From 509f822680fd7ce90082d7980e931df603f01a82 Mon Sep 17 00:00:00 2001
From: Jing Lin <82669431+linjing-lab@users.noreply.github.com>
Date: Sun, 29 Oct 2023 23:36:05 +0800
Subject: [PATCH] fulfill README

---
 released_box/README.md | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/released_box/README.md b/released_box/README.md
index 14ac206..947dbf8 100644
--- a/released_box/README.md
+++ b/released_box/README.md
@@ -9,7 +9,7 @@ perming: Perceptron Models Are Training on Windows Platform with Default GPU Acc
 
 ## init backend
  
-refer to https://pytorch.org/get-started/locally/ and choose PyTorch that support `cuda` compatible with your Windows.
+refer to https://pytorch.org/get-started/locally/ and choose PyTorch to support `cuda` compatible with your Windows.
 
 tests with: PyTorch 1.7.1+cu101
 
@@ -141,6 +141,14 @@ download stable version:
 ```text
 pip install perming --upgrade
 ```
+download versions without supported *early_stop*:
+```text
+pip install perming==1.3.1
+```
+download versions with supported *early_stop*:
+```text
+pip install perming>=1.4.1
+```
 download version without enhancing *Parallel* and *delayed*:
 ```text
 pip install perming==1.6.1