Added new messages for showing metrics. Brough training to a the model-class. Added comments

Sentiment_Analysis/Readme.md

@@ -120,14 +120,12 @@ python3 -m pip install grpcio-tools googleapis-commonprotos
 
 * Now, run this command inside csv_databroker folder:
 ```commandline
-python3 -m grpc_tools.protoc -I. --python_out=. --
-grpc_python_out=. databroker.proto
+python3 -m grpc_tools.protoc -I. --python_out=. --grpc_python_out=. databroker.proto
 ```
 
 * Now, run this command inside Sentiment_Analysis Command:
 ```commandline
-python3 -m grpc_tools.protoc -I. --python_out=. --
-grpc_python_out=. model.proto
+python3 -m grpc_tools.protoc -I. --python_out=. --grpc_python_out=. model.proto
 ```
 This command used model.proto file to generate the needed stubs to create the
 client/server.

Sentiment_Analysis/sa_databroker/model.proto

@@ -10,7 +10,13 @@ message Text {
   string query = 1;
 }
 
+// Message send to extract dataset metadata
+message DatasetMetaData {
+  string datasetName = 1;
+}
+
 //Define the service
 service Databroker {
     rpc sadatabroker(Empty) returns (Text);
-}
+    rpc log_dataset_metadata(DatasetMetaData) returns (Empty);
+}
\ No newline at end of file

Sentiment_Analysis/sa_databroker/model_pb2.py

@@ -19,7 +19,7 @@ DESCRIPTOR = _descriptor.FileDescriptor(
   syntax='proto3',
   serialized_options=None,
   create_key=_descriptor._internal_create_key,
-  serialized_pb=b'\n\x0bmodel.proto\"\x07\n\x05\x45mpty\"\x15\n\x04Text\x12\r\n\x05query\x18\x01 \x01(\t2+\n\nDatabroker\x12\x1d\n\x0csadatabroker\x12\x06.Empty\x1a\x05.Textb\x06proto3'
+  serialized_pb=b'\n\x0bmodel.proto\"\x07\n\x05\x45mpty\"\x15\n\x04Text\x12\r\n\x05query\x18\x01 \x01(\t\"&\n\x0f\x44\x61tasetMetaData\x12\x13\n\x0b\x64\x61tasetName\x18\x01 \x01(\t2]\n\nDatabroker\x12\x1d\n\x0csadatabroker\x12\x06.Empty\x1a\x05.Text\x12\x30\n\x14log_dataset_metadata\x12\x10.DatasetMetaData\x1a\x06.Emptyb\x06proto3'
 )
 
 
@@ -81,8 +81,41 @@ _TEXT = _descriptor.Descriptor(
   serialized_end=45,
 )
 
+
+_DATASETMETADATA = _descriptor.Descriptor(
+  name='DatasetMetaData',
+  full_name='DatasetMetaData',
+  filename=None,
+  file=DESCRIPTOR,
+  containing_type=None,
+  create_key=_descriptor._internal_create_key,
+  fields=[
+    _descriptor.FieldDescriptor(
+      name='datasetName', full_name='DatasetMetaData.datasetName', index=0,
+      number=1, type=9, cpp_type=9, label=1,
+      has_default_value=False, default_value=b"".decode('utf-8'),
+      message_type=None, enum_type=None, containing_type=None,
+      is_extension=False, extension_scope=None,
+      serialized_options=None, file=DESCRIPTOR,  create_key=_descriptor._internal_create_key),
+  ],
+  extensions=[
+  ],
+  nested_types=[],
+  enum_types=[
+  ],
+  serialized_options=None,
+  is_extendable=False,
+  syntax='proto3',
+  extension_ranges=[],
+  oneofs=[
+  ],
+  serialized_start=47,
+  serialized_end=85,
+)
+
 DESCRIPTOR.message_types_by_name['Empty'] = _EMPTY
 DESCRIPTOR.message_types_by_name['Text'] = _TEXT
+DESCRIPTOR.message_types_by_name['DatasetMetaData'] = _DATASETMETADATA
 _sym_db.RegisterFileDescriptor(DESCRIPTOR)
 
 Empty = _reflection.GeneratedProtocolMessageType('Empty', (_message.Message,), {
@@ -99,6 +132,13 @@ Text = _reflection.GeneratedProtocolMessageType('Text', (_message.Message,), {
   })
 _sym_db.RegisterMessage(Text)
 
+DatasetMetaData = _reflection.GeneratedProtocolMessageType('DatasetMetaData', (_message.Message,), {
+  'DESCRIPTOR' : _DATASETMETADATA,
+  '__module__' : 'model_pb2'
+  # @@protoc_insertion_point(class_scope:DatasetMetaData)
+  })
+_sym_db.RegisterMessage(DatasetMetaData)
+
 
 
 _DATABROKER = _descriptor.ServiceDescriptor(
@@ -108,8 +148,8 @@ _DATABROKER = _descriptor.ServiceDescriptor(
   index=0,
   serialized_options=None,
   create_key=_descriptor._internal_create_key,
-  serialized_start=47,
-  serialized_end=90,
+  serialized_start=87,
+  serialized_end=180,
   methods=[
   _descriptor.MethodDescriptor(
     name='sadatabroker',
@@ -121,6 +161,16 @@ _DATABROKER = _descriptor.ServiceDescriptor(
     serialized_options=None,
     create_key=_descriptor._internal_create_key,
   ),
+  _descriptor.MethodDescriptor(
+    name='log_dataset_metadata',
+    full_name='Databroker.log_dataset_metadata',
+    index=1,
+    containing_service=None,
+    input_type=_DATASETMETADATA,
+    output_type=_EMPTY,
+    serialized_options=None,
+    create_key=_descriptor._internal_create_key,
+  ),
 ])
 _sym_db.RegisterServiceDescriptor(_DATABROKER)
 

Sentiment_Analysis/sa_databroker/model_pb2_grpc.py

@@ -20,6 +20,11 @@ class DatabrokerStub(object):
                 request_serializer=model__pb2.Empty.SerializeToString,
                 response_deserializer=model__pb2.Text.FromString,
                 )
+        self.log_dataset_metadata = channel.unary_unary(
+                '/Databroker/log_dataset_metadata',
+                request_serializer=model__pb2.DatasetMetaData.SerializeToString,
+                response_deserializer=model__pb2.Empty.FromString,
+                )
 
 
 class DatabrokerServicer(object):
@@ -32,6 +37,12 @@ class DatabrokerServicer(object):
         context.set_details('Method not implemented!')
         raise NotImplementedError('Method not implemented!')
 
+    def log_dataset_metadata(self, request, context):
+        """Missing associated documentation comment in .proto file."""
+        context.set_code(grpc.StatusCode.UNIMPLEMENTED)
+        context.set_details('Method not implemented!')
+        raise NotImplementedError('Method not implemented!')
+
 
 def add_DatabrokerServicer_to_server(servicer, server):
     rpc_method_handlers = {
@@ -40,6 +51,11 @@ def add_DatabrokerServicer_to_server(servicer, server):
                     request_deserializer=model__pb2.Empty.FromString,
                     response_serializer=model__pb2.Text.SerializeToString,
             ),
+            'log_dataset_metadata': grpc.unary_unary_rpc_method_handler(
+                    servicer.log_dataset_metadata,
+                    request_deserializer=model__pb2.DatasetMetaData.FromString,
+                    response_serializer=model__pb2.Empty.SerializeToString,
+            ),
     }
     generic_handler = grpc.method_handlers_generic_handler(
             'Databroker', rpc_method_handlers)
@@ -67,3 +83,20 @@ class Databroker(object):
             model__pb2.Text.FromString,
             options, channel_credentials,
             insecure, call_credentials, compression, wait_for_ready, timeout, metadata)
+
+    @staticmethod
+    def log_dataset_metadata(request,
+            target,
+            options=(),
+            channel_credentials=None,
+            call_credentials=None,
+            insecure=False,
+            compression=None,
+            wait_for_ready=None,
+            timeout=None,
+            metadata=None):
+        return grpc.experimental.unary_unary(request, target, '/Databroker/log_dataset_metadata',
+            model__pb2.DatasetMetaData.SerializeToString,
+            model__pb2.Empty.FromString,
+            options, channel_credentials,
+            insecure, call_credentials, compression, wait_for_ready, timeout, metadata)

Sentiment_Analysis/sa_databroker/sa_databroker_client.py

@@ -15,7 +15,6 @@ def run():
         stub = model_pb2_grpc.DatabrokerStub(channel)
         ui_request = model_pb2.Empty()
         response = stub.sadatabroker(ui_request)
-
     print("Greeter client received: ")
     print(response)
 

Sentiment_Analysis/sa_databroker/sa_databroker_server.py

@@ -19,7 +19,14 @@ class DatabrokerServicer(model_pb2_grpc.DatabrokerServicer):
     def __init__(self):
         super().__init__()
         self.send_data=True
+        # call the function retrieving metadata on first generation of the container.
 
+        ##########
+        # for testing only:
+        request = model_pb2.DatasetMetaData()
+        request.datasetName = "test Dataset"
+        self.get_dataset_metadata(request, None)
+        ###########
 
     def sadatabroker(self, request, context):
         # this simple example without streaming sends only one data record, then the pipeline must be run again
@@ -34,6 +41,13 @@ class DatabrokerServicer(model_pb2_grpc.DatabrokerServicer):
         return response
 
 
+    # Getting meta data of the given dataset.
+    def log_dataset_metadata(self, request, context):
+        datasetName = request.datasetName
+        logger.debug("test logging metadata")
+        logger.debug(datasetName)
+
+
 def serve():
     server = grpc.server(futures.ThreadPoolExecutor(max_workers=10))
     model_pb2_grpc.add_DatabrokerServicer_to_server(DatabrokerServicer(), server)

Sentiment_Analysis/sa_model/app.py

@@ -22,7 +22,6 @@ def hello():
     headers = ['Text', 'Classification']
     return render_template("index.html", headers=headers, results=results)
 
-
 # if __name__ == '__main__':
 def app_run():
     app.secret_key = "samodel"

Sentiment_Analysis/sa_model/model.proto

 //Define the used version of proto
 syntax = "proto3";
 
+message Empty {
+
+}
+
 //Define a message to hold the features input by the client
 message Text {
   string query = 1;
@@ -11,7 +15,13 @@ message Review_Classify {
   float review = 1 ;
 }
 
+message predictionMetric {
+  float loss = 1;
+  float accuracy = 2;
+}
+
 //Define the service
 service sentiment_analysis_model {
   rpc classify_review(Text) returns (Review_Classify);
+  rpc calculate_metrics(Empty) returns (predictionMetric);
 }
\ No newline at end of file

Sentiment_Analysis/sa_model/model.py

@@ -2,20 +2,96 @@ from keras.models import load_model
 from keras.datasets import imdb
 from keras.preprocessing import sequence
 
+from keras.models import Sequential
+from keras.layers import Dense, Embedding, Conv1D, GlobalMaxPooling1D
+from keras.datasets import imdb
+
+class SentimentAnalysis():
+    def __init__(self):
+        # The maximum length of the input sequences.
+        self.maxlen = 100
+        # The maximum number of words in the vocabulary.
+        self.max_features = 5000
+        # The dimension of the embedding vectors.
+        self.embedding_dims = 32
+        
+        ######
+        # NN settings
+        # The number of filters to use in the convolutional layer. This is the number of output channels that the convolutional layer will produce.
+        self.filters = 128
+        # The size of the filter kernel. This is the number of input elements that the filter will be applied to.
+        self.kernel_size = 3
+        # The stride of the convolutional layer. This is the number of input elements to skip between each application of the filter kernel.
+        self.strides = 1
+        #The type of padding to use. The default is valid, which means that the output of the convolutional layer will be the same size as the input.
+        self.padding = 'valid'
+
+        self.batch_size = 64
+        self.epochs = 5
+
+        self.model_saving_path = "model.h5"
+        
+
+    def predict(self,input):
+        model = load_model(self.model_saving_path)
+        d = imdb.get_word_index()
+        words = input.split()
+        review = []
+        for word in words:
+            if word not in d:
+                review.append(2)
+            else:
+                review.append(d[word] + 3)
+
+        review = sequence.pad_sequences([review],
+                                        truncating='pre', padding='pre', maxlen=self.maxlen)
+        prediction = model.predict(review)
+        return prediction[0][0]
+
+
+    def calculate_metrics(self):
+        model = load_model(self.model_saving_path)
+        _, (x_test, y_test) = imdb.load_data(num_words=self.max_features)
+        x_test = sequence.pad_sequences(x_test, maxlen=self.maxlen)
+        score = model.evaluate(x_test, y_test, verbose=0)
+        score_dic = {
+            "loss": score[0],
+            "accuracy": score[1]
+        }
+        return score_dic
+    
+
+    def train(self):
+
+        (x_train, y_train), _ = imdb.load_data(num_words=self.max_features)
+
+        x_train = sequence.pad_sequences(x_train, maxlen=self.maxlen)
+
+        model = Sequential()
+        # The embedding layer will convert each word in the input sequence to a dense vector of fixed size. 
+        # This vector will represent the meaning of the word in a more meaningful way than the one-hot encoded vector.
+        model.add(Embedding(self.max_features, self.embedding_dims, input_length=self.maxlen))
+        model.add(Conv1D(self.filters, self.kernel_size, padding=self.padding, activation='relu', strides=self.strides))
+        # Global max pooling is a type of pooling operation that takes the maximum value over the entire input sequence. 
+        # This means that the output of the global max pooling layer will be a single vector, regardless of the length of the input sequence.
+        model.add(GlobalMaxPooling1D())
+        # Adds a dense layer with 128 neurons and a sigmoid activation function
+        # Dense layers are the most common type of layer in neural networks. They take a vector of inputs and produce a vector of outputs. 
+        # Each neuron in a dense layer is connected to all of the neurons in the previous layer.
+        # The sigmoid activation function is a non-linear function that squashes its input to a value between 0 and 1. 
+        # It is often used in classification tasks, where the output of the model should be a probability between 0 and 1.
+
+        ## makes no sense to have a Dense with 128 nerons after pooling?
+        #model.add(Dense(128, activation='sigmoid'))
+        model.add(Dense(1, activation='sigmoid'))
+
+        model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['acc'])
+
+        history = model.fit(x_train, y_train, batch_size=self.batch_size, epochs=self.epochs)
+        model.save(self.model_saving_path)
 
-def classify_review(review):
-    maxlen = 100
-    model = load_model('model.h5')
-    d = imdb.get_word_index()
-    words = review.split()
-    review = []
+        print("Saved model to disk")
+        print("Accuracy in training:")
+        print(history.history['acc'])
+        print("loss in training:")
+        print(history.history['loss'])
\ No newline at end of file
-    for word in words:
-        if word not in d:
-            review.append(2)
-        else:
-            review.append(d[word] + 3)
-
-    review = sequence.pad_sequences([review],
-                                    truncating='pre', padding='pre', maxlen=maxlen)
-    prediction = model.predict(review)
-    return prediction[0][0]

Sentiment_Analysis/sa_model/model_pb2.py

@@ -19,12 +19,37 @@ DESCRIPTOR = _descriptor.FileDescriptor(
   syntax='proto3',
   serialized_options=None,
   create_key=_descriptor._internal_create_key,
-  serialized_pb=b'\n\x0bmodel.proto\"\x15\n\x04Text\x12\r\n\x05query\x18\x01 \x01(\t\"!\n\x0fReview_Classify\x12\x0e\n\x06review\x18\x01 \x01(\x02\x32\x46\n\x18sentiment_analysis_model\x12*\n\x0f\x63lassify_review\x12\x05.Text\x1a\x10.Review_Classifyb\x06proto3'
+  serialized_pb=b'\n\x0bmodel.proto\"\x07\n\x05\x45mpty\"\x15\n\x04Text\x12\r\n\x05query\x18\x01 \x01(\t\"!\n\x0fReview_Classify\x12\x0e\n\x06review\x18\x01 \x01(\x02\"2\n\x10predictionMetric\x12\x0c\n\x04loss\x18\x01 \x01(\x02\x12\x10\n\x08\x61\x63\x63uracy\x18\x02 \x01(\x02\x32v\n\x18sentiment_analysis_model\x12*\n\x0f\x63lassify_review\x12\x05.Text\x1a\x10.Review_Classify\x12.\n\x11\x63\x61lculate_metrics\x12\x06.Empty\x1a\x11.predictionMetricb\x06proto3'
 )
 
 
 
 
+_EMPTY = _descriptor.Descriptor(
+  name='Empty',
+  full_name='Empty',
+  filename=None,
+  file=DESCRIPTOR,
+  containing_type=None,
+  create_key=_descriptor._internal_create_key,
+  fields=[
+  ],
+  extensions=[
+  ],
+  nested_types=[],
+  enum_types=[
+  ],
+  serialized_options=None,
+  is_extendable=False,
+  syntax='proto3',
+  extension_ranges=[],
+  oneofs=[
+  ],
+  serialized_start=15,
+  serialized_end=22,
+)
+
+
 _TEXT = _descriptor.Descriptor(
   name='Text',
   full_name='Text',
@@ -52,8 +77,8 @@ _TEXT = _descriptor.Descriptor(
   extension_ranges=[],
   oneofs=[
   ],
-  serialized_start=15,
-  serialized_end=36,
+  serialized_start=24,
+  serialized_end=45,
 )
 
 
@@ -84,14 +109,62 @@ _REVIEW_CLASSIFY = _descriptor.Descriptor(
   extension_ranges=[],
   oneofs=[
   ],
-  serialized_start=38,
-  serialized_end=71,
+  serialized_start=47,
+  serialized_end=80,
+)
+
+
+_PREDICTIONMETRIC = _descriptor.Descriptor(
+  name='predictionMetric',
+  full_name='predictionMetric',
+  filename=None,
+  file=DESCRIPTOR,
+  containing_type=None,
+  create_key=_descriptor._internal_create_key,
+  fields=[
+    _descriptor.FieldDescriptor(
+      name='loss', full_name='predictionMetric.loss', index=0,
+      number=1, type=2, cpp_type=6, label=1,
+      has_default_value=False, default_value=float(0),
+      message_type=None, enum_type=None, containing_type=None,
+      is_extension=False, extension_scope=None,
+      serialized_options=None, file=DESCRIPTOR,  create_key=_descriptor._internal_create_key),
+    _descriptor.FieldDescriptor(
+      name='accuracy', full_name='predictionMetric.accuracy', index=1,
+      number=2, type=2, cpp_type=6, label=1,
+      has_default_value=False, default_value=float(0),
+      message_type=None, enum_type=None, containing_type=None,
+      is_extension=False, extension_scope=None,
+      serialized_options=None, file=DESCRIPTOR,  create_key=_descriptor._internal_create_key),
+  ],
+  extensions=[
+  ],
+  nested_types=[],
+  enum_types=[
+  ],
+  serialized_options=None,
+  is_extendable=False,
+  syntax='proto3',
+  extension_ranges=[],
+  oneofs=[
+  ],
+  serialized_start=82,
+  serialized_end=132,
 )
 
+DESCRIPTOR.message_types_by_name['Empty'] = _EMPTY
 DESCRIPTOR.message_types_by_name['Text'] = _TEXT
 DESCRIPTOR.message_types_by_name['Review_Classify'] = _REVIEW_CLASSIFY
+DESCRIPTOR.message_types_by_name['predictionMetric'] = _PREDICTIONMETRIC
 _sym_db.RegisterFileDescriptor(DESCRIPTOR)
 
+Empty = _reflection.GeneratedProtocolMessageType('Empty', (_message.Message,), {
+  'DESCRIPTOR' : _EMPTY,
+  '__module__' : 'model_pb2'
+  # @@protoc_insertion_point(class_scope:Empty)
+  })
+_sym_db.RegisterMessage(Empty)
+
 Text = _reflection.GeneratedProtocolMessageType('Text', (_message.Message,), {
   'DESCRIPTOR' : _TEXT,
   '__module__' : 'model_pb2'
@@ -106,6 +179,13 @@ Review_Classify = _reflection.GeneratedProtocolMessageType('Review_Classify', (_
   })
 _sym_db.RegisterMessage(Review_Classify)
 
+predictionMetric = _reflection.GeneratedProtocolMessageType('predictionMetric', (_message.Message,), {
+  'DESCRIPTOR' : _PREDICTIONMETRIC,
+  '__module__' : 'model_pb2'
+  # @@protoc_insertion_point(class_scope:predictionMetric)
+  })
+_sym_db.RegisterMessage(predictionMetric)
+
 
 
 _SENTIMENT_ANALYSIS_MODEL = _descriptor.ServiceDescriptor(
@@ -115,8 +195,8 @@ _SENTIMENT_ANALYSIS_MODEL = _descriptor.ServiceDescriptor(
   index=0,
   serialized_options=None,
   create_key=_descriptor._internal_create_key,
-  serialized_start=73,
-  serialized_end=143,
+  serialized_start=134,
+  serialized_end=252,
   methods=[
   _descriptor.MethodDescriptor(
     name='classify_review',
@@ -128,6 +208,16 @@ _SENTIMENT_ANALYSIS_MODEL = _descriptor.ServiceDescriptor(
     serialized_options=None,
     create_key=_descriptor._internal_create_key,
   ),
+  _descriptor.MethodDescriptor(
+    name='calculate_metrics',
+    full_name='sentiment_analysis_model.calculate_metrics',
+    index=1,
+    containing_service=None,
+    input_type=_EMPTY,
+    output_type=_PREDICTIONMETRIC,
+    serialized_options=None,
+    create_key=_descriptor._internal_create_key,
+  ),
 ])
 _sym_db.RegisterServiceDescriptor(_SENTIMENT_ANALYSIS_MODEL)
 

Sentiment_Analysis/sa_model/model_pb2_grpc.py

@@ -20,6 +20,11 @@ class sentiment_analysis_modelStub(object):
                 request_serializer=model__pb2.Text.SerializeToString,
                 response_deserializer=model__pb2.Review_Classify.FromString,
                 )
+        self.calculate_metrics = channel.unary_unary(
+                '/sentiment_analysis_model/calculate_metrics',
+                request_serializer=model__pb2.Empty.SerializeToString,
+                response_deserializer=model__pb2.predictionMetric.FromString,
+                )
 
 
 class sentiment_analysis_modelServicer(object):
@@ -32,6 +37,12 @@ class sentiment_analysis_modelServicer(object):
         context.set_details('Method not implemented!')
         raise NotImplementedError('Method not implemented!')
 
+    def calculate_metrics(self, request, context):
+        """Missing associated documentation comment in .proto file."""
+        context.set_code(grpc.StatusCode.UNIMPLEMENTED)
+        context.set_details('Method not implemented!')
+        raise NotImplementedError('Method not implemented!')
+
 
 def add_sentiment_analysis_modelServicer_to_server(servicer, server):
     rpc_method_handlers = {
@@ -40,6 +51,11 @@ def add_sentiment_analysis_modelServicer_to_server(servicer, server):
                     request_deserializer=model__pb2.Text.FromString,
                     response_serializer=model__pb2.Review_Classify.SerializeToString,
             ),
+            'calculate_metrics': grpc.unary_unary_rpc_method_handler(
+                    servicer.calculate_metrics,
+                    request_deserializer=model__pb2.Empty.FromString,
+                    response_serializer=model__pb2.predictionMetric.SerializeToString,
+            ),
     }
     generic_handler = grpc.method_handlers_generic_handler(
             'sentiment_analysis_model', rpc_method_handlers)
@@ -67,3 +83,20 @@ class sentiment_analysis_model(object):
             model__pb2.Review_Classify.FromString,
             options, channel_credentials,
             insecure, call_credentials, compression, wait_for_ready, timeout, metadata)
+
+    @staticmethod
+    def calculate_metrics(request,
+            target,
+            options=(),
+            channel_credentials=None,
+            call_credentials=None,
+            insecure=False,
+            compression=None,
+            wait_for_ready=None,
+            timeout=None,
+            metadata=None):
+        return grpc.experimental.unary_unary(request, target, '/sentiment_analysis_model/calculate_metrics',
+            model__pb2.Empty.SerializeToString,
+            model__pb2.predictionMetric.FromString,
+            options, channel_credentials,
+            insecure, call_credentials, compression, wait_for_ready, timeout, metadata)

Sentiment_Analysis/sa_model/sentiment_analysis_server.py

@@ -16,11 +16,16 @@ results = []
 
 
 class sentiment_analysis_modelServicer(model_pb2_grpc.sentiment_analysis_modelServicer):
+
+    def __init__(self):
+        self.psp_obj = psp.SentimentAnalysis()
+
+
     def classify_review(self, request, context):
         # define the buffer of the response :
         response = model_pb2.Review_Classify()
         # get the value of the response by calling the desired function :
-        response.review = psp.classify_review(request.query)
+        response.review = self.psp_obj.predict(request.query)
         result = [request.query, response.review]
         with open("results.txt", mode="a+") as f:
             # for e0, e1, e2, e3, e4, e5 in result:
@@ -29,6 +34,22 @@ class sentiment_analysis_modelServicer(model_pb2_grpc.sentiment_analysis_modelSe
         return response
 
 
+    #The function will get the metrics calculated in predict_sale_price.py    
+    def calculate_metrics(self, request, context):
+        # get the metrics
+        score_dic = self.psp_obj.calculate_metrics()
+
+        # make a grcp message object from the proto file.
+        response = model_pb2.predictionMetric()
+        response.loss = score_dic["loss"]
+        response.accuracy = score_dic["accuracy"]
+
+        logging.debug('Test loss:', score_dic["loss"])
+        logging.debug('Test accuracy:', score_dic["accuracy"])
+
+        return response
+
+
 def serve():
     server = grpc.server(futures.ThreadPoolExecutor(max_workers=10))
     model_pb2_grpc.add_sentiment_analysis_modelServicer_to_server(sentiment_analysis_modelServicer(), server)
@@ -37,8 +58,9 @@ def serve():
     threading.Thread(target=app_run()).start()
     server.wait_for_termination()
 
-
+    
 if __name__ == '__main__':
+    
     logging.basicConfig()
     open('results.txt', 'w').close()
     serve()

Sentiment_Analysis/sa_model/trail.csv

-MSSubClass,LotArea,YearBuilt,BedroomAbvGr,TotRmsAbvGrd
-60,8450,2003,3,8
-20,9600,1976,3,6
-60,11250,2001,3,6
-70,9550,1915,3,7
-60,14260,2000,4,9
-50,14115,1993,1,5
-20,10084,2004,3,7
-60,10382,1973,3,7
-50,6120,1931,2,8
-190,7420,1939,2,5