fix: MemoryError when plotting log-log data (#264) + some cleanup

CHANGELOG

@@ -1,3 +1,5 @@
+1.0.1
+ - fix: MemoryError when plotting log-log data (#264)
 1.0.0
  - fix: change the way measurements are recognized (.rtdc files were
    falsely grouped together when located in the same directory) (#262)

shapeout/analysis.py

@@ -141,7 +141,9 @@ def _complete_config(self, measurements=None):
                     if var not in pltng:
                         data = mm[kk]
                         if sc == "log":
-                            data = np.log(data)
+                            # this does not seem to be necessary (issue #264)
+                            #data = np.log(data)
+                            pass
                         acc = l(data) * mult
                         # round to make it look pretty in the GUI
                         accr = float("{:.1e}".format(acc))
@@ -161,10 +163,10 @@ def _doanes_formula_acc(a):
         bad = np.isnan(a) | np.isinf(a)
         data = a[~bad]
         n = data.size
-        g1 = scipy.stats.skew(data)
-        sigma_g1 = np.sqrt(6 * (n - 2) / ((n + 1) * (n + 3)))
-        k = 1 + np.log2(n) + np.log2(1 + np.abs(g1) / sigma_g1)
-        if len(data):
+        if n:
+            g1 = scipy.stats.skew(data)
+            sigma_g1 = np.sqrt(6 * (n - 2) / ((n + 1) * (n + 3)))
+            k = 1 + np.log2(n) + np.log2(1 + np.abs(g1) / sigma_g1)
             acc = (data.max() - data.min()) / k
         else:
             acc = 1

shapeout/gui/plot_common.py

@@ -81,8 +81,8 @@ def get_kde_kwargs(x, y, kde_type, xacc, yacc):
             biny = naninfminmaxdiff(y)/(1.8*yacc)
         except:
             biny = 5
-        binx = int(max(5, binx))
-        biny = int(max(5, biny))
+        binx = int(max(5, binx)) #if binx < x.size else 20
+        biny = int(max(5, biny)) #if biny < y.size else 20
         kde_kwargs["bins"] = [binx, biny]
     return kde_kwargs
 
@@ -113,7 +113,7 @@ def my_log_auto_ticks(data_low, data_high,
 
 
 def naninfminmaxdiff(x):
-    bad = np.isnan(x)+np.isinf(x)
+    bad = np.isnan(x) | np.isinf(x)
     x = x[~bad]
     diff = (x.max()-x.min())
     return diff

shapeout/gui/plot_scatter.py

@@ -214,13 +214,14 @@ def set_scatter_data(plot, mm):
                                             downsample=downsample,
                                             xscale=scalex,
                                             yscale=scaley,
+                                            remove_invalid=True,
                                             ret_mask=True,
                                             )
     if lx == x.shape:
         positions = None
     else:
         print("...Downsampled from {} to {} in {:.2f}s".format(lx, x.shape[0], time.time()-a))
-        positions = np.vstack([x.ravel(), y.ravel()])
+        positions = (x, y)
 
 
     kde_type = mm.config["plotting"]["kde"].lower()
@@ -236,10 +237,10 @@ def set_scatter_data(plot, mm):
     density = mm.get_kde_scatter(xax=xax,
                                  yax=yax,
                                  positions=positions,
-                                 xscale=scalex,
-                                 yscale=scaley,
                                  kde_type=kde_type,
                                  kde_kwargs=kde_kwargs,
+                                 xscale=scalex,
+                                 yscale=scaley,
                                  )
     print("...KDE scatter time {}: {:.2f}s".format(kde_type, time.time()-a))