Trusted, Bureaucrats
1,382
edits
Changes
Jump to navigation
Jump to search
Make a '''06_check_intermediate_iteration_chi2.py ''' file, with this content. {| class="mw-collapsible mw-collapsed wikitable"! See file content|-|<source lang="python"># Python module imports.import os # relax module imports.from pipe_control import pipesimport lib.iofrom specific_analyses.api import return_api # Read the state with the setupvar = 'result_06_check_intermediate'results_dir = os.getcwd() + os.sep + var + os.sep + 'final'# Load the state with setup data.state.load(state='results.bz2', dir=results_dir, force=True) # Show pipespipe.display()pipe_name = pipes.cdp_name()pipe_bundle = pipes.get_bundle(pipe_name) # Define write outwrite_out = results_dir + os.sep + 'grace' # chi2 per iteration? But does not work?grace.write(x_data_type='iter', y_data_type='chi2', file='iter_chi2.agr', dir=write_out, force=True) ############# # This does not do what we want. So let us try manually.var_ori = 'result_06'results_dir_ori = os.getcwd() + os.sep + var_ori dir_list = os.listdir(results_dir_ori) all_models = ['local_tm', 'sphere', 'prolate', 'oblate', 'ellipsoid']opt_models = []for model in all_models: if model in dir_list: opt_models.append(model) # Loop over models MII to MV.out_results = []for model in ['sphere', 'prolate', 'oblate', 'ellipsoid']: # Skip missing models. if model not in opt_models: continue # Make the model dir mdir = results_dir_ori + os.sep + model rdir = [ name for name in os.listdir(mdir) if os.path.isdir(os.path.join(mdir, name)) ] rdirs = lib.io.sort_filenames(rdir) # Loop over rounds for rd in rdirs: if "round_" in rd: dir_model_round = mdir + os.sep + rd + os.sep + 'opt' if os.path.isdir(dir_model_round): # Create pipe to read data pipe_name_rnd = "%s_%s" % (model, rd) pipe.create(pipe_name_rnd, 'mf', bundle="temp") results.read(file='results', dir=dir_model_round) # Get info round_i = rd.split("_")[-1] cdp_iter = str(cdp.iter) chi2 = str(cdp.chi2) tm = str(cdp.diff_tensor.tm) # Get the api to get number of parameters api = return_api(pipe_name=pipe_name) model_loop = api.model_loop model_desc = api.model_desc model_statistics = api.model_statistics for model_info in model_loop(): desc = model_desc(model_info) # Num_params_(k) # Num_data_sets_(n) k_glob, n_glob, chi2_glob = model_statistics(model_info, global_stats=True) break k_glob = str(k_glob) n_glob = str(n_glob) chi2_glob = str(chi2_glob) # Append to results out_results.append([pipe_name_rnd, model, round_i, cdp_iter, chi2, tm, k_glob, n_glob, chi2_glob]) print("\n# Data:") print(out_results[-1]) # Change back to original pipepipe.switch(pipe_name)cdp.out_results = out_results #print resultfor res in out_resultshttps: print res # Write filefile_name = "results_collected//github.txt"file_path = libcom/tlinnet/relax_modelfree_scripts/blob/master/06_check_intermediate_iteration_chi2.iopy 06_check_intermediate_iteration_chi2.get_file_path(file_name, results_dir)file = lib.io.open_write_file(file_path, force=True) # Write the file.headings = ["pipe_name", "model", "round_i", "cdp_iter", "chi2", "tm", "k_glob_Num_params", "n_glob_Num_data_sets", "chi2_glob"py]lib.io.write_data(out=file, headings=headings, data=out_results)file.close() # Save the statestate.save(state='results_collected.bz2', dir=results_dir, force=True)</source>|}
This will also get '''k''' The global number parameters and '''n''' the global number of data sets.
Run with relax
