Merge pull request apache#1149 from apache/dev-postgresql

merge development branch to master branch for V4.2

.gitignore

@@ -29,3 +29,4 @@ test/samples/
 # Sphinx and Doxygen Doc-Site
 doc/_build/*
 doc/en/docs/model_zoo/
+cmake-build-debug/*

CMakeLists.txt

@@ -29,10 +29,10 @@ LIST(APPEND CMAKE_MODULE_PATH ${PROJECT_SOURCE_DIR}/cmake/Thirdparty)
 #string(REGEX REPLACE "^[0-9]+\\.[0-9]+\\.([0-9]+).*" "\\1" VERSION_PATCH "${VERSION}")
 
 
-SET(PACKAGE_VERSION 4.1.0) # ${VERSION})
-SET(VERSION 4.1.0)
+SET(PACKAGE_VERSION 4.2.0) # ${VERSION})
+SET(VERSION 4.2.0)
 SET(SINGA_MAJOR_VERSION 4)
-SET(SINGA_MINOR_VERSION 1)
+SET(SINGA_MINOR_VERSION 2)
 SET(SINGA_PATCH_VERSION 0)
 #SET(SINGA_MAJOR_VERSION ${VERSION_MAJOR})  # 0 -
 #SET(SINGA_MINOR_VERSION ${VERSION_MINOR})  # 0 - 9

NOTICE

@@ -30,16 +30,4 @@ developers of Apache SINGA under Apache License, Version 2.0.
 ./doc/_static/images/sgd.png
 ./doc/_static/images/singa.png
 ./doc/_static/images/singav1-sw.png
-./examples/model_selection/TRAILS-Database-Native-Model-Selection/documents/image-20231020174425377.png
-./examples/model_selection/TRAILS-Database-Native-Model-Selection/documents/image-20231020174945226.png
-./examples/model_selection/TRAILS-Database-Native-Model-Selection/internal/ml/model_selection/documents/imgs/image-20230421214835152.png
-./examples/model_selection/TRAILS-Database-Native-Model-Selection/internal/ml/model_selection/documents/imgs/image-20230421220338391.png
-./examples/model_selection/TRAILS-Database-Native-Model-Selection/internal/ml/model_selection/documents/imgs/image-20230421220443231.png
-./examples/model_selection/TRAILS-Database-Native-Model-Selection/internal/ml/model_selection/documents/imgs/image-20230702035554579.png
-./examples/model_selection/TRAILS-Database-Native-Model-Selection/internal/ml/model_selection/documents/imgs/image-20230702035622198.png
-./examples/model_selection/TRAILS-Database-Native-Model-Selection/internal/ml/model_selection/documents/imgs/image-20230702035639502.png
-./examples/model_selection/TRAILS-Database-Native-Model-Selection/internal/ml/model_selection/documents/imgs/image-20230702035806963.png
-./examples/model_selection/TRAILS-Database-Native-Model-Selection/internal/ml/model_selection/documents/imgs/image-20230722202555763.png
-./examples/model_selection/TRAILS-Database-Native-Model-Selection/internal/ml/model_selection/documents/imgs/image-20230722205244718.png
-./examples/model_selection/TRAILS-Database-Native-Model-Selection/internal/ml/model_selection/documents/imgs/image-20230724111325368.png
-./examples/model_selection/TRAILS-Database-Native-Model-Selection/internal/ml/model_selection/documents/imgs/image-20230724111659545.png
+./examples/model_selection/Trails/documents/ai_db.001.jpeg

RELEASE_NOTES

@@ -1,3 +1,33 @@
+Release Notes - SINGA - Version singa-4.2.0
+
+SINGA is a distributed deep learning library.
+
+This release includes following changes:
+
+  * Add support for deep learning models running on top of PolarDB
+    * Implement efficient model selection for a given dataset stored in the database.
+    * Add support for dynamic model creation.
+    * Add support for flexible setting of model training configurations.
+    * Optimize the in-database analytics modules for scalability, efficiency and memory consumption.
+
+  * New example
+    * Add a horizontal federated learning example using the Bank dataset.
+
+  * Enhance examples
+    * Add sample training data for testing the model selection application.
+
+  * Update the website
+    * Update the star button in the main page.
+    * Refine the display of star statistics.
+
+  * Update the python versions for wheel files
+
+  * Fix bugs
+    * Fix the rat check files.
+    * Update the license files.
+
+----------------------------------------------------------------------------------------------
+
 Release Notes - SINGA - Version singa-4.1.0
 
 SINGA is a distributed deep learning library.

examples/hfl/README.md

@@ -0,0 +1,63 @@
+<!--
+    Licensed to the Apache Software Foundation (ASF) under one
+    or more contributor license agreements.  See the NOTICE file
+    distributed with < this work for additional information
+    regarding copyright ownership.  The ASF licenses this file
+    to you under the Apache License, Version 2.0 (the
+    "License"); you may not use this file except in compliance
+    with the License.  You may obtain a copy of the License at
+
+      http://www.apache.org/licenses/LICENSE-2.0
+
+    Unless required by applicable law or agreed to in writing,
+    software distributed under the License is distributed on an
+    "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+    KIND, either express or implied.  See the License for the
+    specific language governing permissions and limitations
+    under the License.
+-->
+
+
+# Horizontal Federated Learning Example
+
+This is an example of federated learning (FL) using the Singa framework. In FL, there is a server and a set of clients. Each client has a local dataset. 
+In each iteration, each client trains the model using its local dataset and uploads the model gradient to the server, which aggregates to get the global
+gradient using the Federated Average algorithm. The server sends the global gradient to all clients for iterative model training. 
+This example uses the Bank dataset and an MLP model in FL.
+
+## Preparation
+
+Go to the Conda environment that contains the Singa library, and run
+
+```bash
+pip install -r requirements.txt
+```
+
+Download the bank dataset and split it into 3 partitions.
+
+```bash
+# 1. download the data from https://archive.ics.uci.edu/ml/datasets/bank+marketing
+# 2. put it under the /data folder
+# 3. run the following command which:
+#    (1) splits the dataset into N subsets
+#    (2) splits each subsets into train set and test set (8:2)
+python -m bank N
+```
+
+## Run the example
+
+Run the server first (set the number of epochs to 3)
+
+```bash
+python -m src.server -m 3 --num_clients 3
+```
+
+Then, start 3 clients in different terminal
+
+```bash
+python -m src.client --model mlp --data bank -m 3 -i 0 -d non-iid
+python -m src.client --model mlp --data bank -m 3 -i 1 -d non-iid
+python -m src.client --model mlp --data bank -m 3 -i 2 -d non-iid
+```
+
+Finally, the server and clients finish the FL training. 

examples/hfl/config/.gitignore

@@ -0,0 +1,8 @@
+# Default ignored files
+/shelf/
+/workspace.xml
+# Datasource local storage ignored files
+/dataSources/
+/dataSources.local.xml
+# Editor-based HTTP Client requests
+/httpRequests/

examples/hfl/config/Singa-HFL.iml

@@ -0,0 +1,29 @@
+<!--
+    Licensed to the Apache Software Foundation (ASF) under one
+    or more contributor license agreements.  See the NOTICE file
+    distributed with < this work for additional information
+    regarding copyright ownership.  The ASF licenses this file
+    to you under the Apache License, Version 2.0 (the
+    "License"); you may not use this file except in compliance
+    with the License.  You may obtain a copy of the License at
+
+      http://www.apache.org/licenses/LICENSE-2.0
+
+    Unless required by applicable law or agreed to in writing,
+    software distributed under the License is distributed on an
+    "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+    KIND, either express or implied.  See the License for the
+    specific language governing permissions and limitations
+    under the License.
+-->
+
+<?xml version="1.0" encoding="UTF-8"?>
+<module type="PYTHON_MODULE" version="4">
+  <component name="NewModuleRootManager">
+    <content url="file://$MODULE_DIR$">
+      <excludeFolder url="file://$MODULE_DIR$/venv" />
+    </content>
+    <orderEntry type="inheritedJdk" />
+    <orderEntry type="sourceFolder" forTests="false" />
+  </component>
+</module>

examples/hfl/requirements.txt

@@ -0,0 +1,3 @@
+pandas
+scikit-learn
+protobuf

examples/hfl/src/__init__.py

@@ -0,0 +1,19 @@
+#
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements.  See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership.  The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License.  You may obtain a copy of the License at
+#
+#   http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+# KIND, either express or implied.  See the License for the
+# specific language governing permissions and limitations
+# under the License.
+#
+

examples/hfl/src/bank.py

@@ -0,0 +1,97 @@
+#
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements.  See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership.  The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License.  You may obtain a copy of the License at
+#
+#   http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+# KIND, either express or implied.  See the License for the
+# specific language governing permissions and limitations
+# under the License.
+#
+
+# https://github.com/zhengzangw/Fed-SINGA/blob/main/src/client/data/bank.py
+
+import pandas as pd
+import numpy as np
+import sys
+from pandas.api.types import is_numeric_dtype
+from sklearn.model_selection import train_test_split
+from sklearn.utils import shuffle
+
+
+def encode(df):
+    res = pd.DataFrame()
+    for col in df.columns.values:
+        if not is_numeric_dtype(df[col]):
+            tmp = pd.get_dummies(df[col], prefix=col)
+        else:
+            tmp = df[col]
+        res = pd.concat([res, tmp], axis=1)
+    return res
+
+
+def load(device_id):
+    fn_train = "data/bank_train_" + str(device_id) + ".csv"
+    fn_test = "data/bank_test_" + str(device_id) + ".csv"
+
+    train = pd.read_csv(fn_train, sep=',')
+    test = pd.read_csv(fn_test, sep=',')
+
+    train_x = train.drop(['y'], axis=1)
+    train_y = train['y']
+    val_x = test.drop(['y'], axis=1)
+    val_y = test['y']
+
+    train_x = np.array((train_x), dtype=np.float32)
+    val_x = np.array((val_x), dtype=np.float32)
+    train_y = np.array((train_y), dtype=np.int32)
+    val_y = np.array((val_y), dtype=np.int32)
+
+    train_x, val_x = normalize(train_x, val_x)
+    num_classes = 2
+
+    return train_x, train_y, val_x, val_y, num_classes
+
+
+def normalize(X_train, X_test):
+    from sklearn.preprocessing import MinMaxScaler
+    scaler = MinMaxScaler()
+    X_train_scaled = scaler.fit_transform(X_train)
+    X_test_scaled = scaler.transform(X_test)
+    return X_train_scaled, X_test_scaled
+
+
+def split(num):
+    filepath = "../data/bank-additional-full.csv"
+    df = pd.read_csv(filepath, sep=';')
+    df['y'] = (df['y'] == 'yes').astype(int)
+    data = encode(df)
+    data = shuffle(data)
+    train, test = train_test_split(data, test_size=0.2)
+
+    train.to_csv("data/bank_train_.csv", index=False)
+    test.to_csv("data/bank_test_.csv", index=False)
+
+    train_per_client = len(train) // num
+    test_per_client = len(test) // num
+
+    print("train_per_client:", train_per_client)
+    print("test_per_client:", test_per_client)
+    for i in range(num):
+        sub_train = train[i * train_per_client:(i + 1) * train_per_client]
+        sub_test = test[i * test_per_client:(i + 1) * test_per_client]
+        sub_train.to_csv("data/bank_train_" + str(i) + ".csv", index=False)
+        sub_test.to_csv("data/bank_test_" + str(i) + ".csv", index=False)
+
+
+if __name__ == "__main__":
+    split(int(sys.argv[1]))
+