From 40eaa2143e40b5f30f9045417b73202fc310c452 Mon Sep 17 00:00:00 2001
From: Ananya Harsh Jha <ananya@pytorchlightning.ai>
Date: Thu, 27 Aug 2020 14:25:39 -0400
Subject: [PATCH] add desc for minimize (#3216)

---
 docs/source/results.rst | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/docs/source/results.rst b/docs/source/results.rst
index a09c3ad544..68f6fdcdb6 100644
--- a/docs/source/results.rst
+++ b/docs/source/results.rst
@@ -80,6 +80,10 @@ The `TrainResult` basic usage is this:
 
 minimize
 ^^^^^^^^
+When using TrainResult, the metric that needs to be minimized is passed to this
+argument. Internally, the tensor is verified to contain gradients and `.backward()`
+is called on it.
+
 .. code-block:: python
 
     def training_step(...):