Change parameters automatically

Hi,
I am using the Inert Doublet Model UFO model which is available at Feynrules website. I want to change a parameter value taking the new values from an external file.
I read something about the "set" command,
set mmH0 value

but it doesn't work:
Command "set mmH0 7E1" interrupted with error:
InvalidCmd : Possible options for set are ['group_subprocesses', 'ignore_six_quark_processes', 'stdout_level', 'fortran_compiler', 'cpp_compiler', 'loop_optimized_output', 'complex_mass_scheme', 'gauge', 'EWscheme', 'max_npoint_for_channel']
-- set options for generation or output.
syntax: set <option_name> <option_value>
Possible options are:
group_subprocesses,ignore_six_quark_processes,stdout_level
fortran_compiler,cpp_compiler,loop_optimized_output
complex_mass_scheme,gauge,EWscheme
Details of each option:
group_subprocesses True/False/Auto:
 > (default Auto) Smart grouping of subprocesses into
   directories, mirroring of initial states, and
   combination of integration channels.
 > Example: p p > j j j w+ gives 5 directories and 184 channels
   (cf. 65 directories and 1048 channels for regular output)
 > Auto means False for decay computation and True for collisions.
ignore_six_quark_processes multi_part_label
 > (default none) ignore processes with at least 6 of any
   of the quarks given in multi_part_label.
 > These processes give negligible contribution to the
   cross section but have subprocesses/channels.
stdout_level DEBUG|INFO|WARNING|ERROR|CRITICAL
 > change the default level for printed information
fortran_compiler NAME
 > (default None) Force a specific fortran compiler.
   If None, it tries first g77 and if not present gfortran
   but loop output use gfortran.
loop_optimized_output True|False
 > Exploits the open loop thechnique for considerable
   improvement.
 > CP relations among helicites are detected and the helicity
   filter has more potential.
loop_color_flows True|False
 > Only relevant for the loop optimized output.
 > Reduces the loop diagrams at the amplitude level
   rendering possible the computation of the loop amplitude
   for a fixed color flow or color configuration.
 > This option can considerably slow down the loop ME
   computation time, especially when summing over all color
   and helicity configuration, hence turned off by default.
gauge unitary|Feynman
 > (default unitary) choose the gauge of the non QCD part.
 > For loop processes, only Feynman gauge is employable.
complex_mass_scheme True|False
 > (default False) Set complex mass scheme.
 > Complex mass scheme is not yet supported for loop processes.
timeout VALUE
 > (default 20) Seconds allowed to answer questions.
 > Note that pressing tab always stops the timer.
cluster_temp_path PATH
 > (default None) [Used in Madevent Output]
 > Allow to perform the run in PATH directory
 > This allow to not run on the central disk.
 > This is not used by condor cluster (since condor has
   its own way to prevent it).
OLP ProgramName
 > (default 'MadLoop') [Used for virtual generation]
 > Chooses what One-Loop Program to use for the virtual
 > matrix element generation via the BLAH accord.
output_dependencies <mode>
 > (default 'external') [Use for NLO outputs]
 > Choses how the external dependences (such as CutTools)
 > of NLO outputs are handled. Possible values are:
     o external: Some of the libraries the output depends
       on are links to their installation in MG5 root dir.
     o internal: All libraries the output depends on are
       copied and compiled locally in the output directory.
     o environment_paths: The location of all libraries the
       output depends on should be found in your env. paths.

What I am doing wrong?
Thanks!
Sebastián