, 10.0]
The percentage of iterations such that the analytical second derivatives is evaluated.
second_derivatives_values = [0.0, 0.5, 1.0]
We run the optimization algorithm with all possible combinations of the parameters.
The results are stored in a Pandas DataFrame called summary.
results = {}
summary = pd.DataFrame(
columns=[
'LogLikelihood',
'GradientNorm',
'Optimization time',
'TerminationCause',
'Status',
]
)
for infeasible_cg, initial_radius, second_derivatives in itertools.product(
infeasible_cg_values, initial_radius_values, second_derivatives_values
):
# Create the Biogeme object
the_biogeme = bio.BIOGEME(database, logprob, number_of_draws=100, seed=1223)
# We set the parameters of the optimization algorithm
the_biogeme.infeasible_cg = infeasible_cg
the_biogeme.initial_radius = initial_radius
the_biogeme.second_derivatives = second_derivatives
# We cancel the generation of the output files
the_biogeme.generate_html = False
the_biogeme.generate_pickle = False
name = (
f'cg_{infeasible_cg}_radius_{initial_radius}_second_deriv_{second_derivatives}'
)
the_biogeme.modelName = f'b05normal_mixture_algo_{name}'.strip()
result_data = {
'InfeasibleCG': infeasible_cg,
'InitialRadius': initial_radius,
'SecondDerivatives': second_derivatives,
'Status': 'Success', # Assume success unless an exception is caught
}
try:
results[name] = the_biogeme.estimate()
opt_time = format_timedelta(
results[name].data.optimizationMessages["Optimization time"]
)
result_data.update(
{
'LogLikelihood': results[name].data.logLike,
'GradientNorm': results[name].data.gradientNorm,
'Optimization time': opt_time,
'TerminationCause': results[name].data.optimizationMessages[
"Cause of termination"
],
}
)
except BiogemeError as e:
print(e)
result_data.update(
{
'Status': 'Failed',
'LogLikelihood': None,
'GradientNorm': None,
'Optimization time': None,
'TerminationCause': str(e),
}
)
results[name] = None
summary = pd.concat([summary, pd.DataFrame([result_data])], ignore_index=True)
Biogeme parameters read from biogeme.toml.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
*** Initial values of the parameters are obtained from the file __b05normal_mixture_algo_cg_True_radius_0.1_second_deriv_0.0.iter
Parameter values restored from __b05normal_mixture_algo_cg_True_radius_0.1_second_deriv_0.0.iter
The number of draws (100) is low. The results may not be meaningful.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
Optimization algorithm: hybrid Newton/BFGS with simple bounds [simple_bounds]
** Optimization: BFGS with trust region for simple bounds
Iter. ASC_CAR ASC_TRAIN B_COST B_TIME B_TIME_S Function Relgrad Radius Rho
0 -0.055 -0.8 -1.2 -1.4 0.9 5.3e+03 0.029 0.1 0.35 +
1 -0.15 -0.7 -1.1 -1.5 0.92 5.2e+03 0.014 0.1 0.57 +
2 -0.055 -0.71 -1.2 -1.6 0.97 5.2e+03 0.012 0.1 0.69 +
3 -0.13 -0.61 -1.2 -1.7 1 5.2e+03 0.015 0.1 0.56 +
4 -0.029 -0.67 -1.3 -1.7 1.1 5.2e+03 0.011 0.1 0.71 +
5 -0.064 -0.57 -1.2 -1.8 1.2 5.2e+03 0.011 0.1 0.77 +
6 0.036 -0.57 -1.2 -1.9 1.2 5.2e+03 0.0075 0.1 0.63 +
7 0.027 -0.47 -1.2 -1.9 1.3 5.2e+03 0.0091 0.1 0.65 +
8 0.066 -0.49 -1.2 -2 1.4 5.2e+03 0.0048 0.1 0.71 +
9 0.1 -0.44 -1.3 -2 1.5 5.2e+03 0.014 0.1 0.12 +
10 0.1 -0.44 -1.3 -2 1.5 5.2e+03 0.014 0.05 -1 -
11 0.068 -0.46 -1.3 -2.1 1.5 5.2e+03 0.0031 0.05 0.72 +
12 0.068 -0.46 -1.3 -2.1 1.5 5.2e+03 0.0031 0.025 -0.13 -
13 0.093 -0.45 -1.3 -2.1 1.5 5.2e+03 0.0059 0.025 0.75 +
14 0.085 -0.45 -1.3 -2.1 1.5 5.2e+03 0.0017 0.025 0.85 +
15 0.1 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0044 0.025 0.7 +
16 0.1 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0016 0.025 0.76 +
17 0.098 -0.42 -1.3 -2.2 1.5 5.2e+03 0.002 0.025 0.46 +
18 0.098 -0.42 -1.3 -2.2 1.5 5.2e+03 0.002 0.013 0.054 -
19 0.11 -0.43 -1.3 -2.2 1.5 5.2e+03 0.0012 0.013 0.75 +
20 0.11 -0.42 -1.3 -2.2 1.6 5.2e+03 0.0024 0.013 0.57 +
21 0.11 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00079 0.013 0.68 +
22 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.0012 0.013 0.28 +
23 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0016 0.013 0.68 +
24 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.0014 0.013 0.1 +
25 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00033 0.12 0.94 ++
26 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00033 0.062 -60 -
27 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00033 0.031 -16 -
28 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00033 0.016 -1.1 -
29 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00032 0.016 0.55 +
30 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00032 0.0078 -4.6 -
31 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00032 0.0039 -0.75 -
32 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00014 0.0039 0.45 +
33 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00014 0.002 -0.6 -
34 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00014 0.00098 -0.64 -
35 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0001 0.00098 0.47 -
/Users/bierlair/Library/CloudStorage/OneDrive-epfl.ch/github/biogeme/docs/source/examples/swissmetro/plot_b05normal_mixture_all_algos.py:166: FutureWarning: The behavior of DataFrame concatenation with empty or all-NA entries is deprecated. In a future version, this will no longer exclude empty or all-NA columns when determining the result dtypes. To retain the old behavior, exclude the relevant entries before the concat operation.
summary = pd.concat([summary, pd.DataFrame([result_data])], ignore_index=True)
Biogeme parameters read from biogeme.toml.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
*** Initial values of the parameters are obtained from the file __b05normal_mixture_algo_cg_True_radius_0.1_second_deriv_0.5.iter
Parameter values restored from __b05normal_mixture_algo_cg_True_radius_0.1_second_deriv_0.5.iter
The number of draws (100) is low. The results may not be meaningful.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
Optimization algorithm: hybrid Newton/BFGS with simple bounds [simple_bounds]
** Optimization: BFGS with trust region for simple bounds
Iter. ASC_CAR ASC_TRAIN B_COST B_TIME B_TIME_S Function Relgrad Radius Rho
0 -0.055 -0.8 -1.2 -1.4 1.5 5.3e+03 0.054 0.1 0.67 +
1 -0.15 -0.9 -1.1 -1.5 1.4 5.3e+03 0.022 0.1 0.52 +
2 -0.084 -0.8 -1.2 -1.6 1.3 5.2e+03 0.021 0.1 0.8 +
3 -0.18 -0.7 -1.2 -1.7 1.2 5.2e+03 0.02 0.1 0.27 +
4 -0.084 -0.6 -1.3 -1.8 1.3 5.2e+03 0.016 0.1 0.86 +
5 -0.084 -0.6 -1.3 -1.8 1.3 5.2e+03 0.016 0.05 0.018 -
6 -0.034 -0.65 -1.3 -1.8 1.3 5.2e+03 0.01 0.05 0.29 +
7 -0.012 -0.6 -1.2 -1.8 1.3 5.2e+03 0.011 0.05 0.85 +
8 -0.011 -0.55 -1.2 -1.9 1.3 5.2e+03 0.0055 0.05 0.63 +
9 0.039 -0.56 -1.2 -1.9 1.3 5.2e+03 0.0053 0.05 0.74 +
10 0.016 -0.51 -1.2 -1.9 1.4 5.2e+03 0.0049 0.05 0.86 +
11 0.066 -0.5 -1.2 -2 1.4 5.2e+03 0.0039 0.05 0.83 +
12 0.062 -0.46 -1.2 -2 1.4 5.2e+03 0.0031 0.05 0.85 +
13 0.095 -0.46 -1.3 -2.1 1.5 5.2e+03 0.0019 0.05 0.86 +
14 0.088 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0019 0.05 0.8 +
15 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00065 0.5 0.97 ++
16 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00065 0.1 -24 -
17 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00065 0.051 -6 -
18 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00065 0.026 0.026 -
19 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00069 0.026 0.65 +
20 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00069 0.013 -0.58 -
21 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00057 0.013 0.44 +
22 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00051 0.013 0.17 +
23 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00041 0.013 0.49 +
24 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00041 0.0064 -0.6 -
25 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00041 0.0032 -0.14 -
26 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00038 0.0032 0.35 +
27 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 7.3e-05 0.0032 1 +
Biogeme parameters read from biogeme.toml.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
*** Initial values of the parameters are obtained from the file __b05normal_mixture_algo_cg_True_radius_0.1_second_deriv_1.0.iter
Parameter values restored from __b05normal_mixture_algo_cg_True_radius_0.1_second_deriv_1.0.iter
The number of draws (100) is low. The results may not be meaningful.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
Optimization algorithm: hybrid Newton/BFGS with simple bounds [simple_bounds]
** Optimization: BFGS with trust region for simple bounds
Iter. ASC_CAR ASC_TRAIN B_COST B_TIME B_TIME_S Function Relgrad Radius Rho
0 -0.055 -0.8 -1.2 -1.4 1.5 5.3e+03 0.054 0.1 0.67 +
1 -0.15 -0.9 -1.1 -1.5 1.4 5.3e+03 0.022 0.1 0.52 +
2 -0.084 -0.8 -1.2 -1.6 1.3 5.2e+03 0.021 0.1 0.8 +
3 -0.18 -0.7 -1.2 -1.7 1.2 5.2e+03 0.02 0.1 0.27 +
4 -0.084 -0.6 -1.3 -1.8 1.3 5.2e+03 0.016 0.1 0.86 +
5 -0.084 -0.6 -1.3 -1.8 1.3 5.2e+03 0.016 0.05 0.024 -
6 -0.034 -0.65 -1.3 -1.8 1.3 5.2e+03 0.01 0.05 0.29 +
7 -0.012 -0.6 -1.2 -1.8 1.3 5.2e+03 0.011 0.05 0.85 +
8 -0.012 -0.55 -1.2 -1.9 1.3 5.2e+03 0.0055 0.05 0.63 +
9 0.038 -0.56 -1.2 -1.9 1.3 5.2e+03 0.0053 0.05 0.74 +
10 0.016 -0.51 -1.2 -1.9 1.4 5.2e+03 0.0049 0.05 0.86 +
11 0.066 -0.51 -1.2 -2 1.4 5.2e+03 0.0039 0.05 0.83 +
12 0.061 -0.46 -1.2 -2 1.4 5.2e+03 0.0031 0.05 0.85 +
13 0.095 -0.46 -1.3 -2.1 1.5 5.2e+03 0.0019 0.05 0.86 +
14 0.087 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0019 0.05 0.8 +
15 0.11 -0.43 -1.3 -2.2 1.6 5.2e+03 0.00066 0.5 0.97 ++
16 0.11 -0.43 -1.3 -2.2 1.6 5.2e+03 0.00066 0.091 -21 -
17 0.11 -0.43 -1.3 -2.2 1.6 5.2e+03 0.00066 0.046 -4.5 -
18 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0013 0.046 0.32 +
19 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0013 0.023 -0.49 -
20 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00089 0.023 0.35 +
21 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00089 0.011 -0.45 -
22 0.13 -0.4 -1.3 -2.2 1.6 5.2e+03 0.00045 0.011 0.63 +
23 0.13 -0.4 -1.3 -2.2 1.6 5.2e+03 0.00045 0.0057 -0.17 -
24 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00023 0.0057 0.5 +
25 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00023 0.0029 -2.6 -
26 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00018 0.0029 0.32 +
27 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00018 0.0014 -2.2 -
28 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00016 0.0014 0.43 +
29 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00011 0.0014 0.41 +
Biogeme parameters read from biogeme.toml.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
*** Initial values of the parameters are obtained from the file __b05normal_mixture_algo_cg_True_radius_1.0_second_deriv_0.0.iter
Parameter values restored from __b05normal_mixture_algo_cg_True_radius_1.0_second_deriv_0.0.iter
The number of draws (100) is low. The results may not be meaningful.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
Optimization algorithm: hybrid Newton/BFGS with simple bounds [simple_bounds]
** Optimization: BFGS with trust region for simple bounds
Iter. ASC_CAR ASC_TRAIN B_COST B_TIME B_TIME_S Function Relgrad Radius Rho
0 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 0.5 -1.8 -
1 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 0.25 -0.55 -
2 0.095 -0.95 -1.3 -1.5 1.4 5.3e+03 0.048 0.25 0.2 +
3 -0.15 -0.7 -1.1 -1.8 1.1 5.2e+03 0.028 0.25 0.37 +
4 -0.15 -0.7 -1.1 -1.8 1.1 5.2e+03 0.028 0.12 -0.047 -
5 -0.03 -0.58 -1.2 -1.7 1.2 5.2e+03 0.033 0.12 0.38 +
6 -0.041 -0.63 -1.2 -1.8 1.2 5.2e+03 0.0069 0.12 0.85 +
7 0.029 -0.51 -1.2 -1.8 1.3 5.2e+03 0.017 0.12 0.66 +
8 0.023 -0.52 -1.2 -2 1.3 5.2e+03 0.0076 0.12 0.46 +
9 0.091 -0.47 -1.2 -2 1.4 5.2e+03 0.0071 0.12 0.72 +
10 0.091 -0.47 -1.2 -2 1.4 5.2e+03 0.0071 0.062 -1.5 -
11 0.091 -0.47 -1.2 -2 1.4 5.2e+03 0.0071 0.031 -0.23 -
12 0.091 -0.47 -1.2 -2 1.4 5.2e+03 0.0071 0.016 -0.16 -
13 0.076 -0.45 -1.2 -2 1.4 5.2e+03 0.0056 0.016 0.46 +
14 0.086 -0.47 -1.2 -2 1.4 5.2e+03 0.0028 0.016 0.73 +
15 0.071 -0.45 -1.3 -2.1 1.4 5.2e+03 0.0023 0.016 0.63 +
16 0.086 -0.47 -1.3 -2.1 1.4 5.2e+03 0.0024 0.016 0.53 +
17 0.074 -0.45 -1.3 -2.1 1.5 5.2e+03 0.0023 0.016 0.77 +
18 0.09 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0057 0.016 0.22 +
19 0.089 -0.45 -1.3 -2.1 1.5 5.2e+03 0.0017 0.016 0.82 +
20 0.1 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0024 0.016 0.83 +
21 0.093 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0017 0.016 0.64 +
22 0.11 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0015 0.016 0.89 +
23 0.1 -0.43 -1.3 -2.2 1.5 5.2e+03 0.0011 0.016 0.85 +
24 0.11 -0.43 -1.3 -2.2 1.6 5.2e+03 0.00071 0.016 0.84 +
25 0.11 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00055 0.16 0.93 ++
26 0.11 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00055 0.023 -3.8 -
27 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00043 0.23 0.91 ++
28 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00043 0.093 -53 -
29 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00043 0.047 -15 -
30 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00043 0.023 -3.8 -
31 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00043 0.012 -0.31 -
32 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0009 0.012 0.25 +
33 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0009 0.0058 -0.16 -
34 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00013 0.0058 0.89 +
35 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00013 0.0029 -1.3 -
36 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00013 0.0015 -0.098 -
37 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00013 0.0015 0.46 +
38 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 5.3e-05 0.0015 0.9 +
Biogeme parameters read from biogeme.toml.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
*** Initial values of the parameters are obtained from the file __b05normal_mixture_algo_cg_True_radius_1.0_second_deriv_0.5.iter
Parameter values restored from __b05normal_mixture_algo_cg_True_radius_1.0_second_deriv_0.5.iter
The number of draws (100) is low. The results may not be meaningful.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
Optimization algorithm: hybrid Newton/BFGS with simple bounds [simple_bounds]
** Optimization: BFGS with trust region for simple bounds
Iter. ASC_CAR ASC_TRAIN B_COST B_TIME B_TIME_S Function Relgrad Radius Rho
0 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.065 0.5 -1.9 -
1 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.065 0.25 -0.56 -
2 0.095 -0.95 -1.3 -1.5 1.4 5.3e+03 0.048 0.25 0.19 +
3 -0.15 -0.7 -1.1 -1.8 1.1 5.2e+03 0.029 0.25 0.37 +
4 -0.15 -0.7 -1.1 -1.8 1.1 5.2e+03 0.029 0.12 -0.048 -
5 -0.03 -0.58 -1.2 -1.7 1.2 5.2e+03 0.033 0.12 0.39 +
6 -0.04 -0.63 -1.2 -1.8 1.2 5.2e+03 0.0069 0.12 0.85 +
7 0.026 -0.51 -1.2 -1.8 1.2 5.2e+03 0.016 0.12 0.67 +
8 0.026 -0.52 -1.2 -2 1.3 5.2e+03 0.0075 0.12 0.47 +
9 0.088 -0.46 -1.2 -2 1.4 5.2e+03 0.0074 0.12 0.73 +
10 0.088 -0.46 -1.2 -2 1.4 5.2e+03 0.0074 0.062 -1 -
11 0.046 -0.45 -1.2 -2.1 1.4 5.2e+03 0.007 0.062 0.1 +
12 0.11 -0.48 -1.3 -2.1 1.5 5.2e+03 0.0068 0.062 0.11 +
13 0.1 -0.42 -1.3 -2.1 1.5 5.2e+03 0.01 0.062 0.28 +
14 0.1 -0.42 -1.3 -2.1 1.5 5.2e+03 0.01 0.031 -0.24 -
15 0.096 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0015 0.031 0.87 +
16 0.11 -0.42 -1.3 -2.1 1.5 5.2e+03 0.0026 0.031 0.8 +
17 0.12 -0.43 -1.3 -2.2 1.5 5.2e+03 0.0014 0.031 0.53 +
18 0.11 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0019 0.031 0.26 +
19 0.11 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0019 0.016 -1.4 -
20 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.0008 0.016 0.57 +
21 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00082 0.016 0.21 +
22 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00082 0.0078 -0.23 -
23 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0012 0.0078 0.5 +
24 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00043 0.0078 0.35 +
25 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00023 0.0078 0.81 +
26 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00023 0.0039 -4.5 -
27 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00023 0.002 -0.82 -
28 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00017 0.002 0.39 +
29 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00018 0.002 0.64 +
30 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00015 0.002 0.19 +
31 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 4.3e-05 0.002 0.88 +
Biogeme parameters read from biogeme.toml.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
*** Initial values of the parameters are obtained from the file __b05normal_mixture_algo_cg_True_radius_1.0_second_deriv_1.0.iter
Parameter values restored from __b05normal_mixture_algo_cg_True_radius_1.0_second_deriv_1.0.iter
The number of draws (100) is low. The results may not be meaningful.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
Optimization algorithm: hybrid Newton/BFGS with simple bounds [simple_bounds]
** Optimization: BFGS with trust region for simple bounds
Iter. ASC_CAR ASC_TRAIN B_COST B_TIME B_TIME_S Function Relgrad Radius Rho
0 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.065 0.5 -1.8 -
1 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.065 0.25 -0.56 -
2 0.095 -0.95 -1.3 -1.5 1.4 5.3e+03 0.048 0.25 0.19 +
3 -0.15 -0.7 -1.1 -1.8 1.1 5.2e+03 0.028 0.25 0.37 +
4 -0.15 -0.7 -1.1 -1.8 1.1 5.2e+03 0.028 0.12 -0.048 -
5 -0.03 -0.58 -1.2 -1.7 1.2 5.2e+03 0.033 0.12 0.39 +
6 -0.04 -0.63 -1.2 -1.8 1.2 5.2e+03 0.0069 0.12 0.85 +
7 0.027 -0.51 -1.2 -1.8 1.3 5.2e+03 0.016 0.12 0.67 +
8 0.025 -0.52 -1.2 -2 1.3 5.2e+03 0.0076 0.12 0.46 +
9 0.089 -0.46 -1.2 -2 1.4 5.2e+03 0.0073 0.12 0.73 +
10 0.089 -0.46 -1.2 -2 1.4 5.2e+03 0.0073 0.062 -1.1 -
11 0.089 -0.46 -1.2 -2 1.4 5.2e+03 0.0073 0.031 0.025 -
12 0.089 -0.46 -1.2 -2 1.4 5.2e+03 0.0073 0.016 0.056 -
13 0.073 -0.47 -1.2 -2 1.4 5.2e+03 0.0032 0.016 0.65 +
14 0.077 -0.45 -1.2 -2 1.4 5.2e+03 0.0047 0.016 0.83 +
15 0.093 -0.47 -1.3 -2.1 1.4 5.2e+03 0.0034 0.016 0.35 +
16 0.077 -0.45 -1.2 -2.1 1.5 5.2e+03 0.0021 0.016 0.78 +
17 0.093 -0.46 -1.3 -2.1 1.5 5.2e+03 0.0023 0.016 0.89 +
18 0.082 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0022 0.016 0.67 +
19 0.097 -0.46 -1.3 -2.1 1.5 5.2e+03 0.0022 0.016 0.59 +
20 0.087 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0024 0.016 0.79 +
21 0.1 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0014 0.016 0.85 +
22 0.099 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0013 0.016 0.76 +
23 0.11 -0.43 -1.3 -2.2 1.5 5.2e+03 0.0011 0.016 0.78 +
24 0.11 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00069 0.016 0.84 +
25 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.0016 0.016 0.7 +
26 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00083 0.016 0.55 +
27 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00083 0.0078 0.068 -
28 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00059 0.0078 0.65 +
29 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00079 0.0078 0.58 +
30 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00029 0.0078 0.27 +
31 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00029 0.0039 -1 -
32 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00025 0.0039 0.67 +
33 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00025 0.002 -1.6 -
34 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00031 0.002 0.54 +
35 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00012 0.002 0.74 +
Biogeme parameters read from biogeme.toml.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
*** Initial values of the parameters are obtained from the file __b05normal_mixture_algo_cg_True_radius_10.0_second_deriv_0.0.iter
Parameter values restored from __b05normal_mixture_algo_cg_True_radius_10.0_second_deriv_0.0.iter
The number of draws (100) is low. The results may not be meaningful.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
Optimization algorithm: hybrid Newton/BFGS with simple bounds [simple_bounds]
** Optimization: BFGS with trust region for simple bounds
Iter. ASC_CAR ASC_TRAIN B_COST B_TIME B_TIME_S Function Relgrad Radius Rho
0 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.065 5 -3 -
1 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.065 2.5 -3.3 -
2 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.065 1.2 -3.2 -
3 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.065 0.62 -2.3 -
4 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.065 0.31 -0.91 -
5 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.065 0.16 -0.0013 -
6 0.0016 -0.86 -1.2 -1.4 1.4 5.3e+03 0.047 0.16 0.49 +
7 -0.15 -0.75 -1.1 -1.6 1.3 5.2e+03 0.014 0.16 0.51 +
8 0.0016 -0.59 -1.2 -1.7 1.1 5.2e+03 0.012 0.16 0.53 +
9 0.0016 -0.59 -1.2 -1.7 1.1 5.2e+03 0.012 0.078 -0.11 -
10 -0.064 -0.52 -1.2 -1.8 1.2 5.2e+03 0.014 0.078 0.11 +
11 0.014 -0.59 -1.2 -1.9 1.2 5.2e+03 0.0083 0.078 0.66 +
12 1.1e-05 -0.52 -1.2 -1.9 1.3 5.2e+03 0.011 0.078 0.66 +
13 0.035 -0.53 -1.2 -1.9 1.3 5.2e+03 0.0047 0.078 0.76 +
14 0.072 -0.47 -1.2 -2 1.4 5.2e+03 0.012 0.078 0.6 +
15 0.049 -0.47 -1.2 -2 1.4 5.2e+03 0.0047 0.078 0.55 +
16 0.049 -0.47 -1.2 -2 1.4 5.2e+03 0.0047 0.039 -0.23 -
17 0.088 -0.47 -1.2 -2 1.4 5.2e+03 0.0046 0.039 0.66 +
18 0.074 -0.45 -1.3 -2.1 1.5 5.2e+03 0.0025 0.039 0.78 +
19 0.1 -0.46 -1.3 -2.1 1.5 5.2e+03 0.0035 0.039 0.24 +
20 0.11 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0056 0.039 0.57 +
21 0.11 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0056 0.02 -0.35 -
22 0.11 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0056 0.0098 0.039 -
23 0.099 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0014 0.0098 0.72 +
24 0.098 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0013 0.0098 0.82 +
25 0.11 -0.43 -1.3 -2.2 1.5 5.2e+03 0.0019 0.0098 0.73 +
26 0.11 -0.44 -1.3 -2.2 1.5 5.2e+03 0.0011 0.0098 0.45 +
27 0.1 -0.43 -1.3 -2.2 1.5 5.2e+03 0.0014 0.0098 0.34 +
28 0.11 -0.43 -1.3 -2.2 1.6 5.2e+03 0.00091 0.098 0.98 ++
29 0.11 -0.43 -1.3 -2.2 1.6 5.2e+03 0.00091 0.049 -0.14 -
30 0.12 -0.4 -1.3 -2.2 1.6 5.2e+03 0.0012 0.049 0.47 +
31 0.12 -0.4 -1.3 -2.2 1.6 5.2e+03 0.0012 0.024 -0.88 -
32 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.0011 0.024 0.34 +
33 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.0011 0.012 -2.9 -
34 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.0011 0.0061 -0.078 -
35 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00053 0.0061 0.37 +
36 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0003 0.0061 0.79 +
37 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0003 0.0031 -0.083 -
38 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00013 0.0031 0.6 +
39 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00012 0.0031 0.68 +
Biogeme parameters read from biogeme.toml.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
*** Initial values of the parameters are obtained from the file __b05normal_mixture_algo_cg_True_radius_10.0_second_deriv_0.5.iter
Parameter values restored from __b05normal_mixture_algo_cg_True_radius_10.0_second_deriv_0.5.iter
The number of draws (100) is low. The results may not be meaningful.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
Optimization algorithm: hybrid Newton/BFGS with simple bounds [simple_bounds]
** Optimization: BFGS with trust region for simple bounds
Iter. ASC_CAR ASC_TRAIN B_COST B_TIME B_TIME_S Function Relgrad Radius Rho
0 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.065 5 -3 -
1 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.065 2.5 -3.3 -
2 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.065 1.2 -3.2 -
3 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.065 0.62 -2.3 -
4 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.065 0.31 -0.91 -
5 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.065 0.16 0.00086 -
6 0.0016 -0.86 -1.2 -1.4 1.4 5.3e+03 0.047 0.16 0.49 +
7 -0.15 -0.75 -1.1 -1.6 1.3 5.2e+03 0.014 0.16 0.51 +
8 0.0016 -0.59 -1.2 -1.7 1.1 5.2e+03 0.012 0.16 0.54 +
9 0.0016 -0.59 -1.2 -1.7 1.1 5.2e+03 0.012 0.078 -0.12 -
10 -0.064 -0.52 -1.2 -1.8 1.2 5.2e+03 0.014 0.078 0.11 +
11 0.014 -0.59 -1.2 -1.9 1.2 5.2e+03 0.0083 0.078 0.66 +
12 -0.00026 -0.52 -1.2 -1.9 1.3 5.2e+03 0.011 0.078 0.66 +
13 0.035 -0.53 -1.2 -1.9 1.3 5.2e+03 0.0047 0.078 0.76 +
14 0.071 -0.47 -1.2 -2 1.4 5.2e+03 0.011 0.078 0.61 +
15 0.05 -0.47 -1.2 -2 1.4 5.2e+03 0.0046 0.078 0.55 +
16 0.05 -0.47 -1.2 -2 1.4 5.2e+03 0.0046 0.039 -0.23 -
17 0.089 -0.47 -1.2 -2 1.4 5.2e+03 0.0046 0.039 0.65 +
18 0.074 -0.45 -1.3 -2.1 1.5 5.2e+03 0.0028 0.039 0.75 +
19 0.1 -0.46 -1.3 -2.1 1.5 5.2e+03 0.0036 0.039 0.22 +
20 0.11 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0058 0.039 0.55 +
21 0.11 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0058 0.02 -0.41 -
22 0.09 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0019 0.02 0.32 +
23 0.11 -0.43 -1.3 -2.2 1.5 5.2e+03 0.0011 0.02 0.89 +
24 0.11 -0.43 -1.3 -2.2 1.5 5.2e+03 0.0011 0.0098 0.00012 -
25 0.1 -0.42 -1.3 -2.2 1.5 5.2e+03 0.0013 0.0098 0.36 +
26 0.11 -0.43 -1.3 -2.2 1.5 5.2e+03 0.00075 0.098 0.93 ++
27 0.14 -0.39 -1.3 -2.3 1.6 5.2e+03 0.0011 0.098 0.2 +
28 0.14 -0.39 -1.3 -2.3 1.6 5.2e+03 0.0011 0.049 -7.8 -
29 0.14 -0.39 -1.3 -2.3 1.6 5.2e+03 0.0011 0.024 -0.8 -
30 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0018 0.024 0.34 +
31 0.13 -0.4 -1.3 -2.2 1.6 5.2e+03 0.0004 0.024 0.72 +
32 0.13 -0.4 -1.3 -2.2 1.6 5.2e+03 0.0004 0.012 -7 -
33 0.13 -0.4 -1.3 -2.2 1.6 5.2e+03 0.0004 0.0061 -0.14 -
34 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00017 0.0061 0.81 +
35 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00017 0.0031 -2.3 -
36 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00017 0.0015 -0.79 -
37 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00015 0.0015 0.29 +
38 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00012 0.0015 0.7 +
Biogeme parameters read from biogeme.toml.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
*** Initial values of the parameters are obtained from the file __b05normal_mixture_algo_cg_True_radius_10.0_second_deriv_1.0.iter
Parameter values restored from __b05normal_mixture_algo_cg_True_radius_10.0_second_deriv_1.0.iter
The number of draws (100) is low. The results may not be meaningful.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
Optimization algorithm: hybrid Newton/BFGS with simple bounds [simple_bounds]
** Optimization: BFGS with trust region for simple bounds
Iter. ASC_CAR ASC_TRAIN B_COST B_TIME B_TIME_S Function Relgrad Radius Rho
0 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 5 -3 -
1 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 2.5 -3.3 -
2 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 1.2 -3.2 -
3 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 0.62 -2.3 -
4 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 0.31 -0.9 -
5 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 0.16 0.0059 -
6 0.0016 -0.86 -1.2 -1.4 1.5 5.3e+03 0.048 0.16 0.49 +
7 -0.15 -0.75 -1.1 -1.6 1.3 5.2e+03 0.015 0.16 0.52 +
8 0.0016 -0.6 -1.2 -1.7 1.1 5.2e+03 0.012 0.16 0.54 +
9 0.0016 -0.6 -1.2 -1.7 1.1 5.2e+03 0.012 0.078 -0.15 -
10 -0.065 -0.52 -1.2 -1.8 1.2 5.2e+03 0.014 0.078 0.1 +
11 0.013 -0.6 -1.2 -1.9 1.2 5.2e+03 0.0084 0.078 0.64 +
12 -0.001 -0.52 -1.2 -1.9 1.3 5.2e+03 0.011 0.078 0.67 +
13 0.035 -0.53 -1.2 -1.9 1.3 5.2e+03 0.0047 0.078 0.76 +
14 0.07 -0.46 -1.2 -2 1.4 5.2e+03 0.011 0.078 0.61 +
15 0.052 -0.47 -1.2 -2 1.4 5.2e+03 0.0043 0.078 0.57 +
16 0.052 -0.47 -1.2 -2 1.4 5.2e+03 0.0043 0.039 -0.24 -
17 0.091 -0.47 -1.3 -2 1.4 5.2e+03 0.0046 0.039 0.61 +
18 0.073 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0034 0.039 0.66 +
19 0.097 -0.46 -1.3 -2.1 1.5 5.2e+03 0.0036 0.039 0.22 +
20 0.11 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0061 0.039 0.45 +
21 0.11 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0061 0.02 -0.45 -
22 0.093 -0.44 -1.2 -2.1 1.5 5.2e+03 0.0017 0.02 0.47 +
23 0.1 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0019 0.02 0.87 +
24 0.11 -0.43 -1.3 -2.2 1.5 5.2e+03 0.0015 0.02 0.41 +
25 0.11 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00085 0.02 0.76 +
26 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.0022 0.02 0.58 +
27 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.0014 0.02 0.34 +
28 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0007 0.02 0.78 +
29 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0007 0.0098 -2.9 -
30 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0007 0.0049 -0.6 -
31 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0003 0.0049 0.4 +
32 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00011 0.0049 0.75 +
Biogeme parameters read from biogeme.toml.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
*** Initial values of the parameters are obtained from the file __b05normal_mixture_algo_cg_False_radius_0.1_second_deriv_0.0.iter
Parameter values restored from __b05normal_mixture_algo_cg_False_radius_0.1_second_deriv_0.0.iter
The number of draws (100) is low. The results may not be meaningful.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
Optimization algorithm: hybrid Newton/BFGS with simple bounds [simple_bounds]
** Optimization: BFGS with trust region for simple bounds
Iter. ASC_CAR ASC_TRAIN B_COST B_TIME B_TIME_S Function Relgrad Radius Rho
0 -0.055 -0.8 -1.2 -1.4 1.5 5.3e+03 0.054 0.1 0.67 +
1 -0.15 -0.9 -1.1 -1.5 1.4 5.3e+03 0.022 0.1 0.52 +
2 -0.084 -0.8 -1.2 -1.6 1.3 5.2e+03 0.021 0.1 0.8 +
3 -0.18 -0.7 -1.2 -1.7 1.2 5.2e+03 0.02 0.1 0.27 +
4 -0.084 -0.6 -1.3 -1.8 1.3 5.2e+03 0.016 0.1 0.86 +
5 -0.084 -0.6 -1.3 -1.8 1.3 5.2e+03 0.016 0.05 0.026 -
6 -0.034 -0.65 -1.3 -1.8 1.3 5.2e+03 0.01 0.05 0.29 +
7 -0.012 -0.6 -1.2 -1.8 1.3 5.2e+03 0.011 0.05 0.85 +
8 -0.012 -0.55 -1.2 -1.9 1.3 5.2e+03 0.0055 0.05 0.63 +
9 0.038 -0.56 -1.2 -1.9 1.3 5.2e+03 0.0053 0.05 0.74 +
10 0.016 -0.51 -1.2 -1.9 1.4 5.2e+03 0.0049 0.05 0.86 +
11 0.066 -0.51 -1.2 -2 1.4 5.2e+03 0.004 0.05 0.83 +
12 0.061 -0.46 -1.2 -2 1.4 5.2e+03 0.0031 0.05 0.85 +
13 0.095 -0.46 -1.3 -2.1 1.5 5.2e+03 0.0019 0.05 0.86 +
14 0.087 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0019 0.05 0.8 +
15 0.11 -0.43 -1.3 -2.2 1.6 5.2e+03 0.00066 0.5 0.97 ++
16 0.11 -0.43 -1.3 -2.2 1.6 5.2e+03 0.00066 0.087 -20 -
17 0.11 -0.43 -1.3 -2.2 1.6 5.2e+03 0.00066 0.044 -4.1 -
18 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0012 0.044 0.41 +
19 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0012 0.022 -0.45 -
20 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00089 0.022 0.31 +
21 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00089 0.011 -0.35 -
22 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00038 0.011 0.8 +
23 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00038 0.0055 -1.1 -
24 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00011 0.0055 0.27 -
Biogeme parameters read from biogeme.toml.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
*** Initial values of the parameters are obtained from the file __b05normal_mixture_algo_cg_False_radius_0.1_second_deriv_0.5.iter
Parameter values restored from __b05normal_mixture_algo_cg_False_radius_0.1_second_deriv_0.5.iter
The number of draws (100) is low. The results may not be meaningful.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
Optimization algorithm: hybrid Newton/BFGS with simple bounds [simple_bounds]
** Optimization: BFGS with trust region for simple bounds
Iter. ASC_CAR ASC_TRAIN B_COST B_TIME B_TIME_S Function Relgrad Radius Rho
0 -0.055 -0.8 -1.2 -1.4 1.5 5.3e+03 0.055 0.1 0.68 +
1 -0.15 -0.9 -1.1 -1.5 1.4 5.3e+03 0.022 0.1 0.52 +
2 -0.084 -0.8 -1.2 -1.6 1.3 5.2e+03 0.021 0.1 0.8 +
3 -0.18 -0.7 -1.2 -1.7 1.2 5.2e+03 0.02 0.1 0.28 +
4 -0.084 -0.6 -1.3 -1.8 1.2 5.2e+03 0.0081 1 0.96 ++
5 -0.084 -0.6 -1.3 -1.8 1.2 5.2e+03 0.0081 0.5 -5.1 -
6 -0.084 -0.6 -1.3 -1.8 1.2 5.2e+03 0.0081 0.25 -2.1 -
7 -0.084 -0.6 -1.3 -1.8 1.2 5.2e+03 0.0081 0.12 -0.53 -
8 0.041 -0.59 -1.2 -1.9 1.3 5.2e+03 0.0091 0.12 0.47 +
9 0.041 -0.59 -1.2 -1.9 1.3 5.2e+03 0.0091 0.062 -0.094 -
10 -0.012 -0.53 -1.2 -1.9 1.4 5.2e+03 0.0091 0.062 0.53 +
11 0.041 -0.53 -1.2 -2 1.4 5.2e+03 0.0039 0.62 0.91 ++
12 0.041 -0.53 -1.2 -2 1.4 5.2e+03 0.0039 0.31 -1.9 -
13 0.041 -0.53 -1.2 -2 1.4 5.2e+03 0.0039 0.16 -0.41 -
14 0.11 -0.37 -1.2 -2.1 1.5 5.2e+03 0.0063 0.16 0.28 +
15 0.089 -0.47 -1.3 -2.2 1.6 5.2e+03 0.0029 0.16 0.45 +
16 0.089 -0.47 -1.3 -2.2 1.6 5.2e+03 0.0029 0.078 -2.4 -
17 0.089 -0.47 -1.3 -2.2 1.6 5.2e+03 0.0029 0.039 -0.18 -
18 0.12 -0.43 -1.3 -2.2 1.6 5.2e+03 0.0079 0.039 0.15 +
19 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00076 0.039 0.77 +
20 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00076 0.02 -3.7 -
21 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00076 0.0098 -0.88 -
22 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00045 0.0098 0.41 +
23 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00051 0.0098 0.46 +
24 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00051 0.0049 -0.35 -
25 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00049 0.0049 0.23 +
26 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00033 0.0049 0.44 +
27 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00012 0.0049 0.63 +
28 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00012 0.0024 -0.71 -
29 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 5.2e-05 0.0024 0.25 -
Biogeme parameters read from biogeme.toml.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
*** Initial values of the parameters are obtained from the file __b05normal_mixture_algo_cg_False_radius_0.1_second_deriv_1.0.iter
Parameter values restored from __b05normal_mixture_algo_cg_False_radius_0.1_second_deriv_1.0.iter
The number of draws (100) is low. The results may not be meaningful.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
Optimization algorithm: hybrid Newton/BFGS with simple bounds [simple_bounds]
** Optimization: BFGS with trust region for simple bounds
Iter. ASC_CAR ASC_TRAIN B_COST B_TIME B_TIME_S Function Relgrad Radius Rho
0 -0.055 -0.8 -1.2 -1.4 1.5 5.3e+03 0.055 0.1 0.67 +
1 -0.15 -0.9 -1.1 -1.5 1.4 5.3e+03 0.022 0.1 0.52 +
2 -0.084 -0.8 -1.2 -1.6 1.3 5.2e+03 0.021 0.1 0.8 +
3 -0.18 -0.7 -1.2 -1.7 1.2 5.2e+03 0.02 0.1 0.28 +
4 -0.084 -0.6 -1.3 -1.8 1.3 5.2e+03 0.016 0.1 0.85 +
5 -0.084 -0.6 -1.3 -1.8 1.3 5.2e+03 0.016 0.05 -0.013 -
6 -0.034 -0.65 -1.3 -1.8 1.3 5.2e+03 0.01 0.05 0.3 +
7 -0.011 -0.6 -1.2 -1.8 1.3 5.2e+03 0.011 0.05 0.85 +
8 -0.01 -0.55 -1.2 -1.9 1.3 5.2e+03 0.0055 0.05 0.63 +
9 0.04 -0.56 -1.2 -1.9 1.3 5.2e+03 0.0052 0.05 0.74 +
10 0.018 -0.51 -1.2 -2 1.4 5.2e+03 0.0048 0.05 0.86 +
11 0.068 -0.5 -1.2 -2 1.4 5.2e+03 0.0038 0.05 0.84 +
12 0.063 -0.46 -1.2 -2 1.4 5.2e+03 0.0029 0.05 0.85 +
13 0.095 -0.46 -1.3 -2.1 1.5 5.2e+03 0.0018 0.05 0.85 +
14 0.089 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0019 0.05 0.81 +
15 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00054 0.5 0.95 ++
16 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00054 0.17 -29 -
17 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00054 0.084 -8.8 -
18 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00054 0.042 -2.3 -
19 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00054 0.021 -0.3 -
20 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00069 0.021 0.46 +
21 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0017 0.021 0.47 +
22 0.13 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00084 0.021 0.3 +
23 0.13 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00084 0.011 -0.97 -
24 0.13 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00084 0.0053 -0.033 -
25 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00034 0.0053 0.62 +
26 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00011 0.0053 0.5 +
Biogeme parameters read from biogeme.toml.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
*** Initial values of the parameters are obtained from the file __b05normal_mixture_algo_cg_False_radius_1.0_second_deriv_0.0.iter
Parameter values restored from __b05normal_mixture_algo_cg_False_radius_1.0_second_deriv_0.0.iter
The number of draws (100) is low. The results may not be meaningful.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
Optimization algorithm: hybrid Newton/BFGS with simple bounds [simple_bounds]
** Optimization: BFGS with trust region for simple bounds
Iter. ASC_CAR ASC_TRAIN B_COST B_TIME B_TIME_S Function Relgrad Radius Rho
0 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.065 0.5 -1.8 -
1 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.065 0.25 -0.56 -
2 0.095 -0.95 -1.3 -1.5 1.4 5.3e+03 0.048 0.25 0.19 +
3 -0.15 -0.7 -1.1 -1.8 1.1 5.2e+03 0.028 0.25 0.37 +
4 -0.15 -0.7 -1.1 -1.8 1.1 5.2e+03 0.028 0.12 -0.048 -
5 -0.03 -0.58 -1.2 -1.7 1.2 5.2e+03 0.033 0.12 0.39 +
6 -0.04 -0.63 -1.2 -1.8 1.2 5.2e+03 0.0069 0.12 0.85 +
7 0.027 -0.51 -1.2 -1.8 1.3 5.2e+03 0.017 0.12 0.67 +
8 0.025 -0.52 -1.2 -2 1.3 5.2e+03 0.0076 0.12 0.46 +
9 0.089 -0.46 -1.2 -2 1.4 5.2e+03 0.0073 0.12 0.73 +
10 0.089 -0.46 -1.2 -2 1.4 5.2e+03 0.0073 0.062 -1.2 -
11 0.089 -0.46 -1.2 -2 1.4 5.2e+03 0.0073 0.031 -0.013 -
12 0.089 -0.46 -1.2 -2 1.4 5.2e+03 0.0073 0.016 0.073 -
13 0.074 -0.47 -1.2 -2 1.4 5.2e+03 0.0032 0.016 0.66 +
14 0.078 -0.45 -1.2 -2 1.4 5.2e+03 0.0047 0.016 0.84 +
15 0.094 -0.47 -1.3 -2.1 1.4 5.2e+03 0.0033 0.016 0.31 +
16 0.078 -0.45 -1.2 -2.1 1.5 5.2e+03 0.0021 0.016 0.79 +
17 0.094 -0.46 -1.3 -2.1 1.5 5.2e+03 0.0024 0.016 0.89 +
18 0.088 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0021 0.016 0.84 +
19 0.099 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0018 0.016 0.89 +
20 0.098 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0015 0.016 0.88 +
21 0.11 -0.43 -1.3 -2.2 1.5 5.2e+03 0.0011 0.016 0.89 +
22 0.11 -0.42 -1.3 -2.2 1.5 5.2e+03 0.0013 0.16 0.97 ++
23 0.11 -0.42 -1.3 -2.2 1.5 5.2e+03 0.0013 0.078 -3 -
24 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00091 0.078 0.73 +
25 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00091 0.039 -8.7 -
26 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00091 0.02 -4.2 -
27 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00091 0.0098 -1.8 -
28 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00091 0.0049 -0.29 -
29 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00091 0.0024 -0.045 -
30 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00046 0.0024 0.62 +
31 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00027 0.0024 0.75 +
32 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00021 0.0024 0.61 +
33 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00015 0.0024 0.59 +
34 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00014 0.0024 0.18 +
35 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0001 0.0024 0.48 +
Biogeme parameters read from biogeme.toml.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
*** Initial values of the parameters are obtained from the file __b05normal_mixture_algo_cg_False_radius_1.0_second_deriv_0.5.iter
Parameter values restored from __b05normal_mixture_algo_cg_False_radius_1.0_second_deriv_0.5.iter
The number of draws (100) is low. The results may not be meaningful.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
Optimization algorithm: hybrid Newton/BFGS with simple bounds [simple_bounds]
** Optimization: BFGS with trust region for simple bounds
Iter. ASC_CAR ASC_TRAIN B_COST B_TIME B_TIME_S Function Relgrad Radius Rho
0 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 0.5 -1.8 -
1 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 0.25 -0.55 -
2 0.095 -0.95 -1.3 -1.5 1.4 5.3e+03 0.048 0.25 0.2 +
3 -0.15 -0.7 -1.1 -1.8 1.1 5.2e+03 0.028 0.25 0.37 +
4 -0.15 -0.7 -1.1 -1.8 1.1 5.2e+03 0.028 0.12 -0.045 -
5 -0.03 -0.58 -1.2 -1.7 1.2 5.2e+03 0.033 0.12 0.38 +
6 -0.03 -0.7 -1.2 -1.8 1.2 5.2e+03 0.017 0.12 0.29 +
7 -0.038 -0.58 -1.2 -1.8 1.2 5.2e+03 0.011 0.12 0.87 +
8 0.064 -0.57 -1.2 -1.9 1.2 5.2e+03 0.0096 0.12 0.38 +
9 0.029 -0.45 -1.2 -2 1.3 5.2e+03 0.007 0.12 0.57 +
10 0.029 -0.45 -1.2 -2 1.3 5.2e+03 0.007 0.062 -0.025 -
11 0.092 -0.47 -1.2 -2 1.4 5.2e+03 0.0071 0.062 0.6 +
12 0.06 -0.48 -1.2 -2.1 1.4 5.2e+03 0.0062 0.062 0.3 +
13 0.067 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0091 0.062 0.21 +
14 0.067 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0091 0.031 -0.039 -
15 0.098 -0.46 -1.3 -2.1 1.4 5.2e+03 0.0035 0.031 0.53 +
16 0.084 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0017 0.031 0.88 +
17 0.12 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0048 0.031 0.1 +
18 0.098 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0014 0.031 0.81 +
19 0.11 -0.41 -1.3 -2.2 1.5 5.2e+03 0.0042 0.031 0.28 +
20 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.0016 0.031 0.56 +
21 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00045 0.031 0.72 +
22 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00045 0.016 -1.9 -
23 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00045 0.0078 -0.48 -
24 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0007 0.0078 0.38 +
25 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00026 0.0078 0.68 +
26 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00026 0.0039 -3.7 -
27 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00026 0.002 -0.46 -
28 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 6.8e-05 0.002 0.69 -
Biogeme parameters read from biogeme.toml.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
*** Initial values of the parameters are obtained from the file __b05normal_mixture_algo_cg_False_radius_1.0_second_deriv_1.0.iter
Parameter values restored from __b05normal_mixture_algo_cg_False_radius_1.0_second_deriv_1.0.iter
The number of draws (100) is low. The results may not be meaningful.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
Optimization algorithm: hybrid Newton/BFGS with simple bounds [simple_bounds]
** Optimization: BFGS with trust region for simple bounds
Iter. ASC_CAR ASC_TRAIN B_COST B_TIME B_TIME_S Function Relgrad Radius Rho
0 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 0.5 -1.8 -
1 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 0.25 -0.55 -
2 0.095 -0.95 -1.3 -1.5 1.4 5.3e+03 0.048 0.25 0.2 +
3 -0.15 -0.7 -1.1 -1.8 1.1 5.2e+03 0.028 0.25 0.37 +
4 -0.15 -0.7 -1.1 -1.8 1.1 5.2e+03 0.028 0.12 -0.047 -
5 -0.03 -0.58 -1.2 -1.7 1.2 5.2e+03 0.033 0.12 0.38 +
6 -0.04 -0.63 -1.2 -1.8 1.2 5.2e+03 0.0069 0.12 0.85 +
7 0.028 -0.51 -1.2 -1.8 1.3 5.2e+03 0.017 0.12 0.66 +
8 0.024 -0.52 -1.2 -2 1.3 5.2e+03 0.0076 0.12 0.46 +
9 0.091 -0.47 -1.2 -2 1.4 5.2e+03 0.0072 0.12 0.72 +
10 0.091 -0.47 -1.2 -2 1.4 5.2e+03 0.0072 0.062 -1.4 -
11 0.091 -0.47 -1.2 -2 1.4 5.2e+03 0.0072 0.031 -0.15 -
12 0.059 -0.46 -1.2 -2 1.4 5.2e+03 0.0042 0.031 0.12 +
13 0.091 -0.47 -1.2 -2.1 1.4 5.2e+03 0.0034 0.031 0.88 +
14 0.066 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0039 0.031 0.42 +
15 0.097 -0.45 -1.3 -2.1 1.5 5.2e+03 0.0019 0.031 0.72 +
16 0.092 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0019 0.031 0.83 +
17 0.11 -0.44 -1.3 -2.2 1.5 5.2e+03 0.0019 0.031 0.52 +
18 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.0039 0.031 0.47 +
19 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.0039 0.016 -0.89 -
20 0.11 -0.43 -1.3 -2.2 1.6 5.2e+03 0.0012 0.016 0.55 +
21 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00091 0.16 0.91 ++
22 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00091 0.078 -11 -
23 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00091 0.039 -3.8 -
24 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00091 0.02 -0.75 -
25 0.11 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0014 0.02 0.2 +
26 0.11 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0014 0.0098 -1.5 -
27 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00049 0.0098 0.69 +
28 0.12 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00049 0.0049 -0.46 -
29 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00091 0.0049 0.44 +
30 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0003 0.0049 0.6 +
31 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00019 0.049 0.94 ++
32 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00019 0.013 -4.6 -
33 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00027 0.013 0.4 +
34 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00027 0.0064 -13 -
35 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00027 0.0032 -2.4 -
36 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00012 0.0032 0.54 -
Biogeme parameters read from biogeme.toml.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
*** Initial values of the parameters are obtained from the file __b05normal_mixture_algo_cg_False_radius_10.0_second_deriv_0.0.iter
Parameter values restored from __b05normal_mixture_algo_cg_False_radius_10.0_second_deriv_0.0.iter
The number of draws (100) is low. The results may not be meaningful.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
Optimization algorithm: hybrid Newton/BFGS with simple bounds [simple_bounds]
** Optimization: BFGS with trust region for simple bounds
Iter. ASC_CAR ASC_TRAIN B_COST B_TIME B_TIME_S Function Relgrad Radius Rho
0 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 5 -3 -
1 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 2.5 -3.3 -
2 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 1.2 -3.2 -
3 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 0.62 -2.3 -
4 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 0.31 -0.9 -
5 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 0.16 0.0043 -
6 0.0016 -0.86 -1.2 -1.4 1.5 5.3e+03 0.048 0.16 0.49 +
7 -0.15 -0.75 -1.1 -1.6 1.3 5.2e+03 0.015 0.16 0.51 +
8 0.0016 -0.59 -1.2 -1.7 1.1 5.2e+03 0.012 0.16 0.54 +
9 0.0016 -0.59 -1.2 -1.7 1.1 5.2e+03 0.012 0.078 -0.14 -
10 -0.065 -0.52 -1.2 -1.8 1.2 5.2e+03 0.014 0.078 0.11 +
11 0.014 -0.59 -1.2 -1.9 1.2 5.2e+03 0.0084 0.078 0.65 +
12 -0.00075 -0.52 -1.2 -1.9 1.3 5.2e+03 0.011 0.078 0.67 +
13 0.035 -0.53 -1.2 -1.9 1.3 5.2e+03 0.0047 0.078 0.76 +
14 0.071 -0.46 -1.2 -2 1.4 5.2e+03 0.011 0.078 0.61 +
15 0.051 -0.47 -1.2 -2 1.4 5.2e+03 0.0044 0.078 0.56 +
16 0.051 -0.47 -1.2 -2 1.4 5.2e+03 0.0044 0.039 -0.24 -
17 0.09 -0.47 -1.2 -2 1.4 5.2e+03 0.0046 0.039 0.62 +
18 0.074 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0032 0.039 0.69 +
19 0.098 -0.46 -1.3 -2.1 1.5 5.2e+03 0.0036 0.039 0.22 +
20 0.11 -0.43 -1.3 -2.1 1.5 5.2e+03 0.006 0.039 0.49 +
21 0.11 -0.43 -1.3 -2.1 1.5 5.2e+03 0.006 0.02 -0.53 -
22 0.092 -0.44 -1.2 -2.1 1.5 5.2e+03 0.0018 0.02 0.37 +
23 0.11 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0015 0.2 0.91 ++
24 0.11 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0015 0.098 -1 -
25 0.11 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0015 0.049 -0.0026 -
26 0.1 -0.42 -1.3 -2.2 1.6 5.2e+03 0.0025 0.049 0.3 +
27 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.003 0.049 0.27 +
28 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.003 0.024 -0.84 -
29 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.003 0.012 -0.055 -
30 0.13 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00081 0.012 0.74 +
31 0.13 -0.42 -1.3 -2.2 1.6 5.2e+03 0.00081 0.0061 -0.082 -
32 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0013 0.0061 0.24 +
33 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00021 0.0061 0.59 +
34 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00021 0.0031 -0.039 -
35 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00018 0.0031 0.5 +
36 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00018 0.0015 -1.2 -
37 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 6.1e-05 0.0015 0.47 -
Biogeme parameters read from biogeme.toml.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
*** Initial values of the parameters are obtained from the file __b05normal_mixture_algo_cg_False_radius_10.0_second_deriv_0.5.iter
Parameter values restored from __b05normal_mixture_algo_cg_False_radius_10.0_second_deriv_0.5.iter
The number of draws (100) is low. The results may not be meaningful.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
Optimization algorithm: hybrid Newton/BFGS with simple bounds [simple_bounds]
** Optimization: BFGS with trust region for simple bounds
Iter. ASC_CAR ASC_TRAIN B_COST B_TIME B_TIME_S Function Relgrad Radius Rho
0 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 5 -3 -
1 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 2.5 -3.3 -
2 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 1.2 -3.2 -
3 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 0.62 -2.3 -
4 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 0.31 -0.91 -
5 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 0.16 0.0018 -
6 0.0016 -0.86 -1.2 -1.4 1.4 5.3e+03 0.047 0.16 0.49 +
7 -0.15 -0.75 -1.1 -1.6 1.3 5.2e+03 0.014 0.16 0.51 +
8 0.0016 -0.59 -1.2 -1.7 1.1 5.2e+03 0.012 0.16 0.54 +
9 0.0016 -0.59 -1.2 -1.7 1.1 5.2e+03 0.012 0.078 -0.13 -
10 -0.064 -0.52 -1.2 -1.8 1.2 5.2e+03 0.014 0.078 0.11 +
11 0.014 -0.59 -1.2 -1.9 1.2 5.2e+03 0.0083 0.078 0.65 +
12 -0.00039 -0.52 -1.2 -1.9 1.3 5.2e+03 0.011 0.078 0.66 +
13 0.035 -0.53 -1.2 -1.9 1.3 5.2e+03 0.0047 0.078 0.76 +
14 0.071 -0.46 -1.2 -2 1.4 5.2e+03 0.011 0.078 0.61 +
15 0.05 -0.47 -1.2 -2 1.4 5.2e+03 0.0046 0.078 0.56 +
16 0.05 -0.47 -1.2 -2 1.4 5.2e+03 0.0046 0.039 -0.23 -
17 0.089 -0.47 -1.2 -2 1.4 5.2e+03 0.0046 0.039 0.64 +
18 0.074 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0029 0.039 0.74 +
19 0.1 -0.46 -1.3 -2.1 1.5 5.2e+03 0.0036 0.039 0.22 +
20 0.11 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0058 0.039 0.53 +
21 0.11 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0058 0.02 -0.43 -
22 0.091 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0018 0.02 0.4 +
23 0.11 -0.43 -1.3 -2.2 1.5 5.2e+03 0.001 0.02 0.88 +
24 0.11 -0.43 -1.3 -2.2 1.5 5.2e+03 0.001 0.0098 0.099 -
25 0.1 -0.42 -1.3 -2.2 1.5 5.2e+03 0.0014 0.0098 0.36 +
26 0.11 -0.43 -1.3 -2.2 1.5 5.2e+03 0.00075 0.098 0.93 ++
27 0.14 -0.39 -1.3 -2.3 1.6 5.2e+03 0.0011 0.098 0.21 +
28 0.14 -0.39 -1.3 -2.3 1.6 5.2e+03 0.0011 0.049 -8.7 -
29 0.14 -0.39 -1.3 -2.3 1.6 5.2e+03 0.0011 0.024 -1 -
30 0.12 -0.41 -1.3 -2.2 1.6 5.2e+03 0.0015 0.024 0.39 +
31 0.13 -0.4 -1.3 -2.2 1.6 5.2e+03 0.0007 0.024 0.5 +
32 0.13 -0.4 -1.3 -2.2 1.6 5.2e+03 0.0007 0.012 -3 -
33 0.13 -0.4 -1.3 -2.2 1.6 5.2e+03 0.0007 0.0061 -0.68 -
34 0.13 -0.4 -1.3 -2.2 1.6 5.2e+03 0.00067 0.0061 0.24 +
35 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00045 0.0061 0.46 +
36 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00029 0.0061 0.2 +
37 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00031 0.0061 0.72 +
38 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00031 0.0031 -2.1 -
39 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00031 0.0015 -0.7 -
40 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00019 0.0015 0.36 +
41 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 5.4e-05 0.0015 0.74 +
Biogeme parameters read from biogeme.toml.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
*** Initial values of the parameters are obtained from the file __b05normal_mixture_algo_cg_False_radius_10.0_second_deriv_1.0.iter
Parameter values restored from __b05normal_mixture_algo_cg_False_radius_10.0_second_deriv_1.0.iter
The number of draws (100) is low. The results may not be meaningful.
As the model is rather complex, we cancel the calculation of second derivatives. If you want to control the parameters, change the name of the algorithm in the TOML file from "automatic" to "simple_bounds"
Optimization algorithm: hybrid Newton/BFGS with simple bounds [simple_bounds]
** Optimization: BFGS with trust region for simple bounds
Iter. ASC_CAR ASC_TRAIN B_COST B_TIME B_TIME_S Function Relgrad Radius Rho
0 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 5 -3 -
1 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 2.5 -3.3 -
2 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 1.2 -3.2 -
3 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 0.62 -2.3 -
4 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 0.31 -0.9 -
5 -0.15 -0.7 -1.1 -1.3 1.6 5.3e+03 0.066 0.16 0.0051 -
6 0.0016 -0.86 -1.2 -1.4 1.5 5.3e+03 0.048 0.16 0.49 +
7 -0.15 -0.75 -1.1 -1.6 1.3 5.2e+03 0.015 0.16 0.51 +
8 0.0016 -0.59 -1.2 -1.7 1.1 5.2e+03 0.012 0.16 0.54 +
9 0.0016 -0.59 -1.2 -1.7 1.1 5.2e+03 0.012 0.078 -0.14 -
10 -0.065 -0.52 -1.2 -1.8 1.2 5.2e+03 0.014 0.078 0.1 +
11 0.013 -0.59 -1.2 -1.9 1.2 5.2e+03 0.0084 0.078 0.65 +
12 -0.00086 -0.52 -1.2 -1.9 1.3 5.2e+03 0.011 0.078 0.67 +
13 0.035 -0.53 -1.2 -1.9 1.3 5.2e+03 0.0047 0.078 0.76 +
14 0.07 -0.46 -1.2 -2 1.4 5.2e+03 0.011 0.078 0.61 +
15 0.051 -0.47 -1.2 -2 1.4 5.2e+03 0.0044 0.078 0.57 +
16 0.051 -0.47 -1.2 -2 1.4 5.2e+03 0.0044 0.039 -0.24 -
17 0.09 -0.47 -1.2 -2 1.4 5.2e+03 0.0046 0.039 0.62 +
18 0.073 -0.44 -1.3 -2.1 1.5 5.2e+03 0.0033 0.039 0.68 +
19 0.098 -0.46 -1.3 -2.1 1.5 5.2e+03 0.0036 0.039 0.22 +
20 0.11 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0061 0.039 0.47 +
21 0.11 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0061 0.02 -0.51 -
22 0.093 -0.44 -1.2 -2.1 1.5 5.2e+03 0.0017 0.02 0.41 +
23 0.11 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0017 0.2 0.9 ++
24 0.11 -0.43 -1.3 -2.1 1.5 5.2e+03 0.0017 0.098 -0.76 -
25 0.14 -0.42 -1.3 -2.2 1.6 5.2e+03 0.0035 0.098 0.24 +
26 0.14 -0.42 -1.3 -2.2 1.6 5.2e+03 0.0035 0.049 -3.2 -
27 0.14 -0.42 -1.3 -2.2 1.6 5.2e+03 0.0035 0.024 -1.1 -
28 0.14 -0.42 -1.3 -2.2 1.6 5.2e+03 0.0035 0.012 -0.29 -
29 0.12 -0.4 -1.3 -2.2 1.6 5.2e+03 0.00098 0.012 0.44 +
30 0.12 -0.4 -1.3 -2.2 1.6 5.2e+03 0.00098 0.0061 0.093 -
31 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00022 0.061 0.91 ++
32 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00022 0.031 -7.6 -
33 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00022 0.015 -2.5 -
34 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00022 0.0076 -0.33 -
35 0.13 -0.4 -1.3 -2.2 1.6 5.2e+03 0.00035 0.0076 0.43 +
36 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00023 0.0076 0.56 +
37 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00023 0.0038 -1.9 -
38 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00023 0.0019 -0.65 -
39 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 0.00035 0.0019 0.23 +
40 0.13 -0.41 -1.3 -2.2 1.6 5.2e+03 8e-05 0.0019 0.59 +
summary
SUMMARY_FILE = '05normalMixture_allAlgos.csv'
summary.to_csv(SUMMARY_FILE, index=False)
print(f'Summary reported in file {SUMMARY_FILE}')
Summary reported in file 05normalMixture_allAlgos.csv
Total running time of the script: (4 minutes 0.277 seconds)