add testing script

test.py

@@ -0,0 +1,18 @@
+import keras.backend as K
+from model import SiameseNet
+from data_loader import SiameseImageLoader
+import matplotlib.pyplot as plt
+from keras import optimizers
+from keras.models import Model, load_model
+import albumentations as A
+import cv2
+import time
+
+
+model = SiameseNet('configs/road_signs.yml')
+model.load_model('{}best_model_4.h5'.format(model.weights_save_path))
+model.load_encodings('{}encodings.pkl'.format(model.encodings_path))
+
+
+model_accuracy = model.calculate_prediction_accuracy()
+print('Model accuracy on validation set: {}'.format(model_accuracy))

train.py

@@ -5,14 +5,12 @@
 from keras.callbacks import EarlyStopping, ReduceLROnPlateau, ModelCheckpoint
 
 
-n_epochs = 5
-n_steps_per_epoch = 6
-batch_size = 1
+n_epochs = 1000
+n_steps_per_epoch = 600
+batch_size = 8
 val_steps = 100
 
-# model = SiameseNet('configs/road_signs.yml')
-model = SiameseNet('configs/plates.yml')
-
+model = SiameseNet('configs/road_signs.yml')
 
 initial_lr = 1e-4
 decay_factor = 0.99
@@ -23,7 +21,7 @@
                           decay_factor ** np.floor(x/step_size)),
     EarlyStopping(patience=50, verbose=1),
     TensorBoard(log_dir=model.tensorboard_log_path),
-    ModelCheckpoint(filepath=os.path.join(model.weights_save_path, 'best_model.h5'),
+    ModelCheckpoint(filepath=os.path.join(model.weights_save_path, 'best_model_4.h5'),
                     verbose=1, monitor='val_loss', save_best_only=True)
 ]
 
@@ -32,7 +30,6 @@
 
 
 model.generate_encodings()
-# model.load_encodings('encodings/encodings.pkl')
 
 model_accuracy = model.calculate_prediction_accuracy()
 print('Model accuracy on validation set: {}'.format(model_accuracy))