maxVal = max(target) # we still need to implement validation so 0% is used for validation # 70% of the data is used for training and 30% for testing here segmented_input = parser.segmentation(data, 0.8, 0.0, 0.2) segmented_target = parser.segmentation(target, 0.8, 0.0, 0.2) train_input = segmented_input[0] train_target = segmented_target[0] train_size = len(train_input) train_input = np.array(train_input) train_target = np.array(train_target) train_target = train_target.reshape(train_size, 1) ## Find Optimal NN setup ## validation.k_folds(3, train_input, train_target, feature_value_range, minVal, maxVal) # test_input = segmented_input[2] # test_target = segmented_target[2] # train_size = len(train_input) # test_size = len(test_input) # # we need to convert the format of the data to be # # compliant with the neurolab API, print out the # # values of inp and tar to see format # train_input = np.array(train_input) # train_target = np.array(train_target) # train_target = train_target.reshape(train_size, 1) # # Create network with 3 layers with 5, 5, and 1 neruon(s) in each layer
