From f18c3699249f94073543d9737890724a2053838a Mon Sep 17 00:00:00 2001
From: madhur-ob <155637867+madhur-ob@users.noreply.github.com>
Date: Sun, 7 Jan 2024 01:56:51 +0530
Subject: [PATCH] non default channels for conda packages (#111)

* tip on using non-default conda channel

* add a newline
---
 docs/scaling/dependencies/libraries.md | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/docs/scaling/dependencies/libraries.md b/docs/scaling/dependencies/libraries.md
index f53d0603..62f6c2dd 100644
--- a/docs/scaling/dependencies/libraries.md
+++ b/docs/scaling/dependencies/libraries.md
@@ -177,6 +177,20 @@ $ python scikitflow.py --environment=conda run
 You can search all available public Conda packages at [anaconda.org](https://anaconda.org)
 :::
 
+:::tip
+By default, all conda packages are searched in the default channel i.e. [conda-forge](https://conda-forge.org/feedstock-outputs/). However, some packages may exist in other channels.
+
+To specify a custom channel for a particular package, use the following syntax: `{channel_name}::{package_name}`.
+
+Example: The package `pytorch` exists in the `pytorch` channel. Thus, the dependency will be specified as follows:
+```
+@conda(
+    python='3.9.13',
+    packages={'pytorch::pytorch': '2.1.2'}
+)
+```
+:::
+
 ## Using the same packages in all steps
 
 Sometimes you may want to use the same set of packages in all steps. Repeating the same