Merge pull request #161 from leondavi/onn600

[OPENNN_6.0.0] upgrade opennn to version 6.0.0 - compilation passed

.gitmodules

@@ -7,3 +7,6 @@
 [submodule "simple-cpp-logger"]
 	path = src_cpp/simple-cpp-logger
 	url = https://github.com/nadrino/simple-cpp-logger
+[submodule "nifpp"]
+	path = src_cpp/nifpp
+	url = https://github.com/goertzenator/nifpp.git

CMakeLists.txt

@@ -5,23 +5,30 @@
 # @copyright Copyright (c) 2021 Nerlnet
 # *****************************************************/ 
 
-project(nerlnet DESCRIPTION "Nerlnet Distributed Machine Learning Research Platform")
 cmake_minimum_required(VERSION 3.16.3 FATAL_ERROR)
-
-find_package(OpenMP REQUIRED)
+project(nerlnet DESCRIPTION "Nerlnet Distributed Machine Learning Research Platform")
 
 set(PROJECT_BINARY_DIR build)
 set (CMAKE_CXX_STANDARD 11)
-set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -latomic -fopenmp")#${OpenMP_C_FLAGS}")
-set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -latomic -fopenmp")#${OpenMP_CXX_FLAGS}")
+set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -latomic")
+set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -latomic")
 set(CMAKE_CXX_FLAGS_RELEASE "-O3")
-#set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} ${OpenMP_C_FLAGS}")
-#set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} ${OpenMP_CXX_FLAGS}")
+
+#add_compile_definitions(EIGEN_MAX_ALIGN_BYTES=8) #Open this line for RASPI
+
+option(USE_OpenMP "Use-OpenMP" ON)
+if(USE_OpenMP)
+    find_package(OpenMP)
+    if(OPENMP_FOUND)
+        set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} ${OpenMP_C_FLAGS}")
+        set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} ${OpenMP_CXX_FLAGS}")
+    endif()
+endif()
 
 ###########################
 # adding sub directories 
 ###########################
 add_subdirectory(src_cpp) # actually adding opennnBridge
 
 add_library(${PROJECT_NAME} SHARED $<TARGET_OBJECTS:opennnBridge>)
-target_link_libraries(${PROJECT_NAME} PUBLIC opennn opennnBridge)
+target_link_libraries(${PROJECT_NAME} PUBLIC opennnBridge)

NerlnetBuild.sh

@@ -71,6 +71,16 @@ done
 
 NERLNET_BUILD_PREFIX="[Nerlnet Build] "
 
+OPTION="add_compile_definitions(EIGEN_MAX_ALIGN_BYTES=8)"
+is_rasp="$(grep -c raspbian /etc/os-release)"
+if [ $is_rasp -gt "0" ]; then 
+        echo "$NERLNET_BUILD_PREFIX Detected raspberrypi => setting alignment to 8"
+        sed -i "s/^.*#\(${OPTION}\)/\1/" CMakeLists.txt
+else 
+        echo "$NERLNET_BUILD_PREFIX Using default alignment"
+        sed -i "s/^.*\(${OPTION}.*$\)/#\1/" CMakeLists.txt
+fi
+
 echo "$NERLNET_BUILD_PREFIX Building Nerlnet Library"
 echo "$NERLNET_BUILD_PREFIX Cmake command of Nerlnet NIFPP"
 cmake -S . -B build/release -DCMAKE_BUILD_TYPE=RELEASE

NerlnetGetData.sh

@@ -9,7 +9,7 @@ if [ -d "$INPUT_DATA_DIR" ]; then
 else
         echo "$NERLNET_PREFIX $INPUT_DATA_DIR is generated and is empty!"
         wget https://github.com/halfway258/NerlnetData/archive/refs/heads/main.zip
-        unzip main.zip -d inputDataDir
+        unzip main.zip -d $INPUT_DATA_DIR
         rm main.zip
         echo "$NERLNET_PREFIX Add input data to $INPUT_DATA_DIR"
 fi

README.md

@@ -3,7 +3,7 @@
 
 
 <p align="center">
-  <img src="Nerlnet_logo.jpg" width="350" title="NerlNet">
+  <img src="NerlnetLogo.png" width="200" title="NerlNet">
 </p>
 
 Nerlnet is an open-source library for research of distributed machine learning algorithms on IoT devices that gives full insights into both edge devices that run neural network models and network performance and statistics. Nerlnet can simulate distributed ML architectures and deploy them on various IoT devices. <br><br>
@@ -62,8 +62,9 @@ After running the installation script, NErlNet directory can be accessed via the
 ## Python API and Jupyter (For GUI Server): 
 Minimum Python version: 3.8
 
-1. Create and activate a virtual environment for Nerlnet https://docs.python.org/3/library/venv.html
-2. Install required modules ```pip install -r src_py/requirements.txt```
+1. Create and activate a virtual environment for Nerlnet: (https://docs.python.org/3/library/venv.html)  
+  ```python -m venv VENV_PATH/VENV_NAME  ==== then ===> source VENV/bin/acticvate``` 
+2. Install required modules while in venv ```pip install -r src_py/requirements.txt```
 3. Install Jupyter Notebook ```pip install jupyterlab```
 4. Call Jupyter environment creator script with an experiment directory ```./NerlnetJupyterEnvGenerator.sh --j <experiment_direcotry>```
 5. Run Jupyter notebook and import the ApiServer module to the notebook. 

inputJsonFiles/Architecture/arch_1PC1WorkerHealth.json

@@ -35,25 +35,26 @@
   "workers": [
     {
       "name": "w1",
-      "modelId": "0",
+      "modelId": "1",
       "_comment0": "E_APPROXIMATION = 1, E_CLASSIFICATION = 2, E_FORECASTING = 3 , E_NCODER_DECODER = 4, E_CUSTOMNN = 5, E_AE = 6, E_AEC = 7",
       "modelType": "5",
-      "scalingMethod": "0",
-      "_comment1": "NoScaling = 1 , MinimumMaximum = 2 , MeanStandardDeviation = 3 , StandardDeviation = 4 , Logarithm = 5",
-      "layerTypesList":             "[6,1,1,1,1,1,1,1,1]",
+      "_comment1": "Non-Scaling = 1 , MinimumMaximum = 2 , MeanStandardDeviation = 3 , StandardDeviation = 4 , Logarithm = 5",
+      "scalingMethod": "1",
+      "_commnet_layer_type": "E_LAYER_TYPE_DEFAULT(perceptron) = 0, E_LAYER_TYPE_SCALING = 1, E_LAYER_TYPE_CONVOLUTIONAL = 2 , E_LAYER_TYPE_PERCEPTRON = 3 , E_LAYER_TYPE_POOLING = 4 , E_LAYER_TYPE_PROBABILISTIC = 5 , E_LAYER_TYPE_LSTM = 6 , E_LAYER_TYPE_RECURRENT = 7 , E_LAYER_TYPE_UNSCALING = 8 , E_LAYER_TYPE_BOUNDING = 9 ",
+      "layerTypesList":             "[4,1,1,1,0,0,0]",
       "layersSizes":                "[4,1,1,13,10,6,5]",
-      "_comment2": "Activation functions explain: Threshold = 1, SymmetricThreshold = 2 ,Logistic = 3 ,HyperbolicTangent = 4 ,Linear = 5 ,RectifiedLinear = 6 ,ExponentialLinear = 7 ,ScaledExponentialLinear = 8 ,SoftPlus = 9 ,SoftSign = 10 ,HardSigmoid = 11",
-      "layersActivationFunctions":  "[6,1,1,5,9,5,9,5,11]",
+      "_comment2": "Activation functions explain: Threshold = 1, SymmetricThreshold = 2 ,Logistic = 3 ,HyperbolicTangent = 4 ,Linear = 5,RectifiedLinear = 6 ,ExponentialLinear = 7 ,ScaledExponentialLinear = 8 ,SoftPlus = 9 ,SoftSign = 10 ,HardSigmoid = 11",
+      "layersActivationFunctions":  "[4,1,1,6,6,11,11]",
       "federatedMode": "0",
       "countLimit": "10",
       "_comment4": "LOSS: SUM_SQUARED_ERROR = 1, MSE = 2 /* MSE - Mean Squared Error */, E_LOSS_METHOD_NSE = 3 /* NSE - Normalized Squared Error */, MINKOWSKI_ERROR = 4, WSE = 5 /* WSE - Weighted Squared Error */, CEE = 6 /* CEE - Cross Entropy Error */",
       "lossMethod": "2",
       "_comment5": "TODOTODO explain learningRate",
       "learningRate": "0.01",
       "_comment3": "OptimizerExplain: 0- NONE, 1- SGD, 2- MINI_BATCH_SGD, 3- MOMENTUM, 4- NAG, 5- ADAGRAD, 6- ADAM",
-      "optimizer":"5",
+      "optimizer":"6",
       "features": "13",
-      "labels": "1"
+      "labels": "5"       
     }
   ],
   "clients": [

inputJsonFiles/Architecture/arch_2PC2Workers.json

@@ -11,7 +11,7 @@
       "entities": "mainServer,c1,s1,r1"
     },
     {
-      "host": "192.168.0.102",
+      "host": "192.168.0.103",
       "entities": "c2"
     }
   ],

inputJsonFiles/experimentsFlow/exp_1Worker1SourceHealth.json

@@ -1,31 +1,33 @@
 {
-     "CSV path": "health_comb",
-     "Batches per source": 
-     {
-        "Training": 50,
-        "Prediction": 50
-     },
-     "Training":  [
-         {
-            "source name": "s1",
-            "workers": "w1",
-            "CSV path": "health_comb_train"
-         }
-     ],
-
-     "Prediction": [
-         {
-            "source name": "s1",
-            "workers": "w1",
-            "CSV path": "health_comb_predict"
-         }
-    ],
-
-     "Statistics": [
-         {
-            "workers": "w1",
-            "routers": "r1"
-         }
-    ]
+   "Features": 13,
+   "Labels": [1, 2, 3, 4, 5],
+   "CSV path": "health_comb",
+   "Batches per source": 
+   {
+      "Training": 50,
+      "Prediction": 20
+   },
+   "Training":  [
+      {
+         "source name": "s1",
+         "workers": "w1",
+         "CSV path": "health_comb_train"
+      }
+   ],
+
+   "Prediction": [
+      {
+         "source name": "s1",
+         "workers": "w1",
+         "CSV path": "health_comb_predict"
+      }
+   ],
+
+   "Statistics": [
+      {
+         "workers": "w1",
+         "routers": "r1"
+      }
+   ]
 }