remove plt

Author: victordibia

metrics/ae/histogram.png

@@ -1 +1 @@
-{"acc": 0.961, "precision": 0.8536906854130053, "recall": 0.9715, "f1": 0.9087932647333957, "f2": 0.9454067730634488, "roc": 0.9560253750000001, "threshold": 0.053}
+{"acc": 0.8462, "precision": 0.697098976109215, "recall": 0.4085, "f1": 0.5151324085750315, "f2": 0.4453772350632359, "roc": 0.955555625, "threshold": 0.056}

metrics/ae/metrics.json

@@ -20,7 +20,6 @@
 
 import os
 from utils import train_utils
-import matplotlib.pyplot as plt
 
 import numpy as np
 import random
@@ -134,16 +133,7 @@ def train(self, in_train, in_val):
         kwargs['verbose'] = 1
         kwargs['callbacks'] = [train_utils.TimeHistory()]
 
-        history = self.model.fit(X_train, X_trains, **kwargs)
-        # Plot training & validation loss values
-        # plt.plot(history.history['loss'])
-        # plt.plot(history.history['val_loss'])
-        # plt.title('Model loss')
-        # plt.ylabel('Loss')
-        # plt.xlabel('Epoch')
-        # plt.legend(['Train', 'Test'], loc='upper left')
-        # # plt.show()
-        # plt.close()
+        history = self.model.fit(X_train, X_train, **kwargs)
 
     def compute_anomaly_score(self, df):
         preds = self.model.predict(df)

metrics/ae/metrics.png

@@ -19,7 +19,6 @@
 from tensorflow.keras.layers import LeakyReLU, ReLU
 import os
 
-import matplotlib.pyplot as plt
 
 import numpy as np
 import random

metrics/ae/roc.png

@@ -16,7 +16,6 @@
 from tensorflow.keras.optimizers import Adam
 
 from utils import train_utils
-import matplotlib.pyplot as plt
 
 import numpy as np
 import random
@@ -169,12 +168,12 @@ def train(self, in_train, in_test):
         history = self.model.fit(
             [encoder_input_data, decoder_input_data], decoder_target_data, **kwargs)
         # Plot training & validation loss values
-        plt.plot(history.history['loss'])
-        plt.plot(history.history['val_loss'])
-        plt.title('Model loss')
-        plt.ylabel('Loss')
-        plt.xlabel('Epoch')
-        plt.legend(['Train loss', 'Val Loss'], loc='upper left')
+        # plt.plot(history.history['loss'])
+        # plt.plot(history.history['val_loss'])
+        # plt.title('Model loss')
+        # plt.ylabel('Loss')
+        # plt.xlabel('Epoch')
+        # plt.legend(['Train loss', 'Val Loss'], loc='upper left')
         # plt.show()
 
     def decode_sequence(self, input_seq):

models/ae.py

@@ -19,7 +19,7 @@
 
 
 import logging
-logging.basicConfig(level=logging.DEBUG)
+logging.basicConfig(level=logging.INFO)
 
 
 parser = argparse.ArgumentParser(description='Process train parameters')

models/bigan.py

@@ -7,17 +7,18 @@
 # =============================================================================
 #
 
+from multiprocessing import Queue, Pool
+import json
+import os
+import logging
 import numpy as np
 from sklearn.metrics import precision_recall_fscore_support
 from sklearn.metrics import confusion_matrix, accuracy_score, f1_score, precision_score, recall_score, fbeta_score, roc_curve, auc, roc_auc_score
 import pandas as pd
-import matplotlib.pyplot as plt
-import logging
-import os
-import json
-
-from multiprocessing import Queue, Pool
+import matplotlib
+matplotlib.use('Agg')
 
+plt = matplotlib.pyplot
 fig_size = (9, 6)
 fig_font = 15
 plot_params = {'legend.fontsize': 'large',