gets lack of physio bars working

billbrod · Oct 21, 2024 · 316bfb3 · 316bfb3
1 parent 8734716
commit 316bfb3
Show file tree

Hide file tree

Showing 2 changed files with 11 additions and 10 deletions.
diff --git a/Snakefile b/Snakefile
@@ -51,7 +51,6 @@ wildcard_constraints:
     save_all='|_saveall',
     gammacorrected='|_gamma-corrected',
     plot_focus='|_focus-subject|_focus-image',
-    physio='|_physio-bars',
     ecc_mask="|_eccmask-[0-9]+",
     logscale="log|linear",
     mcmc_model="partially-pooled|unpooled|partially-pooled-interactions-[.0-9]+|partially-pooled-interactions",
@@ -2436,13 +2435,13 @@ rule mcmc_performance_comparison_figure:
         op.join(config['DATA_DIR'], 'dacey_data',
                 'Dacey1992_mcmc_line-nooffset_step-.1_prob-.8_depth-10_c-4_d-1000_w-1000_s-10.nc'),
     output:
-        op.join(config['DATA_DIR'], 'figures', '{context}', 'mcmc{scaling_extended}_{mcmc_model}_{mcmc_plot_type}_{focus}{physio}.{ext}'),
+        op.join(config['DATA_DIR'], 'figures', '{context}', 'mcmc{scaling_extended}_{mcmc_model}_{mcmc_plot_type}_{focus}.{ext}'),
     log:
         op.join(config['DATA_DIR'], 'logs', 'figures', '{context}',
-                'mcmc{scaling_extended}_{mcmc_model}_{mcmc_plot_type}_{focus}{physio}_{ext}.log')
+                'mcmc{scaling_extended}_{mcmc_model}_{mcmc_plot_type}_{focus}_{ext}.log')
     benchmark:
         op.join(config['DATA_DIR'], 'logs', 'figures', '{context}',
-                'mcmc{scaling_extended}_{mcmc_model}_{mcmc_plot_type}_{focus}{physio}_{ext}_benchmark.txt')
+                'mcmc{scaling_extended}_{mcmc_model}_{mcmc_plot_type}_{focus}_{ext}_benchmark.txt')
     run:
         import foveated_metamers as fov
         import contextlib
@@ -2539,7 +2538,7 @@ rule mcmc_performance_comparison_figure:
                                                                logscale_xaxis=True,
                                                                increase_size=False)
                     g.fig.canvas.draw()
-                    if 'bars' in wildcards.physio:
+                    if 'physio-bars' in wildcards.focus:
                         fov.plotting.add_physiological_scaling_bars(g.ax, az.from_netcdf(input[-1]))
                     fig = g.fig
                     if 'line' in wildcards.focus:
@@ -2660,11 +2659,11 @@ rule performance_comparison_figure:
         op.join(config['DATA_DIR'], 'dacey_data',
                 'Dacey1992_mcmc_line-offset_step-.1_prob-.8_depth-10_c-4_d-1000_w-1000_s-10.nc'),
     output:
-        op.join(config['DATA_DIR'], 'figures', '{context}', 'performance_{focus}{physio}.svg')
+        op.join(config['DATA_DIR'], 'figures', '{context}', 'performance_{focus}.svg')
     log:
-        op.join(config['DATA_DIR'], 'logs', 'figures', '{context}', 'performance_{focus}{physio}.log')
+        op.join(config['DATA_DIR'], 'logs', 'figures', '{context}', 'performance_{focus}.log')
     benchmark:
-        op.join(config['DATA_DIR'], 'logs', 'figures', '{context}', 'performance_{focus}{physio}_benchmark.txt')
+        op.join(config['DATA_DIR'], 'logs', 'figures', '{context}', 'performance_{focus}_benchmark.txt')
     run:
         import foveated_metamers as fov
         import pandas as pd
@@ -2718,7 +2717,7 @@ rule performance_comparison_figure:
                                                  style='trial_type',
                                                  aspect=2 if col is None else 1,
                                                  logscale_xaxis=logscale_xaxis)
-                if col is None and 'bars' in wildcards.physio:
+                if col is None and 'physio-bars' in wildcards.focus:
                     # need to draw so that the following code can check text size
                     g.fig.canvas.draw()
                     fov.plotting.add_physiological_scaling_bars(g.ax, az.from_netcdf(input[-1]))

diff --git a/foveated_metamers/plotting.py b/foveated_metamers/plotting.py
@@ -1373,7 +1373,9 @@ def lineplot_like_pointplot(data, x, y, col=None, row=None, hue=None, ci=95,
     # used in pointplot (which uses scatter). I also added an extra factor of
     # sqrt(2) (by changing the 2 to a 4 in the sqrt below), which looks
     # necessary
-    ms = kwargs.pop('markersize', np.sqrt(np.pi * np.square(lw) * 4))
+    ms = kwargs.pop('markersize', None)
+    if ms is None:
+        ms = np.sqrt(np.pi * np.square(lw) * 4)
     if ax is None:
         if col is not None:
             col_order = kwargs.pop('col_order', sorted(data[col].unique()))