from ai4materials.dataprocessing.preprocessing import prepare_dataset_STEM ### YBC #from ai4materials.descriptors.diffraction2d import Diffraction2D from ai4materials.utils.utils_config import set_configs from ai4materials.utils.utils_config import setup_logger from ai4materials.utils.utils_crystals import create_supercell from ai4materials.utils.utils_crystals import create_vacancies from ai4materials.wrappers import calc_descriptor from ai4materials.wrappers import load_descriptor import os.path from sklearn import preprocessing import numpy as np from PIL import Image # set configs configs = set_configs(main_folder='./') logger = setup_logger(configs, level='INFO', display_configs=False) # setup folder and files dataset_folder = os.path.join(configs['io']['main_folder'], 'my_datasets') desc_file_name = 'fcc_bcc_hcp_example' # calculate the descriptor for the list of structures images_list = [] targets_list = [] f_list = open("list_shuf_test.txt", 'r') while True: line = f_list.readline() if not line: break
from ai4materials.utils.utils_config import set_configs from ai4materials.utils.utils_config import setup_logger from ai4materials.utils.utils_data_retrieval import read_ase_db from ai4materials.wrappers import load_descriptor from ai4materials.wrappers import calc_model from ai4materials.wrappers import calc_descriptor from ai4materials.descriptors.atomic_features import AtomicFeatures from ai4materials.descriptors.atomic_features import get_table_atomic_features from ai4materials.utils.utils_config import get_data_filename from ai4materials.visualization.viewer import read_control_file import numpy as np import pandas as pd # modify this path if you want to save the calculation results in another location configs = set_configs(main_folder='./l1_l0_example') logger = setup_logger(configs, level='INFO') # setup folder and files lookup_file = os.path.join(configs['io']['main_folder'], 'lookup.dat') materials_map_plot_file = os.path.join(configs['io']['main_folder'], 'binaries_l1_l0_map_prl2015.png') # define descriptor - atomic features in this case kwargs = {'energy_unit': 'eV', 'length_unit': 'angstrom'} descriptor = AtomicFeatures(configs=configs, **kwargs) # ============================================================================= # Descriptor calculation # ============================================================================= desc_file_name = 'atomic_features_binaries'
from functools import partial from ai4materials.utils.utils_config import set_configs from ai4materials.dataprocessing.preprocessing import load_dataset_from_file from ai4materials.models.cnn_architectures import cnn_nature_comm_ziletti2018 from ai4materials.models.cnn_architectures import cnn_architecture_ai4STEM # YBC from ai4materials.models.cnn_nature_comm_ziletti2018 import load_datasets from ai4materials.models.STEM_CNN_segmentation import train_neural_network #YBC from ai4materials.utils.utils_config import setup_logger from sklearn import preprocessing import numpy as np import os configs = set_configs() logger = setup_logger(configs, level='DEBUG', display_configs=False) #dataset_folder = configs['io']['main_folder'] dataset_folder = os.path.join(configs['io']['main_folder'], 'my_datasets') # ============================================================================= # Download the dataset from the online repository and load it # ============================================================================= #x_pristine, y_pristine, dataset_info_pristine, x_vac25, y_vac25, dataset_info_vac25 = load_datasets(dataset_folder) train_set_name = 'STEM_monocrystalline_train' path_to_x_pristine = os.path.join(dataset_folder, train_set_name + '_x.pkl') path_to_y_pristine = os.path.join(dataset_folder, train_set_name + '_y.pkl') path_to_summary_pristine = os.path.join(dataset_folder, train_set_name + '_summary.json') test_set_name = 'STEM_monocrystalline_test' path_to_x_vac25 = os.path.join(dataset_folder, test_set_name + '_x.pkl') path_to_y_vac25 = os.path.join(dataset_folder, test_set_name + '_y.pkl')
def test_setup_logger(self): logger = setup_logger(configs=None, level=None, display_configs=False) self.assertIsInstance(logger, logging.Logger)