from umbmid import get_proj_path, verify_path, get_script_logger
from umbmid.loadsave import save_pickle, save_mat, load_pickle
from umbmid.content import get_class_labels
from umbmid.ai.traintestsplit import split_to_train_test
###############################################################################
# Define the directory where the clean dataset is located
__DATA_DIR = os.path.join(get_proj_path(), 'datasets/gen-one/clean/')
###############################################################################
# Define the output directory where the train/test set files will
# be saved
__OUTPUT_DIR = os.path.join(get_proj_path(), 'datasets/gen-one/clean/')
verify_path(__OUTPUT_DIR)
###############################################################################
if __name__ == '__main__':
logger = get_script_logger(__file__) # Get logger
# Load the dataset to be split and its metadata
fd_data = load_pickle(os.path.join(__DATA_DIR, 'fd_data_s11_emp.pickle'))
metadata = load_pickle(os.path.join(__DATA_DIR, 'md_list_s11_emp.pickle'))
labels = get_class_labels(metadata) # Get the class labels
# Split into the train and test sets, and return the random seed
# NOTE: The random seed that was used to make the train/test sets
# that produced the results presented in the EuCAP2020 paper
def make_clean_files(gen='one',
cal_type='emp',
sparams='s11',
logger=null_logger):
"""Makes and saves the clean .mat and .pickle files
Parameters
----------
gen : str
The generation of data to be used, must be in ['one', 'two']
cal_type : str
The type of calibration to be performed, must be in
['emp', 'adi'
sparams : str
The type of sparam to save, must be in ['s11', 's21']
logger :
A logger for logging progress
"""
assert gen in ['one', 'two', 'three'], \
"Error: gen must be in ['one', 'two', 'three']"
assert sparams in ['s11', 's21'], \
"Error: sparams must be in ['s11', 's21']"
# Load the frequency-domain dataset
logger.info('\tImporting FD data and metadata...')
fd_data, fd_md = import_fd_cal_dataset(cal_type=cal_type,
prune=True,
gen=gen,
sparams=sparams,
logger=logger)
logger.info('\tImport complete. Saving to .pickle and .mat files...')
# Define an output dir for this generation of dataset
this_output_dir = os.path.join(__OUTPUT_DIR, 'gen-%s/clean/' % gen)
verify_path(this_output_dir) # Verify that this dir exists
logger.info('Num data samples:\t\t%s' % np.size(fd_data, axis=0))
logger.info('Length of metadata:\t\t%s' % len(fd_md))
# Save the frequency-domain data and metadata
save_pickle(
fd_md,
os.path.join(this_output_dir,
'md_list_%s_%s.pickle' % (sparams, cal_type)))
save_pickle(
fd_data,
os.path.join(this_output_dir,
'fd_data_%s_%s.pickle' % (sparams, cal_type)))
save_mat(
fd_data, 'fd_data_%s' % sparams,
os.path.join(this_output_dir,
'fd_data_%s_%s.mat' % (sparams, cal_type)))
save_mat(
fd_md, 'md_%s' % sparams,
os.path.join(this_output_dir,
'md_list_%s_%s.mat' % (sparams, cal_type)))
logger.info('\tComplete saving clean data files.')
University of Manitoba
October 21st, 2019
"""
import os
import numpy as np
from umbmid import get_proj_path, verify_path, get_script_logger
from umbmid.loadsave import save_pickle, save_mat
from umbmid.build import (import_fd_dataset, import_metadata,
import_metadata_df)
###############################################################################
__OUTPUT_DIR = os.path.join(get_proj_path(), 'datasets/')
verify_path(__OUTPUT_DIR)
__DATA_DIR = os.path.join(get_proj_path(), 'datasets/')
###############################################################################
# The possible sparams for each generation of dataset
possible_sparams = {
'one': ['s11'],
'two': ['s11', 's21'],
}
###############################################################################
if __name__ == '__main__':
University of Manitoba
October 21st, 2019
"""
import os
import numpy as np
from umbmid import get_proj_path, verify_path, get_script_logger
from umbmid.loadsave import save_pickle, save_mat
from umbmid.build import (import_fd_dataset, import_metadata,
import_metadata_df)
###############################################################################
__OUTPUT_DIR = os.path.join(get_proj_path(), 'datasets/')
verify_path(__OUTPUT_DIR)
__DATA_DIR = os.path.join(get_proj_path(), 'datasets/')
###############################################################################
# The possible sparams for each generation of dataset
possible_sparams = {
'one': ['s11'],
'two': ['s11', 's21'],
'three': ['s11', 's21'],
}
###############################################################################
if __name__ == '__main__':