You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@singa.apache.org by zh...@apache.org on 2023/03/19 13:25:51 UTC

[singa] branch dev updated: Release Notes for v4.0.0

This is an automated email from the ASF dual-hosted git repository.

zhaojing pushed a commit to branch dev
in repository https://gitbox.apache.org/repos/asf/singa.git


The following commit(s) were added to refs/heads/dev by this push:
     new df63ca20 Release Notes for v4.0.0
     new ea7e1ea0 Merge pull request #1047 from lzjpaul/23-3-18-zj-release-note
df63ca20 is described below

commit df63ca2084a5225248e2667dab39d5fec025c128
Author: zhaojing <zh...@comp.nus.edu.sg>
AuthorDate: Sat Mar 18 22:52:52 2023 +0800

    Release Notes for v4.0.0
---
 RELEASE_NOTES | 37 +++++++++++++++++++++++++++++++++++++
 1 file changed, 37 insertions(+)

diff --git a/RELEASE_NOTES b/RELEASE_NOTES
index d426ac2b..bbe67726 100644
--- a/RELEASE_NOTES
+++ b/RELEASE_NOTES
@@ -1,3 +1,40 @@
+Release Notes - SINGA - Version singa-4.0.0
+
+SINGA is a distributed deep learning library.
+
+This release includes following changes:
+
+  * Enhance distributed training
+    * Add support for configuration of number of GPUs to be used.
+    * Increase max epoch for better convergence.
+    * Print intermediate mini-batch information.
+    * Add support for switching between CPU and GPU devices.
+
+  * Enhance example code
+    * Update the args of normalize forward function in the transforms of the BloodMnist example.
+    * Update the xceptionnet in the cnn example.
+    * Add arguments for weight decay, momentum and learning rates in the cnn example.
+    * Add training scripts for more datasets and model types in the cnn example.
+    * Add resnet dist version for the large dataset cnn example.
+    * Add cifar 10 multi process for the large dataset cnn example.
+    * Add sparsification implementation for mnist in the large dataset cnn example.
+    * Update the cifar datasets downloading to local directories.
+    * Extend the cifar datasets load function for customized directorires.
+
+  * Enhance the webpage
+    * Update online documentation for distributed training.
+
+  * Promote code quality
+    * Update inline comments for prepreocessing and data loading
+
+  * Update the PIL image module
+
+  * Update the runtime Dockerfile
+
+  * Update the conda files
+
+----------------------------------------------------------------------------------------------
+
 Release Notes - SINGA - Version singa-3.3.0
 
 SINGA is a distributed deep learning library.