codestyle ...

Author: AndreasGocht

benchmark/compare.py

@@ -16,29 +16,29 @@
 with open(args.right, "rb") as f:
     right = pickle.load(f)
 
-if left.keys()!=right.keys():
+if left.keys() != right.keys():
     raise ValueError("Different Experiments")
 
 experiment = right.keys()
 
 for exp in experiment:
     print("\nExperiment: {}".format(exp))
-    if left[exp].keys()!=right[exp].keys():
+    if left[exp].keys() != right[exp].keys():
         raise ValueError("Different Instrumenters")
     instrumenters = left[exp].keys()
-    
+
     for inst in instrumenters:
         print("\n\tInstrumenter: {}".format(inst))
-        if left[exp][inst].keys()!=right[exp][inst].keys():
+        if left[exp][inst].keys() != right[exp][inst].keys():
             raise ValueError("Different Iterations")
         iterations = left[exp][inst].keys()
         Y_left = []
-        Y_right =  []
+        Y_right = []
         X = []
         for it in iterations:
             left_val = left[exp][inst][it]
             right_val = right[exp][inst][it]
-            if len(left_val)!=len(right_val):
+            if len(left_val) != len(right_val):
                 raise ValueError("Different Repetitons")
 
             Y_left.append(numpy.mean(left_val))
@@ -48,12 +48,15 @@
             if args.s:
                 print("\t\tInterations {}".format(it))
                 print("\t\tMean:     {:>7.4f} s  {:>7.4f} s".format(numpy.mean(left_val), numpy.mean(right_val)))
-                print("\t\tMedian:   {:>7.4f} s  {:>7.4f} s".format(numpy.quantile(left_val, 0.50), numpy.quantile(right_val, 0.50)))
-                print("\t\t5%:       {:>7.4f} s  {:>7.4f} s".format(numpy.quantile(left_val, 0.05), numpy.quantile(right_val, 0.05)))
-                print("\t\t95%:      {:>7.4f} s  {:>7.4f} s".format(numpy.quantile(left_val, 0.95), numpy.quantile(right_val, 0.95)))
-        Y_left = numpy.asarray(Y_left,dtype=float).flatten()
-        Y_right = numpy.asarray(Y_right,dtype=float).flatten()
-        X = numpy.asarray(X,dtype=float).flatten()
+                print("\t\tMedian:   {:>7.4f} s  {:>7.4f} s".format(
+                    numpy.quantile(left_val, 0.50), numpy.quantile(right_val, 0.50)))
+                print("\t\t5%:       {:>7.4f} s  {:>7.4f} s".format(
+                    numpy.quantile(left_val, 0.05), numpy.quantile(right_val, 0.05)))
+                print("\t\t95%:      {:>7.4f} s  {:>7.4f} s".format(
+                    numpy.quantile(left_val, 0.95), numpy.quantile(right_val, 0.95)))
+        Y_left = numpy.asarray(Y_left, dtype=float).flatten()
+        Y_right = numpy.asarray(Y_right, dtype=float).flatten()
+        X = numpy.asarray(X, dtype=float).flatten()
 
         cost_left = numpy.polyfit(X, Y_left, 1)
         cost_right = numpy.polyfit(X, Y_right, 1)
@@ -62,6 +65,3 @@
             print("")
         print("\tSlope     {:>7.4f} us {:>7.4f} us".format(cost_left[0] * 1e6, cost_right[0] * 1e6))
         print("\tIntercept {:>7.4f} s  {:>7.4f} s".format(cost_left[1], cost_right[1]))
-            
-        
-