0Pricing
Deep Learning Academy · Lesson

Binary Cross-Entropy with Logits

The stable loss for two-class problems.

Two Classes, One Decision

When your model answers a yes or no question like spam or not spam, you reach for binary cross-entropy, the loss built for two-class problems. ✅

From Score to Probability

A model first outputs a raw score called a logit. The sigmoid function squashes that logit into a probability between 0 and 1.

import torch
prob = torch.sigmoid(logit)

All lessons in this course

  1. MSE & MAE for Regression
  2. Binary Cross-Entropy with Logits
  3. Cross-Entropy for Multiclass
  4. Class Weights for Imbalanced Data
← Back to Deep Learning Academy