Scores to probabilities

Logistic regression turns a score into a probability with the sigmoid. Convert some scores.

What to build

A program that prints the sigmoid of each score and the class it predicts.

Requirements

  • For each score print Z -> P (C), with Z written as in the input, P with 4 decimal places and C the class, 1 or 0.
  • The class is 1 when the probability is at least 0.5.

Input

One line of scores.

Output

One line per score.

Constraints

1 to 100 scores between −100 and 100.

Example

Input:

0 2 -2

Output:

0 -> 0.5000 (1)
2 -> 0.8808 (1)
-2 -> 0.1192 (0)

Why it works

The sigmoid is 0.5 at z = 0 and rises smoothly towards 1 for positive scores and falls towards 0 for negative ones, so the sign of the score decides the class.

Examples

Example 1 · Input
0 2 -2
Expected output
0 -> 0.5000 (1)
2 -> 0.8808 (1)
-2 -> 0.1192 (0)
Example 2 · Input
5
Expected output
5 -> 0.9933 (1)

Prompts inside input() are ignored when checking, so input("n: ") is fine.

Hints

Hint 1

σ(z) = 1 / (1 + e^(−z)); use `math.exp`.

Hint 2

The class is 1 when σ(z) ≥ 0.5, which is the same as z ≥ 0.

Solution

Show one correct solution (try it yourself first)
import math

for token in input().split():
    p = 1 / (1 + math.exp(-float(token)))
    print(f"{token} -> {p:.4f} ({1 if p >= 0.5 else 0})")

Lesson: Logistic regression →