Skip to content

Define Metrics

Edward Junprung edited this page Sep 20, 2022 · 3 revisions

Metrics are the building blocks for the reward function. Metrics can embody important simulation metrics such as revenue and cost. These metrics are likely what you directly seek to optimize.

Step 1

Open the Pathmind Helper properties and locate the Metrics field.

metrics

Step 2

Input your metrics.

class Reward {
	double reward_0 = rew0;
	double reward_1 = rew1;
	double reward_2 = rew2;
}

This should output an array that looks something like: [0.0, 0.0, 0.0]. The values 0.0 are just placeholders; in actuality, they will contain varied values from your rewards.

Finally, you can audit your metrics by following the instructions here.

Why do we need metrics?

Metrics alone serve no purpose. These metrics will become the building blocks for your reward function.