Izmenjen prediktor i omogućeno čuvanje metrika.

2 files changed, 42 insertions, 32 deletions

diff --git a/backend/microservice/api/controller.py b/backend/microservice/api/controller.py
index 6f483008..0eee571c 100644
--- a/backend/microservice/api/controller.py
+++ b/backend/microservice/api/controller.py
@@ -69,15 +69,26 @@ def train():
     
 
     #dataset, paramsModel, paramsExperiment, callback)
-    filepath,result,finalMetrics= newmlservice.train(data, paramsModel, paramsExperiment,paramsDataset, train_callback)
+    filepath,histMetrics= newmlservice.train(data, paramsModel, paramsExperiment,paramsDataset, train_callback)
     """
     f = request.json['filepath']
     dataset = pd.read_csv(f)
     filepath,result=newmlservice.train(dataset,request.json['model'],train_callback)
     print(result)
     """
-
-
+    #m = []
+    #for attribute, value in result.items():
+       #m.append(histMetrics(attribute,str(value)).__dict__)
+    '''
+    m = []
+    for attribute, value in result.items():
+        m.append({"Name" : attribute, "JsonValue" : value}))
+    
+    print("**************************************************************")
+    print(m)
+    
+    print("**************************************************************")
+    '''
     url = config.api_url + "/file/h5"
     files = {'file': open(filepath, 'rb')}
     r=requests.post(url, files=files,data={"uploaderId":paramsExperiment['uploaderId']})
@@ -92,15 +103,23 @@ def train():
         "experimentId" : paramsExperiment["_id"],
         "modelId" : paramsModel["_id"],
         "h5FileId" : fileId,
-        "metrics" : result,
-        "finalMetrics":finalMetrics
+        "metricsLoss":histMetrics[0],
+        "metricsValLoss":histMetrics[1],
+        "metricsAcc":histMetrics[2],
+        "metricsValAcc":histMetrics[3],
+        "metricsMae":histMetrics[4],
+        "metricsValMae":histMetrics[5],
+        "metricsMse":histMetrics[6],
+        "metricsValMse":histMetrics[7]
+
+
     }
     #print(predictor)
-   
+    
     url = config.api_url + "/Predictor/add"
     r = requests.post(url, json=predictor).text
     
-    print(r)
+    #print(r)
     return r
 
 @app.route('/predict', methods = ['POST'])
diff --git a/backend/microservice/api/newmlservice.py b/backend/microservice/api/newmlservice.py
index bcff5a33..85be0c2f 100644
--- a/backend/microservice/api/newmlservice.py
+++ b/backend/microservice/api/newmlservice.py
@@ -303,7 +303,7 @@ def train(dataset, paramsModel,paramsExperiment,paramsDataset,callback):
     ###OPTIMIZATORI
     print(paramsModel['optimizer'])
     if(paramsModel['optimizer']=='Adam'):
-        opt=tf.keras.optimizers.Adam(learning_rate=3)
+        opt=tf.keras.optimizers.Adam(learning_rate=float(paramsModel['learningRate']))
 
     elif(paramsModel['optimizer']=='Adadelta'):
         opt=tf.keras.optimizers.Adadelta(learning_rate=float(paramsModel['learningRate']))
@@ -370,7 +370,7 @@ def train(dataset, paramsModel,paramsExperiment,paramsDataset,callback):
 
         
 
-        classifier.compile(loss =paramsModel["lossFunction"] , optimizer =opt, metrics = ['mae','mse'])
+        classifier.compile(loss =paramsModel["lossFunction"] , optimizer =opt, metrics = ['accuracy','mae','mse'])
 
         history=classifier.fit( x=x_train, y=y_train, epochs = paramsModel['epochs'],batch_size=int(paramsModel['batchSize']),callbacks=callback(x_test, y_test,paramsModel['_id']),validation_data=(x_val, y_val))
 
@@ -383,9 +383,9 @@ def train(dataset, paramsModel,paramsExperiment,paramsDataset,callback):
         scores = classifier.evaluate(x_test, y_test)
         #print("\n%s: %.2f%%" % (classifier.metrics_names[1], scores[1]*100))
         
-        
+        '''
         classifier.save(filepath, save_format='h5')
-        metrics={}
+       
         macro_averaged_precision=sm.precision_score(y_test, y_pred, average = 'macro')
         micro_averaged_precision=sm.precision_score(y_test, y_pred, average = 'micro')
         macro_averaged_recall=sm.recall_score(y_test, y_pred, average = 'macro')
@@ -401,12 +401,12 @@ def train(dataset, paramsModel,paramsExperiment,paramsDataset,callback):
                 {"Name":"macro_averaged_f1","JsonValue": str(macro_averaged_f1)},
                 {"Name":"micro_averaged_f1", "JsonValue": str(micro_averaged_f1)}
         ]
-
+        '''
         #vizuelizacija u python-u
         #from ann_visualizer.visualize import ann_viz;
         #ann_viz(classifier, title="My neural network")
         
-        return filepath,hist,metrics
+        return filepath,[hist['loss'],hist['val_loss'],hist['accuracy'],hist['val_accuracy'],hist['mae'],hist['val_mae'],hist['mse'],hist['val_mse']]
 
     elif(problem_type=='binarni-klasifikacioni'):
         #print('*************************************************************************binarni')
@@ -444,6 +444,7 @@ def train(dataset, paramsModel,paramsExperiment,paramsDataset,callback):
 
         history=classifier.fit( x=x_train, y=y_train, epochs = paramsModel['epochs'],batch_size=int(paramsModel['batchSize']),callbacks=callback(x_test, y_test,paramsModel['_id']),validation_data=(x_val, y_val))
         hist=history.history
+        
         y_pred=classifier.predict(x_test)
         y_pred=(y_pred>=0.5).astype('int')
         
@@ -452,7 +453,7 @@ def train(dataset, paramsModel,paramsExperiment,paramsDataset,callback):
         # ann_viz(classifier, title="My neural network")
         
         classifier.save(filepath, save_format='h5')
-
+        """
         accuracy = float(sm.accuracy_score(y_test,y_pred))
         precision = float(sm.precision_score(y_test,y_pred))
         recall = float(sm.recall_score(y_test,y_pred))
@@ -461,22 +462,9 @@ def train(dataset, paramsModel,paramsExperiment,paramsDataset,callback):
         f1 = float(sm.f1_score(y_test,y_pred))
         fpr, tpr, _ = sm.roc_curve(y_test,y_pred)
         logloss = float(sm.log_loss(y_test, y_pred))
-        metrics= [
-            {"Name":"accuracy" ,"JsonValue": str(accuracy)},
-            {"Name":"precision","JsonValue": str(precision)},
-            {"Name":"recall" , "JsonValue":str(recall)},
-            {"Name":"specificity" ,"JsonValue":str(specificity)},
-            {"Name":"f1" ,"JsonValue": str(f1)},
-            {"Name":"tn" , "JsonValue":str(tn)},
-            {"Name":"fp" , "JsonValue":str(fp)},
-            {"Name":"fn" , "JsonValue":str(fn)},
-            {"Name":"tp" , "JsonValue":str(tp)},
-            {"Name":"fpr" ,"JsonValue": str(fpr.tolist())},
-            {"Name":"tpr" , "JsonValue":str(tpr.tolist())},
-            {"Name":"logloss" , "JsonValue":str(logloss)}
-        ]
+        """
 
-        return filepath,hist,metrics
+        return filepath,[hist['loss'],hist['val_loss'],hist['accuracy'],hist['val_accuracy'],hist['mae'],hist['val_mae'],hist['mse'],hist['val_mse']]
 
     elif(problem_type=='regresioni'):
         reg=paramsModel['layers'][0]['regularisation']
@@ -519,8 +507,10 @@ def train(dataset, paramsModel,paramsExperiment,paramsDataset,callback):
         #print(classifier.evaluate(x_test, y_test))
  
         classifier.save(filepath, save_format='h5')
-       
+        '''
+
         mse = float(sm.mean_squared_error(y_test,y_pred))
+        
         mae = float(sm.mean_absolute_error(y_test,y_pred))
         mape = float(sm.mean_absolute_percentage_error(y_test,y_pred))
         rmse = float(np.sqrt(sm.mean_squared_error(y_test,y_pred)))
@@ -531,6 +521,7 @@ def train(dataset, paramsModel,paramsExperiment,paramsDataset,callback):
         n = 40
         k = 2
         adj_r2 = float(1 - ((1-r2)*(n-1)/(n-k-1)))
+       
         metrics= [
             {"Name":"mse","JsonValue":str(mse)},
 
@@ -541,8 +532,8 @@ def train(dataset, paramsModel,paramsExperiment,paramsDataset,callback):
             {"Name":"r2","JsonValue":str( r2)},
             {"Name":"adj_r2","JsonValue":str(adj_r2)}
             ]
-        
-        return filepath,hist,metrics
+        '''
+        return filepath,[hist['loss'],hist['val_loss'],[],[],hist['mae'],hist['val_mae'],hist['mse'],hist['val_mse']]
 
     def roc_auc_score_multiclass(actual_class, pred_class, average = "macro"):