2.11 Threshold

R figure

Code

ranking <- c(1, 1, 1, 0, 1, 0, 0, 1, 0, 0)
x <- c(0, 0, 0, 300, 300, 300, 600, 600, 600, 900, 1200)
y <- c(0, 200, 400, 400, 600, 800, 800, 1000, 1200, 1200, 1200)

h <- 1200
w <- 1200 
grid.step <- 120

plot(
  c(0, w), c(0, h),
  xaxs = "i", yaxs = "i", 
  xaxt = 'n', yaxt = 'n', 
  type = "n",             
  xlab = "False positive rate", 
  ylab = "True positive rate"
)

axis(1, at = x, labels = c('', '', '', '0.25', '', '', '0.50', '', '', '0.75', '1.00'))
axis(2, at = y, labels = c('', '0.17', '0.33', '', '0.50', '0.67', '', '0.83', '1.00', '', ''))

gx <- grid.step
while (gx <= w) {
  abline(v = gx, col = "gray", lty = "dotted")
  gx <- gx + grid.step
}

gy <- grid.step
while (gy <= h) {
  abline(h = gy, col = "gray", lty = "dotted")
  gy <- gy + grid.step
}

lines(x, y, lty = 1, type = 'o')

segments(x[3], y[3], x[9], y[9], lty = 3, col = "red")
segments(x[1], y[6], x[9], y[9], lty = 3, col = "red")

abline(a = 1200, b = -h / w, lty = 3, col = "blue")

segments(0, 960, 240, 960, lty = 3, col = "blue")
segments(0, 12000 / 14, 4800 / 14, 12000 / 14, lty = 3, col = "blue")

Python figure

Code

import matplotlib.pyplot as plt
import numpy as np

ranking = [1, 1, 1, 0, 1, 0, 0, 1, 0, 0]
x = [0, 0, 0, 300, 300, 300, 600, 600, 600, 900, 1200]
y = [0, 200, 400, 400, 600, 800, 800, 1000, 1200, 1200, 1200]

h = 1200
w = 1200
grid_step = 120

fig, ax = plt.subplots()
ax.set_xlim(0, w)

(0.0, 1200.0)

Code

ax.set_ylim(0, h)

(0.0, 1200.0)

Code

ax.set_xticks(x)
ax.set_xticklabels(['', '', '', '0.25', '', '', '0.50', '', '', '0.75', '1.00'])
ax.set_yticks(y)
ax.set_yticklabels(['', '0.17', '0.33', '', '0.50', '0.67', '', '0.83', '1.00', '', ''])

gx = grid_step
while gx <= w:
    ax.axvline(x=gx, color='gray', linestyle='dotted')
    gx += grid_step

gy = grid_step
while gy <= h:
    ax.axhline(y=gy, color='gray', linestyle='dotted')
    gy += grid_step

ax.plot(x, y, linestyle='solid', marker='o')

ax.plot([x[2], x[8]], [y[2], y[8]], linestyle='dashed', color='red')
ax.plot([x[0], x[5]], [y[5], y[8]], linestyle='dashed', color='red')

slope = -h / w
intercept = 1200
x_vals = np.array([0, w])
y_vals = intercept + slope * x_vals
ax.plot(x_vals, y_vals, linestyle='dashed', color='blue')

ax.plot([0, 240], [960, 960], linestyle='dashed', color='blue')
ax.plot([0, 4800 / 14], [12000 / 14, 12000 / 14], linestyle='dashed', color='blue')

ax.set_xlabel("False positive rate")
ax.set_ylabel("True positive rate")
plt.show()