Skip to content
Calcrivo

Mish Activation Calculator

Calculate the Mish activation output: x times tanh(softplus(x)).

Inputs

Mish(x)

1.943959

softplus(x)

2.126928

tanh(softplus(x))

0.971979

Step by step

  1. softplus(x) = ln(1 + e^x)

    ln(1 + e^2)

    = 2.126928

  2. tanh(softplus(x))

    tanh(2.126928)

    = 0.971979

  3. Mish(x) = x × tanh(softplus(x))

    2 × 0.971979

    = 1.943959

How it works

Mish is a smooth, self-gated activation function similar in spirit to Swish: Mish(x) = x·tanh(softplus(x)) = x·tanh(ln(1+e^x)). Like Swish, it is smooth everywhere, non-monotonic (dipping slightly below zero for small negative inputs), and unbounded above for positive inputs, but its softplus-then-tanh construction gives it a slightly different curvature. Mish was introduced as an activation for computer vision architectures (notably used in YOLOv4) and has empirically shown modest accuracy improvements over both ReLU and Swish on several benchmark tasks, though at a higher computational cost due to the combination of exponential, logarithm, and tanh operations.

Formula

Mish(x) = x × tanh(ln(1 + e^x))

x
Input value
tanh
Hyperbolic tangent function
ln
Natural logarithm

Frequently Asked Questions

How is Mish different from Swish?

Both are smooth, self-gated activations of the form x × (gating function), but Swish gates with a sigmoid (x × σ(βx)) while Mish gates with tanh(softplus(x)) — the softplus-tanh combination gives Mish a slightly smoother, more gradual transition near zero and marginally different curvature for negative inputs.

Is Mish more expensive to compute than ReLU or Swish?

Yes — computing Mish requires an exponential, a logarithm (for softplus), and a tanh, making it noticeably more expensive per-element than ReLU's single comparison or even Swish's single sigmoid, which is a practical tradeoff against its reported accuracy benefits.

Where has Mish been used successfully?

Mish gained popularity in computer vision, most notably as the activation function used in YOLOv4's backbone network, where it contributed to measurable accuracy improvements over the ReLU-family activations used in earlier YOLO versions.

You might also need