Added the whole project

doall.sh

@@ -0,0 +1,4 @@
+python3.4 generator/generator.py
+python3.4 main.py
+./graphs/graph.sh source.csv source.png Not downsampled
+./graphs/graph.sh sampled.csv sampled.png Downsampled

generator/generator.py

@@ -0,0 +1,32 @@
+import csv
+import random
+
+
+def signal_generator():
+ seed = 1
+ max_y = 250
+ min_y = 100
+ start_x = 0
+ end_x = 7500
+
+ random.seed(seed)
+ prev_y = random.random()*(max_y-min_y)+min_y
+
+ with open('source.csv', 'w') as f:
+ csvf = csv.writer(f, delimiter=',')
+ cur = start_x
+ while cur <= end_x:
+ if random.random() > 0.01:
+ sign = random.choice([-1,1])
+ amp = random.random()*5
+ gen = random.random()*amp*sign+prev_y
+ else:
+ gen = random.random()*(max_y-min_y)+min_y
+ prev_y = gen
+ csvf.writerow([cur, gen])
+ cur += 1
+
+
+if __name__ == '__main__':
+ signal_generator()
+ exit()

graphs/graph.sh

@@ -0,0 +1,61 @@
+#!/usr/bin/env bash
+
+if [ -z $1 ] || ! [ -f $1 ]; then
+ echo "Missing or incorrect input file"
+ exit
+fi
+
+if [ -z $2 ]; then
+ echo "Missing output file"
+ exit
+fi
+
+if [ -z $3 ]; then
+ echo "Missing title"
+ exit
+fi
+
+gnuplot << EOF
+ # reset config
+ reset
+ 
+ # define data delimiter
+ set datafile separator ","
+ 
+ # define terminal
+ set terminal pngcairo size 720,480 enhanced font 'Verdana,10'
+ 
+ # define axis
+ # remove border on top and right and set color to gray
+ set style line 11 lc rgb '#808080' lt 1
+ set border 3 back ls 11
+ set tics nomirror
+ 
+ # define grid
+ set style line 12 lc rgb '#808080' lt 0 lw 1
+ set grid back ls 12
+ 
+ # define legend
+ set key off
+ 
+ # define line styles
+ set style line 1 lt 1 lc rgb '#483D8B' lw 2 # Color: DarkSlateBlue
+ set style line 2 lt 1 lc rgb '#3CB371' lw 2 # Color: MediumSeaGreen
+ set style line 3 lt 1 lc rgb '#CD853F' lw 2 # Color: Peru
+ set style line 4 lt 1 lc rgb '#663399' lw 2 # Color: RebeccaPurple
+ set style line 5 lt 1 lc rgb '#6B8E23' lw 2 # Color: OliveDrab
+ set style line 6 lt 1 lc rgb '#DC143C' lw 2 # Color: Crimson
+ set style line 7 lt 1 lc rgb '#FF4500' lw 2 # Color: OrangeRed
+ 
+ # define ranges
+ set yrange [0:500]
+ 
+ # define output
+ set output "$2"
+ # define labels
+ set title "${@:3}"
+ set xlabel "Time (s)"
+ set ylabel "Y axis"
+ # plotting
+ plot "$1" using 1:2 ls 6 with lines
+EOF

lttb/lttb.py

@@ -0,0 +1,80 @@
+import math
+
+
+def largest_triangle_three_buckets(data, threshold):
+ """
+ Return a downsampled version of data.
+
+ Parameters
+ ----------
+ data: list of lists
+ data must be formated this way: [[x,y], [x,y], [x,y], ...]
+ threshold: int
+ threshold must be >= 2 and <= to the len of data
+
+ Returns
+ -------
+ data, but downsampled using threshold
+ """
+
+ # Check if data and threshold are valid
+ if not isinstance(data, list):
+ raise Exception("incorrect data provided")
+ if not isinstance(threshold, int) or threshold <= 2 or threshold >= len(data):
+ raise Exception("threshold not well defined")
+ for i in data:
+ if not isinstance(i, list) or len(i) != 2:
+ raise Exception("incorrect data provided")
+
+ # Bucket size. Leave room for start and end data points
+ every = (len(data) - 2)/(threshold - 2)
+
+ a = 0 # Initially a is the first point in the triangle
+ next_a = 0
+ max_area_point = (0, 0)
+
+ sampled = [data[0]] # Always add the first point
+
+ for i in range(0, threshold-2):
+ # Calculate point average for next bucket (containing c)
+ avg_x = 0
+ avg_y = 0
+ avg_range_start = math.floor((i+1)*every) + 1
+ avg_range_end = math.floor((i+2)*every) + 1
+ avg_rang_end = avg_range_end if avg_range_end < len(data) else len(data)
+
+ avg_range_length = avg_rang_end - avg_range_start
+
+ while avg_range_start < avg_rang_end:
+ avg_x += data[avg_range_start][0]
+ avg_y += data[avg_range_start][1]
+ avg_range_start += 1
+
+ avg_x /= avg_range_length
+ avg_y /= avg_range_length
+
+ # Get the range for this bucket
+ range_offs = math.floor((i+0)*every) + 1
+ range_to = math.floor((i+1)*every) + 1
+
+ # Point a
+ point_ax = data[a][0]
+ point_ay = data[a][1]
+
+ max_area = -1
+
+ while range_offs < range_to:
+ # Calculate triangle area over three buckets
+ area = math.fabs((point_ax - avg_x)*(data[range_offs][1] - point_ay) - (point_ax - data[range_offs][0])*(avg_y-point_ay))*0.5
+ if area > max_area:
+ max_area = area
+ max_area_point = data[range_offs]
+ next_a = range_offs # Next a is this b
+ range_offs += 1
+
+ sampled.append(max_area_point) # Pick this point from the bucket
+ a = next_a # This a is the next a (chosen b)
+
+ sampled.append(data[len(data)-1]) # Always add last
+
+ return sampled

main.py

@@ -0,0 +1,25 @@
+from lttb.lttb import largest_triangle_three_buckets
+import csv
+
+
+def main():
+ with open('source.csv', 'r') as f:
+ c = 0
+ data = []
+ csvf = csv.reader(f, delimiter=',')
+ for row in csvf:
+ if c == 0:
+ c += 1
+ continue
+ data.append([int(row[0]), float(row[1])])
+ c += 1
+ sampled = largest_triangle_three_buckets(data, 500)
+ with open('sampled.csv', 'w') as f2:
+ csvf2 = csv.writer(f2, delimiter=',')
+ for row in sampled:
+ csvf2.writerow(row)
+
+
+if __name__ == '__main__':
+ main()
+ exit()