Trusted, Bureaucrats
1,382
edits
Changes
Jump to navigation
Jump to search
* View -> Data pipe editor
* Right click on pipe, and select "Associate with a new auto-analysis"
=== Try fast run ===
Now we try a fast run, to see if everything is setup
Copy '''03_save_state_inspect_GUI.py''' to '''04_run_default_with_tolerance_lim.py''' and modify last lines:
<source lang="python">
# The results dir.
var = 'result_04'
results_dir = os.getcwd() + os.sep + var
# Save the state before running. Open and check in GUI!
state.save(state=var+'_ini.bz2', dir=results_dir, force=True)
# To check in GUI
# relax -g
# File -> Open relax state
# In folder "result_03" open "result_03_ini.bz2"
# View -> Data pipe editor
# Right click on pipe, and select "Associate with a new auto-analysis"
dAuvergne_protocol(pipe_name=name, pipe_bundle=bundle_name, results_dir=results_dir, diff_model=DIFF_MODEL, mf_models=MF_MODELS, local_tm_models=LOCAL_TM_MODELS, grid_inc=GRID_INC, min_algor=MIN_ALGOR, mc_sim_num=MC_NUM, max_iter=MAX_ITER, conv_loop=CONV_LOOP)
</source>
Before running, is worth to note, which values are NOT set to default values in the GUI.
* dAuvergne_protocol.opt_func_tol = 1e-10 # Standard: opt_func_tol = 1e-25
* dAuvergne_protocol.opt_max_iterations = int(1e5) # Standard: opt_max_iterations = int(1e7)
These 2 values is used in the '''minfx''' python package, and is an instruction to the minimiser function, to continue changing parameter values,
UNTIL either the difference in chi2 values between "2 steps" is less than 1e-10, OR if the number all steps is larger than 10^5.
It's an instruction not to be tooooo pedantic, here in the exploration phase. When finalising for publication, these values
should be set to their standard value.
* MC_NUM = 20
Number of Monte-Carlo simulations. The protocol will find optimum parameter values in this protocol, but error
estimation will not be very reliable. Standard is 500.
Run with
<source lang="bash">
relax 04_run_default_with_tolerance_lim.py -t 04_run_default_with_tolerance_lim.log
</source>
== See also ==
[[Category:Tutorials]]
[[Category:Model-free_analysis]]
