RF saved

RF/Python/RF_training.py

@@ -124,8 +124,9 @@ def display_model_performance_metrics(true_labels, predicted_labels, classes=[1,
 
 
 
-def train_RF(saveWeighs=False, saveAlgo=False):
-    # dataset = r'features.csv'
+def train_RF(saveAlgo=False):
+    count_estimator = 0 #number of estimators processes in the Random Forest
+
     dataset = "build/features.csv"
 
     crop_df = pd.read_csv(dataset, header=0)
@@ -222,27 +223,80 @@ def train_RF(saveWeighs=False, saveAlgo=False):
         # print(e.n_outputs_)
         report = export_text(e)
         # print(report)
-    
     # help(sklearn.tree._tree.Tree)
-
-    # Write in cpp header file CARTTrainedcpp.h
-    # text = str(model_best)
-    # with open('/home/estellearrc/Documents/Cours/ENSTA_Bretagne_2020_2022/3A_2021_2022/S5/Embedded_machine_learning/embedded-machine-learning/RF/RFTrained.h', 'w') as f:
-    #     f.write("  # ifndef RFTRAINED_H\n#define RFTRAINED_H\n#include <string>\n#include <numeric>\n#include <fstream>\n#include <map>\n#include \"../Helpers/globals.h\"\n#include <typeinfo>\n")
-    #     f.write(
-    #         "std::string RFModel(std::map < FTYPE, DataVector > &features){" + "DataVector featureVector = features[FTYPE::BINAVG];featureVector.insert(featureVector.end(), features[FTYPE::BINSTDEV].begin(), features[FTYPE::BINSTDEV].end());"+text + "return 0;}\n#endif")
-    # print("Save algo in embedded-machine-learning/RF/RFTrained.h file!")
-
+    file = '/home/estellearrc/Documents/Cours/ENSTA_Bretagne_2020_2022/3A_2021_2022/S5/Embedded_machine_learning/embedded-machine-learning/RF/RFTrained.h'
+    open(file, 'w').close()
+    with open(file, 'a') as f:
+        f.write("#ifndef RFTRAINED_H\n#define RFTRAINED_H\n#include <string>\n#include <numeric>\n#include <fstream>\n#include <map>\n#include \"../Helpers/globals.h\"\n#include <typeinfo>\n")
+        f.write("const int N_ESTIMATORS = "+str(model_best.n_estimators)+";\nconst int MAX_DEPTH = "+str(model_best.max_depth)+";\n")
+        f.write("float RFModel(int num_estimator, std::map < FTYPE, DataVector > &features){\n") 
+        f.write("DataVector featureVector = features[FTYPE::BINAVG];\nfeatureVector.insert(featureVector.end(), features[FTYPE::BINSTDEV].begin(), features[FTYPE::BINSTDEV].end());\n")
+        f.close()
+    for e in model_best.estimators_:
+        # Write in cpp header file RFTrainedcpp.h
+        with open(file, 'a') as f:
+            text = ""
+            text = str_tree(text, e, model_best.n_estimators, 0, 1)
+            f.write("\nif(num_estimator == "+str(count_estimator)+"){\n" + text + "}\n")
+            f.close()
+            count_estimator+=1
+    with open(file, 'a') as f:
+        f.write("return 0;}\n#endif")
+        f.close()
+    print("Save algo in embedded-machine-learning/RF/RFTrained.h file!")
     return model_best, history_best
 
-def __str__(tree):
-    if tree.left == tree.right == None:
-        return "return " + str(tree.label)+" ;"
+def compute_depth(tree, node):
+    """
+    Returns the depth of the subtree rooted in node.
+    """
+    def compute_depth_(current_node, current_depth,
+                       children_left, children_right, depths):
+        depths += [current_depth]
+        left = children_left[current_node]
+        right = children_right[current_node]
+        if left != -1 and right != -1:
+            compute_depth_(left, current_depth+1,
+                           children_left, children_right, depths)
+            compute_depth_(right, current_depth+1,
+                           children_left, children_right, depths)
+
+    depths = []
+    compute_depth_(node, 1, tree.children_left, tree.children_right, depths)
+    return max(depths)
+
+def str_tree(text,decision_tree, max_depth, node, depth):
+    tree_ = decision_tree.tree_
+    class_names = decision_tree.classes_
+    value = None
+    if tree_.n_outputs == 1:
+        value = tree_.value[node][0]
+    else:
+        value = tree_.value[node].T[0]
+    class_name = np.argmax(value)
+
+    if (tree_.n_classes[0] != 1 and
+            tree_.n_outputs == 1):
+        class_name = class_names[class_name]
+
+    if depth <= max_depth+1:
+        # print("hello")
+        if tree_.children_right[node] != tree_.children_left[node]:
+            left = str_tree(text,decision_tree, max_depth, tree_.children_left[node], depth+1)
+            right = str_tree(text,decision_tree, max_depth, tree_.children_right[node], depth+1)
+            secondeLayerIF = "if(" + "featureVector["+str(tree_.feature[node])+"]" + "<="+str(tree_.threshold[node]) + \
+            "){\n"+ str(left) + "}\n"
+            secondeLayerELSE = "else{\n"+str(right)+"}\n"
+            text += secondeLayerIF + secondeLayerELSE
+            print(text)
+        else:  # leaf
+            text+= "return " + str(class_name)+";\n"
     else:
-        secondeLayerIF = "if(" + "featureVector["+str(tree.feature)+"]" + "<="+str(tree.threshold) + \
-            "){"+str(tree.left.__str__()) + "}"
-        secondeLayerELSE = "else{"+str(tree.right.__str__())+"}"
-        return secondeLayerIF + secondeLayerELSE
+        subtree_depth = compute_depth(tree_, node)
+        if subtree_depth == 1:
+            text+= "return " + str(class_name)+";\n"
+    # print(text)
+    return text
 
 def transformArrayToStr(array):
     array = array.T
@@ -271,4 +325,5 @@ def save_file_as_csv(data, name_folder):
 
 
 if __name__ == "__main__":
-    model, history = train_RF(saveAlgo=False, saveWeighs=True)
+    model, history = train_RF(saveAlgo=False)
+    

RF/main.cpp

+// HEADER=all_headers.h
+// echo "#ifndef __ALL_HEADERS__" > $HEADER
+// echo "#define __ALL_HEADERS__" >> $HEADER
+// for file in 'dir/*.h'
+// do
+//     echo "#include <$file>" >> $HEADER
+// done
+// echo "#endif" >> $HEADER
 #include "RFTrained.h"
 #include "../Extraction/features_extraction.cpp"
 #include <string>