plot.py

#!/usr/bin/env python3

import matplotlib.pyplot as plt
from matplotlib.cm import get_cmap
import matplotlib
import statistics
from ast import literal_eval as make_tuple
from scipy import interpolate
import numpy as np

plt.plot([], [], ' ', label='Dota 2 xiph.org footage\n 1920*1080 Frames: 180, 6 different scenes')

# AV1
plt.plot([], [], ' ', label='AOM 10.05.2020 CQ range: cq 30-60, step 5')
with open('aom_dotatest.mkvdata.txt', 'r') as f:
        raw = f.read().split('),(')
        raw = [x for x in raw]
        d = raw[1:-1] + [raw[0][1:], raw[-1][:-2]]

data = []

for x in d:
    try:
        data.append(make_tuple('(' + x + ')'))
    except:
        print(x)


cmap = plt.get_cmap('Purples')
colors = list(cmap(np.linspace(0.4,0.8, len(set([x[2] for x in data])))))

cpus = sorted(set([x[2] for x in data]))

for cpu in cpus:
    dt = [x for x in data if x[2] == cpu]
    x = sorted([x[4] for x in dt])
    y = sorted([x[5] for x in dt])
    time = sum([x[1] for x in dt])/len(dt)
    f = interpolate.interp1d(x, y, kind='cubic')
    xnew = np.linspace(min(x), max(x), max(x) - min(x) )
    plt.plot(xnew, f(xnew), label=f'AOM {cpu}, Average Time: {round(time)}', linewidth=3, c=colors.pop())




# x264
plt.plot([], [], ' ', label='FFmpeg 14.05.2020 x264: CRF range: 35 - 17, step 3')
with open('x264_dotatest.mkvdata.txt') as f:
    raw = f.read().split('),(')
    raw = [x for x in raw]
    d = raw[1:-1] + [raw[0][1:], raw[-1][:-2]]

x264 = []

for x in d:
    try:
        x264.append(make_tuple('(' + x + ')'))
    except:
        print(x)

speeds = (('ultrafast', 0), ('superfast', 1), ('veryfast', 2), ('faster', 3), ('fast', 4), ('medium', 5), ('slow', 6), ('slower', 7), ('veryslow', 8), ('placebo',9))


in_file = set([x[2] for x in x264])
speeds = reversed([x for x in speeds if x[0] in in_file])

cmap = plt.get_cmap('Reds')
colors2 = list(cmap(np.linspace(0.4,0.9, len(in_file))))

for speed, _ in speeds:
    ks = [x for x in x264 if x[2] == speed]
    x = sorted([x[4] for x in ks])
    y = sorted(x[5] for x in ks)
    time = sum(x[1] for x in ks)/len(x)
    f = interpolate.interp1d(x, y, kind='cubic')
    xnew = np.linspace(min(x), max(x), max(x) - min(x) )
    plt.plot(xnew, f(xnew), label=f'x264 {speed}, Average Time: {round(time)}',linestyle='-.', linewidth=3, c=colors2.pop())

# XVID
plt.plot([], [], ' ', label='XVID 25-3, step 5')
with open('xvid_dotatest.mkvdata.txt', 'r') as f:
        raw = f.read().split('),(')
        raw = [x for x in raw]
        d = raw[1:-1] + [raw[0][1:], raw[-1][:-2]]

data = []

for x in d:
    try:
        data.append(make_tuple('(' + x + ')'))
    except:
        print(x)


cmap = plt.get_cmap('Greens')
colors = list(cmap(np.linspace(0.4,0.8, len(set([x[2] for x in data])))))

cpus = sorted(set([x[2] for x in data]))

for cpu in cpus:
    dt = [x for x in data if x[2] == cpu]
    x = sorted([x[4] for x in dt])
    y = sorted([x[5] for x in dt])
    time = sum([x[1] for x in dt])/len(dt)
    f = interpolate.interp1d(x, y, kind='cubic')
    xnew = np.linspace(min(x), max(x), max(x) - min(x) )
    plt.plot(xnew, f(xnew), label=f'XVID, Average Time: {round(time)}', linewidth=3, c=colors.pop())




# Plot

plt.xticks([x for x in range(0, 20000, 500)], [float(x)/1000 for x in range(0, 20000, 500) ],fontsize=26)
plt.yticks([x for x in range(30, 101, 1)],fontsize=28)
[plt.axvline(i, color='grey', linewidth=0.3) for i in range(0, 40000, 500)]
[plt.axhline(i, color='grey', linewidth=0.5) for i in range(21, 100, 2)]
[plt.axhline(i, color='black', linewidth=1) for i in range(22, 100, 2)]
plt.ylabel('Vmaf', size=30)
plt.xlabel('Bitrate Mbps', size=30)
plt.title('AOM vs x264 vs XVID', size=30)
plt.legend(prop={'size': 24}, loc="lower right")
plt.xlim(1000,15000)
plt.ylim((80, 100))
plt.tight_layout()
plt.show()