added wilson plot + d to second x-axis + improved log file

reduction_tools/partialator.py

@@ -38,6 +38,8 @@ from tqdm import tqdm
 import regex as re
 import matplotlib.pyplot as plt
 from scipy.optimize import curve_fit
+import warnings
+warnings.filterwarnings( "ignore", category=RuntimeWarning ) 
 from loguru import logger
 
 def submit_job( job_file, reservation ):
@@ -195,9 +197,15 @@ def get_metric( d2_series, cc_series, cut_off ):
     # get curve for plotting
     cc_tanh = tanh( d2_series, r, s0 )
 
-    return cc_stat, cc_tanh
+    return round( cc_stat, 3 ), cc_tanh
 
-def summary_fig( stats_df, cc_tanh, ccstar_tanh ):
+def summary_fig( stats_df, cc_tanh, ccstar_tanh, cc_cut, ccstar_cut ):
+
+    def dto1_d( x ):
+        return 1/x
+        
+    def dto1_d2( x ):
+        return 1/x**2
 
     # plot results
     cc_fig, axs = plt.subplots(2, 2)
@@ -212,9 +220,11 @@ def summary_fig( stats_df, cc_tanh, ccstar_tanh ):
     # plot cc
     axs[0,0].plot( stats_df[ "1_d" ], stats_df.cc, color=color )
     # plot fit
-    axs[0,0].plot( stats_df[ "1_d2" ], cc_tanh, color="tab:grey", linestyle = "dashed" )
-    axs[0,0].xticks( stats_df[ "1_d2" ].iloc[::5, :], stats_df[ "d" ].iloc[::5, :] )
+    axs[0,0].plot( stats_df[ "1_d" ], cc_tanh, color="tab:grey", linestyle = "dashed" )
+    sax1 = axs[0,0].secondary_xaxis( 'top', functions=( dto1_d, dto1_d ) )
+    sax1.set_xlabel('d (A)')
     axs[0,0].tick_params( axis="y", labelcolor=color )
+    axs[0,0].text( 0.1, 0.1, "CC @ 0.2 = {0}".format( cc_cut ), fontsize = 8 )
 
     # cc* plot
     color = "tab:blue"
@@ -224,15 +234,19 @@ def summary_fig( stats_df, cc_tanh, ccstar_tanh ):
     axs[0,1].axhline( y = 0.7, color="black", linestyle = "dashed" )
     axs[0,1].plot( stats_df[ "1_d" ], stats_df.ccstar, color=color )
     # plot fit
-    axs[0,0].plot( stats_df[ "1_d2" ], ccstar_tanh, color="tab:grey", linestyle = "dashed" )
-    axs[0,0].xticks( stats_df[ "1_d2" ].iloc[::5, :], stats_df[ "d" ].iloc[::5, :] )
+    axs[0,1].plot( stats_df[ "1_d" ], ccstar_tanh, color="tab:grey", linestyle = "dashed" )
+    sax2 = axs[0,1].secondary_xaxis( 'top', functions=( dto1_d, dto1_d ) )
+    sax2.set_xlabel('d (A)')
     axs[0,1].tick_params( axis='y', labelcolor=color )
+    axs[0,1].text( 0.1, 0.1, "CC* @ 0.7 = {0}".format( ccstar_cut ) , fontsize = 8 )
 
     # rsplit plot
     color = "tab:green"
     axs[1,0].set_xlabel( "1/d (1/A)" )
     axs[1,0].set_ylabel( "Rsplit", color=color )
     axs[1,0].plot( stats_df[ "1_d" ], stats_df.rsplit, color=color )
+    sax3 = axs[1,0].secondary_xaxis( 'top', functions=( dto1_d, dto1_d ) )
+    sax3.set_xlabel('d (A)')
     axs[1,0].tick_params( axis='y', labelcolor=color )
 
 
@@ -241,37 +255,14 @@ def summary_fig( stats_df, cc_tanh, ccstar_tanh ):
     axs[1,1].set_xlabel( "d (A)" )
     axs[1,1].set_ylabel( "lnI", color=color )
     axs[1,1].plot( stats_df[ "1_d2" ], stats_df.lnI, color=color )
-#    axs[1,1].invert_xaxis()
+    sax4 = axs[1,1].secondary_xaxis( 'top', functions=( dto1_d2, dto1_d2 ) )
+    sax4.set_xlabel('d (A)')
     axs[1,1].tick_params( axis='y', labelcolor=color )
 
     # save figure
     plt.tight_layout()
     plt.savefig( "plots.png" )
 
-def get_mean_cell( stream ):
-
-    # get uc values from stream file
-    # example - Cell parameters 7.71784 7.78870 3.75250 nm, 90.19135 90.77553 90.19243 deg
-    # scrub clen and return - else nan
-    try:
-        pattern = r"Cell\sparameters\s(\d+\.\d+)\s(\d+\.\d+)\s(\d+\.\d+)\snm,\s(\d+\.\d+)\s(\d+\.\d+)\s(\d+\.\d+)\sdeg"
-        cell_lst = re.findall( pattern, stream )
-        xtals = len( cell_lst )
-    except AttributeError:
-        return np.nan
-    
-    cols = [ "a", "b", "c", "alpha", "beta", "gamma" ]
-    cell_df = pd.DataFrame( cell_lst, columns=cols )
-
-    mean_a = round( cell_df.a.mean()*10, 3 )
-    mean_b = round( cell_df.b.mean()*10, 3 )
-    mean_c = round( cell_df.c.mean()*10, 3 )
-    mean_alpha = round( cell_df.alpha.mean(), 3 )
-    mean_beta = round( cell_df.beta.mean(), 3 )
-    mean_gamma = round( cell_df.gamma.mean(), 3 )
-
-    return mean_a, mean_b, mean_c, mean_alpha, mean_beta, mean_gamma
-
 def main( cwd, name, stream, pointgroup, model, iterations, cell, shells, part_h_res, adu, reservation ):
 
     # submitted job set
@@ -307,22 +298,23 @@ def main( cwd, name, stream, pointgroup, model, iterations, cell, shells, part_h
 
     # make summary data table
     stats_df = summary_stats( cc_dat, ccstar_dat, mult_dat, rsplit_dat, wilson_dat  )
-    logger.info( "stats table from .dat file =\n{0}".format( stats_df ) )
+    logger.info( "stats table from .dat file =\n{0}".format( stats_df.to_string() ) )
     print_df = stats_df[ [  "1_d", "d", "min",
                             "max", "nref", "poss",
                             "comp", "obs", "mult",
                             "snr", "I", "rsplit", "cc", "ccstar" ] ]
     print_df.to_csv( "summary_table.csv", sep="\t", index=False )
 
-
     # calculate cc metrics
     cc_cut, cc_tanh = get_metric( stats_df[ "1_d2" ], stats_df.cc, 0.3 )
     ccstar_cut, ccstar_tanh = get_metric( stats_df[ "1_d2" ], stats_df.ccstar, 0.7 )
     print( "resolution at CC0.5 at 0.3 = {0}".format( cc_cut ) )
     print( "resolution at CC* at 0.7 = {0}".format( ccstar_cut ) )
+    logger.info( "resolution at CC0.5 at 0.3 = {0}".format( cc_cut ) )
+    logger.info( "resolution at CC* at 0.7 = {0}".format( ccstar_cut ) )
 
     # show plots
-    summary_fig( stats_df, cc_tanh, ccstar_tanh )
+    summary_fig( stats_df, cc_tanh, ccstar_tanh, cc_cut, ccstar_cut )
 
     # move back to top dir
     os.chdir( cwd )