test_acc_array[j] = test_results[1]
predictions = np.empty((len(test_data), 2))
predictions_output = model.predict(test_data)
for k in range(0, len(predictions_output)):
if predictions_output[k][0] > predictions_output[k][1]:
predictions[k] = np.array((1, 0))
else:
predictions[k] = np.array((0, 1))
print(
confusion_matrix(test_labels.argmax(axis=1),
predictions.argmax(axis=1)))
max_test_array[i][count - 1] = np.max(test_acc_array)
count = count + 1
model.load_weights("weights.h5")
# In[ ]:
#plt.plot(train_acc_list)
#plt.plot(test_acc_list)
#plt.title('VGG-16-5 Fold Cross validation-model accuracy')
#plt.ylabel('accuracy')
#plt.xlabel('epoch')
#plt.legend(['train', 'test'], loc='upper left')
#plt.gcf().savefig("Results/VGG-16-5FCV.png")
target_size=(image_size, image_size),
batch_size=1,
shuffle=False)
x = model_list[model[count]].output
x = Dropout(0.25, name='do_akhir')(x)
predictions = Dense(num_classes, activation='softmax')(x)
loaded_model = Model(inputs=model_list[model[count]].input,
outputs=predictions)
loaded_model.compile(
Adam(lr=0.01),
loss='categorical_crossentropy',
metrics=[categorical_accuracy, top_2_accuracy, top_3_accuracy])
loaded_model.load_weights(file)
predictions = loaded_model.predict_generator(test_batches,
steps=num_val_samples,
verbose=1)
test_labels = test_batches.classes
cm = confusion_matrix(test_labels, predictions.argmax(axis=1))
plot_confusion_matrix(cm,
cm_plot_labels,
base_dir=base_dir,
title='Confusion Matrix',
name=model[count])
y_pred = np.argmax(predictions, axis=1)
y_true = test_batches.classes
report = classification_report(y_true, y_pred, target_names=cm_plot_labels)
print(report)
f.write(report)
f.close()
count += 1
plt.xlabel("Epochs")
plt.ylabel("Loss")
plt.legend()
plt.savefig("loss.png")
plt.figure()
plt.plot(epochs, acc, 'g', label='Training acc')
plt.plot(epochs, val_acc, 'b', label='Validation acc')
plt.xlabel("Epochs")
plt.ylabel("Accuracy %")
plt.title('Training and validation cat accuracy')
plt.legend()
plt.figure()
plt.savefig("accuracy.png")
plt.show()
# Confusion Matrix
predictions = models.predict_generator(test_gen, steps=len(df_val), verbose=1)
cm = confusion_matrix(test_gen.classes, predictions.argmax(axis=1))
plot_confusion_matrix(cm, ['0', '1'])
# Classification Report
y_pred = np.argmax(predictions, axis=1)
y_true = test_gen.classes
report = classification_report(y_true, y_pred, target_names=['0', '1'])
print(report)
#save model
model_json = models.to_json()
open('CNN_Model.json', 'w').write(model_json)
