Precision, Recall & F1 From a Confusion Matrixeasy

Precision, Recall & F1 From a Confusion Matrix

easy Classical ML from Scratch 10–15 min

NumPy allowed · plain Python earns 10 bonus points

A vector transforms

Matrix multiplication is the first computation behind every model.

Open lesson →

01vector

02matrix

03transform

04result

[2, −1] × [[1, 2], [−.5, 1]] = [0, −2]

Computed matrix × vector transformation

Implement the three classification metrics that matter far more than raw accuracy whenever classes are imbalanced -- the numbers behind every "high accuracy, useless model" cautionary tale in ML.

Related Skills

Model Evaluation & Metrics

Task

Implement precision_recall_f1(tp, fp, fn), returning a dict with keys 'precision', 'recall', and 'f1', computed from the standard formulas with explicit zero-division guards.

Function Signature

precision_recall_f1(tp: int, fp: int, fn: int) -> dict

Examples

Example 1: Typical Confusion Matrix
Input: tp = 50
fp = 10
fn = 5
Output: {'precision': 0.8333333333333334, 'recall': 0.9090909090909091, 'f1': 0.8695652173913043}

Example 2: No Positive Predictions At All
Input: tp = 0
fp = 0
fn = 5
Output: {'precision': 0, 'recall': 0, 'f1': 0}
Explanation: tp+fp == 0 -- precision is defined as 0.0, not undefined

Constraints

Libraries (standard library only) are allowed and accepted normally.
A plain-Python solution earns 10 bonus points.
precision = tp / (tp + fp), or 0.0 if tp + fp == 0.
recall = tp / (tp + fn), or 0.0 if tp + fn == 0.
f1 = 2 * precision * recall / (precision + recall), or 0.0 if precision + recall == 0.

Python3Saved ✓

def precision_recall_f1(tp: int, fp: int, fn: int) -> dict:

"""tp, fp, fn: true positive / false positive / false negative counts

for the positive class. Return {'precision': ..., 'recall': ...,

'f1': ...}. Define precision = 0.0 if tp+fp == 0, recall = 0.0 if

tp+fn == 0, and f1 = 0.0 if precision+recall == 0, instead of

dividing by zero."""

# Your implementation here

pass

Case 1: Typical Confusion Matrix

Input: tp = 50
fp = 10
fn = 5

Expected: {'precision': 0.8333333333333334, 'recall': 0.9090909090909091, 'f1': 0.8695652173913043}

Case 2: No Positive Predictions At All

Input: tp = 0
fp = 0
fn = 5

Expected: {'precision': 0, 'recall': 0, 'f1': 0}