diff --git a/eagle-mpctool.tar.gz b/eagle-mpctool.tar.gz
index d9b15d68d1471a93c8689e9e0fb01798b189d962..a4a9f2b00f5c61e6d360b35d6558c19a11b3df76 100644
Binary files a/eagle-mpctool.tar.gz and b/eagle-mpctool.tar.gz differ
diff --git a/eagle.spec b/eagle.spec
index 37a2da31ec7b4616a67d9247ed25c4c9a4c22a8c..f9b0389423bd2792a96a8c8653b3ee2e3ad9c708 100644
--- a/eagle.spec
+++ b/eagle.spec
@@ -1,6 +1,6 @@
 Name:             eagle
-Version:          1.0.1
-Release:          2
+Version:          1.0.2
+Release:          1
 Summary:          eagle kit
 License:          Mulan PSL v2
 URL:              https://gitee.com/openeuler/eagle
@@ -36,6 +36,14 @@ Requires:           python3-numpy python3-pandas python3-psutil python3-scipy ip
 %{python3_sitelib}/*
 
 %changelog
+* Thur Nov 30 2023 queyanwen<queyanwen@huawei.com> - 1.0.2-1
+- The current condition for determining whether the system enters the stable state is that
+- the difference between the maximum temperature and the minimum temperature within 60s is less than 0.2 degrees.
+- This judgment is too strict and needs to be changed to 2 degrees.
+- In addition, the average fuzzy processing in five windows is used to prevent data jitter.
+- When the entire system can obtain the CPU power consumption data, the obtained CPU power consumption
+- is directly used during training, instead of obtaining the CPU power consumption by inferring.
+
 * Mon Jun 12 2023 queyanwen<queyanwen@huawei.com> - 1.0.1-2
 - Optimize the accuracy of non-steady-state data prediction and training.
 - The anbient temperature check is added befor the collection is started. If the temperature is