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 2022/05/07 02:43:48 UTC
[singa] branch dev updated: add release_notes for v3.3.0
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 6b1d0eeb add release_notes for v3.3.0
new 946f5df4 Merge pull request #962 from NLGithubWP/dev2
6b1d0eeb is described below
commit 6b1d0eebc9559eaf102a2a274ada453ab10ee77f
Author: NLGithubWP <xi...@gmail.com>
AuthorDate: Fri May 6 21:53:20 2022 +0800
add release_notes for v3.3.0
---
RELEASE_NOTES | 35 +++++++++++++++++++++++++++++++++++
1 file changed, 35 insertions(+)
diff --git a/RELEASE_NOTES b/RELEASE_NOTES
index 6a3cebf9..d426ac2b 100644
--- a/RELEASE_NOTES
+++ b/RELEASE_NOTES
@@ -1,3 +1,38 @@
+Release Notes - SINGA - Version singa-3.3.0
+
+SINGA is a distributed deep learning library.
+
+This release includes following changes:
+
+ * New examples:
+ * Add one CNN example for the BloodMnist dataset, a sub set of MedMNIST.
+ * Add one example for the medical image analysis.
+
+ * Enhance distributed training
+ * Add key information printing, e.g., throughput and communication time, for distributed training.
+ * Optimize printing and logging modules for faster distributed training.
+
+ * Enhance example code
+ * Add more datasets and model implementations for the cifar_distributed_cnn example.
+ * Update the running script for the cifar_distributed_cnn example to include more models.
+ * Update the dataset path for the largedataset_cnn example for more flexibility.
+ * Add more model implementations for the largedataset_cnn example.
+
+ * Enhance the webpage
+ * Reconstruct the singa webpage to include project features.
+ * Update the Git web site by deploying it via .asf.yaml.
+ * Update the Chinese and Vietnamese documentations.
+
+ * Debug and add assertions for input tensor data types in the opt.py.
+
+ * Change pointer type to void for generalizing data types.
+
+ * Fix bugs
+ * Fix the python test error due to operations not implemented for some data types.
+ * Fix the model of pad from bytes to str.
+
+----------------------------------------------------------------------------------------------
+
Release Notes - SINGA - Version singa-3.2.0
SINGA is a distributed deep learning library.