def test(batch_size,img_size,test_images_path,model_path):
test_generator = helpers.generate_test_data(batch_size,img_size,preprocess_input,test_images_path)
model = helpers.get_model(model_path)
yhat = helpers.predict(model,test_generator)
yhat = helpers.get_best_guess(yhat)
acc_nnet_M = helpers.get_accuracy(yhat,test_generator)
class_mapping = helpers.get_class_mapping(test_generator)
final_pred = helpers.final_prediction(class_mapping,yhat)
df_nnet_M = helpers.create_dataframe(['image','class','nnet_M'])
df_nnet_M = helpers.add_records(df_nnet_M,test_generator,final_pred,'nnet_M')
return df_nnet_M,acc_nnet_M
print('Getting model...')
result = None
with open(os.path.join(model_path, 'train.pickle'), 'rb') as f:
result = pickle.load(f)
label_probs = result[0]
probs_per_label = result[1]
words = result[2]
labels = result[3]
predictor = partial(helpers.predict,
label_probs, probs_per_label, words, labels)
if os.path.isdir(input_path):
print('Loading dataset...')
test_target, test_data = helpers.load_dataset(input_path)
print('Testing dataset...')
print("============= RESULT ===============")
accuracy = helpers.get_accuracy(test_data, test_target, predictor)
print('Accuracy: {0:.2f}%'.format(accuracy * 100))
print('=====================================')
else:
print('Testing...')
label, tokens = helpers.load_file(input_path)
predict_label = predictor(tokens)
print('================= RESULT ================')
print('Expected label for the text: {}'.format(label))
print('Predicted label for the text: {}'.format(predict_label))
print('==========================================')