Пример #1
0
 def score(self, input_data, label):
     input_pad = self.prepare_input(input_data)
     label = self.prepare_label(label)
     prediction = self.model.predict(input_pad)
     get_score_senti(label, prediction)
     get_confusion_matrix(label,
                          prediction)  #implemented cofusion matrix for it
Пример #2
0
    def train(self,
              train_input_data,
              train_label,
              valid_input_data,
              valid_label,
              class_weights=None):
        x_train = self.prepare_input(train_input_data)
        y_train = self.prepare_label(train_label)
        # np.set_printoptions(threshold=np.inf)
        x_valid = self.prepare_input(valid_input_data)
        y_valid = self.prepare_label(valid_label)

        print('start training...')
        history = self.model.fit(x=x_train,
                                 y=y_train,
                                 batch_size=self.config.batch_size,
                                 epochs=self.config.n_epochs,
                                 validation_data=(x_valid, y_valid),
                                 callbacks=self.callbacks,
                                 class_weight=class_weights)
        plt.plot(history.history['acc'])
        plt.plot(history.history['val_acc'])
        plt.title('Model accuracy')
        plt.ylabel('Accuracy')
        plt.xlabel('Epoch')
        plt.legend(['Train', 'Test'], loc='upper left')
        plt.savefig('model_acc.png')
        print('training end...')

        print('score over valid data:')
        valid_pred = self.model.predict(x_valid)
        get_score_senti(y_valid, valid_pred)
 def score(self, input_data, label):
     input_pad = self.prepare_input(input_data)
     label = self.prepare_label(label)
     prediction = self.model.predict(input_pad)
     # no need to print them
     # print(prediction)
     get_score_senti(label, prediction)
Пример #4
0
    def train(self, train_input_data, train_label, valid_input_data,
              valid_label):
        x_train = self.prepare_input(train_input_data)
        y_train = self.prepare_label(train_label)
        x_valid = self.prepare_input(valid_input_data)
        y_valid = self.prepare_label(valid_label)

        print('start training...')
        self.model.fit(x=x_train,
                       y=y_train,
                       batch_size=self.config.batch_size,
                       epochs=self.config.n_epochs,
                       validation_data=(x_valid, y_valid),
                       callbacks=self.callbacks)
        print('training end...')

        print('score over valid data:')
        valid_pred = self.model.predict(x_valid)
        get_score_senti(y_valid, valid_pred)
Пример #5
0
    def train(self, train_input_data, train_label, valid_input_data, valid_label,class_weights=None,imBalanced=False):
        print("LENGTH OF THIS STUPID input thing IS",len(train_input_data))
        x_train = self.prepare_input(train_input_data)
        y_train = self.prepare_label(train_label)
        # np.set_printoptions(threshold=np.inf)
        x_valid = self.prepare_input(valid_input_data)
        y_valid = self.prepare_label(valid_label)
        # if imBalanced is True:
        #     # print("Applying ImBalance Technique")
        #     # training_generator, steps_per_epoch = balanced_batch_generator(
        #     #     x_train, y_train, sampler=NearMiss(), batch_size=10, random_state=42)



        if imBalanced is False:
            print('start training (without ImBalance)...')
            history = self.model.fit(x=x_train, y=y_train, batch_size=self.config.batch_size, epochs=self.config.n_epochs,
                           validation_data=(x_valid, y_valid), callbacks=self.callbacks, class_weight=class_weights)
        else:
            print('start training (without ImBalance)...')
            # history = self.model.fit_generator(generator=training_generator,steps_per_epoch=steps_per_epoch, batch_size=self.config.batch_size, epochs=self.config.n_epochs,
            #    validation_data=(x_valid, y_valid), callbacks=self.callbacks, class_weight=class_weights)
            history = self.model.fit(x=(x_train[0],x_train[1]), y=y_train, batch_size=self.config.batch_size, epochs=self.config.n_epochs,
               validation_data=(x_valid, y_valid), callbacks=self.callbacks, class_weight=class_weights)


        


        plt.plot(history.history['acc'])
        plt.plot(history.history['val_acc'])
        plt.title('Model accuracy')
        plt.ylabel('Accuracy')
        plt.xlabel('Epoch')
        plt.legend(['Train', 'Test'], loc='upper left')
        plt.savefig('model_acc.png')
        print('training end...')

        print('score over valid data:')
        valid_pred = self.model.predict(x_valid)
        get_score_senti(y_valid, valid_pred)
Пример #6
0
 def on_epoch_end(self, epoch, logs={}):
     if len(self.validation_data[:-3]) == 1:
         x_valid = self.validation_data[0]
     else:
         x_valid = self.validation_data[:-3]
     y_valid = self.validation_data[-3]
     valid_results = self.model.predict(x_valid)
     _val_acc, _val_f1 = get_score_senti(y_valid, valid_results)
     logs['val_acc'] = _val_acc
     logs['val_f1'] = _val_f1
     self.val_accs.append(_val_acc)
     self.val_f1s.append(_val_f1)
     print('val_acc: %f' % _val_acc)
     print('val_f1: %f' % _val_f1)
     return
Пример #7
0
 def score(self, input_data, label):
     input_pad = self.prepare_input(input_data)
     label = self.prepare_label(label)
     prediction = self.model.predict(input_pad)
     get_score_senti(label, prediction)