Initial commit

Author: AlexGustafsson

README.md

@@ -0,0 +1,210 @@
+# -*- coding: utf-8 -*-
+from timeit import default_timer as timer
+
+###
+# This file will usually not need any alteration.
+###
+
+#
+
+# Class to wrap a benchmark (helper functions)
+class Benchmark(object):
+    def __init__(self):
+        self.totalTime = {}
+        self.iterations = {}
+        self.startTime = {}
+        self.depth = 0
+        self.events = []
+        # True will make all benchmark instances log their actions in an easy to follow manner
+        # Might however use up more memory as each action is logged - defaults to False. Use with benchmark.printEvents()
+        self.verbose = False
+
+    def __str__(self):
+        totalTimePrint = ', '.join('"%s": %.6f'%(key, value) for key, value in self.totalTime.items())
+        iterationsPrint = ', '.join('"%s": %d'%(key, value) for key, value in self.iterations.items())
+        return "Benchmark <Total time: {%s}, iterations: {%s}>"%(totalTimePrint, iterationsPrint)
+    def __repr__(self):
+        totalTimePrint = ', '.join('"%s": %.6f'%(key, value) for key, value in self.totalTime.items())
+        iterationsPrint = ', '.join('"%s": %d'%(key, value) for key, value in self.iterations.items())
+        return "Benchmark <Total time: {%s}, iterations: {%s}>"%(totalTimePrint, iterationsPrint)
+
+    def __add__(self, other):
+        self.mergeDictionaries(self.totalTime, other.totalTime)
+        self.mergeDictionaries(self.iterations, other.iterations)
+        self.mergeDictionaries(self.startTime, other.startTime)
+        return self
+    def __radd__(self, other):
+        return self.__add__(other)
+
+    def asTables(self):
+        timeLatex = """$\\begin{array}{l | l}
+\\text{name} & \\text{time }s\\\\
+\\hline\n """
+        iterationsLatex = """$\\begin{array}{l | l}
+\\text{name} & \\text{iterations}\\\\
+\\hline\n """
+
+        for key, value in self.totalTime.items():
+            timeLatex += "\\text{%s}&%f\\\\\n"%(key, value)
+        for key, value in self.iterations.items():
+                    iterationsLatex += "\\text{%s}&%d\\\\\n"%(key, value)
+
+        timeLatex += "\\end{array}$"
+        iterationsLatex += "\\end{array}$"
+
+        return (timeLatex, iterationsLatex)
+
+    def mergeDictionaries(self, a, b):
+        for key in b.keys():
+            if key not in a:
+                a[key] = b[key]
+            else:
+                a[key] += b[key]
+
+    def start(self, name = "default"):
+        t = timer()
+
+        if ":" in name or "," in name or "-" in name:
+            raise ValueError("A benchmark name may not contain characters ':', '-' or ','")
+
+        if name in self.startTime:
+            raise ValueError("Benchmark for %s is already in progress, please stop benchmark first"%name)
+
+        self.startTime[name] = t
+
+        if self.verbose:
+            self.events.append(("start", name, self.depth, None))
+            self.depth += 1
+
+    def stop(self, name = "default"):
+        t = timer()
+
+        if ":" in name or "," in name or "-" in name:
+            raise ValueError("A benchmark name may not contain characters ':', '-' or ','")
+
+        if name not in self.startTime:
+            raise ValueError("Benchmark for %s cannot be stopped before it has been started"%name)
+
+        if name not in self.totalTime:
+            self.totalTime[name] = 0
+
+        timeTaken = t - self.startTime[name]
+        self.totalTime[name] += timeTaken
+        del self.startTime[name]
+
+        if self.verbose:
+            self.depth -= 1
+            self.events.append(("stop", name, self.depth, timeTaken))
+
+    def iterate(self, name = "default"):
+        if ":" in name or "," in name or "-" in name:
+            raise ValueError("A benchmark name may not contain characters ':', '-' or ','")
+
+        if name not in self.iterations:
+            self.iterations[name] = 0
+
+        self.iterations[name] += 1
+
+        if self.verbose:
+            self.events.append(("iterate", name, self.depth, None))
+
+    def printEvents(self):
+        if self.verbose == False or len(self.events) == 0:
+            print "It seems like no events were captured. Try setting verbose = True on the benchmark instance"
+
+        i = 0
+        while i < len(self.events):
+            type, name, depth, timeTaken = self.events[i]
+            if type == "iterate":
+                start = i
+                # Group identical iterations to preserve output length
+                for j in range(i + 1, len(self.events)):
+                    otherType, otherName, otherDepth, otherTime = self.events[j]
+                    if otherType == type and otherName == name:
+                        i += 1
+                    else:
+                        break
+                # Print number of occurances if multiple iterations followed each other
+                if start == i:
+                    print "\t" * depth + "🔂 %s"%name
+                else:
+                    print "\t" * depth + "🔂 %s x %d"%(name, i - start + 1)
+            elif type == "start":
+                print "\t" * depth + "▶️ %s"%name
+            elif type == "stop":
+                print "\t" * depth + "⏹ %s (%f s)"%(name, timeTaken)
+            i += 1
+
+    @classmethod
+    def max(self, benchmarks):
+        benchmark = Benchmark()
+
+        totalTime = {}
+        iterations = {}
+
+        for b in benchmarks:
+            for key, value in b.totalTime.items():
+                if key not in totalTime:
+                    totalTime[key] = []
+                totalTime[key].append(value)
+            for key, value in b.iterations.items():
+                if key not in iterations:
+                    iterations[key] = []
+                iterations[key].append(value)
+
+        for key, values in totalTime.items():
+            benchmark.totalTime[key] = max(values)
+        for key, values in iterations.items():
+            benchmark.iterations[key] = max(values)
+
+        return benchmark
+
+    @classmethod
+    def min(self, benchmarks):
+        benchmark = Benchmark()
+
+        totalTime = {}
+        iterations = {}
+
+        for b in benchmarks:
+            for key, value in b.totalTime.items():
+                if key not in totalTime:
+                    totalTime[key] = []
+                totalTime[key].append(value)
+            for key, value in b.iterations.items():
+                if key not in iterations:
+                    iterations[key] = []
+                iterations[key].append(value)
+
+        for key, values in totalTime.items():
+            benchmark.totalTime[key] = min(values)
+        for key, values in iterations.items():
+            benchmark.iterations[key] = min(values)
+
+        return benchmark
+
+    @classmethod
+    def average(self, benchmarks):
+        benchmark = Benchmark()
+
+        totalTime = {}
+        iterations = {}
+
+        average = lambda x : sum(x) / len(x)
+
+        for b in benchmarks:
+            for key, value in b.totalTime.items():
+                if key not in totalTime:
+                    totalTime[key] = []
+                totalTime[key].append(value)
+            for key, value in b.iterations.items():
+                if key not in iterations:
+                    iterations[key] = []
+                iterations[key].append(value)
+
+        for key, values in totalTime.items():
+            benchmark.totalTime[key] = average(values)
+        for key, values in iterations.items():
+            benchmark.iterations[key] = average(values)
+
+        return benchmark

benchmark.py

@@ -0,0 +1,36 @@
+#!/usr/bin/env bash
+
+# This scripts take all python source code files and
+# format them to be syntax highlit and includable in LaTex.
+# All output files will be in ./appendices.
+
+# Create the (temporary) directory ./appendices if needed
+mkdir -p appendices
+
+# Create a file to be included in LaTex
+echo "% This file should be included in the article" > appendices/appendices.tex
+
+# For each python file
+ls | grep ".py$" | while read -r filename ; do
+    # Get LaTeX file name
+    safeName=$(echo "$filename" | sed -e "s/_/\\\_/g")
+    out=$(echo "$filename" | sed -e "s/.py/.tex/g")
+    # Write LaTeX heading
+    echo "\\chapter{$safeName}" > "appendices/$out"
+    echo "\\begin{minted}[frame=lines, framesep=2mm, baselinestretch=1.2, bgcolor=white, fontsize=\\footnotesize, linenos]{python}" >> "appendices/$out"
+    # Append file content
+    cat $filename >> "appendices/$out"
+    echo "" >> "appendices/$out"
+    # Append LaTeX footer
+    echo "\end{minted}" >> "appendices/$out"
+    # Append filename to included in appendices
+    echo "\\input{appendices/$out}" >> appendices/appendices.tex
+done
+
+
+
+# Zip the appendices
+zip -r appendices.zip appendices
+
+# Remove the temporary folder
+rm -r appendices