def run_instance(instance,
sparsity,
trade_off,
learning_rate,
max_iter=1000,
epsilon=1e-3,
logger=None):
opt_x_array, _ = optimize(instance, sparsity, trade_off, learning_rate,
max_iter, epsilon)
raw_pred_subgraphs = []
for opt_x in opt_x_array:
pred_subgraph = np.nonzero(opt_x)[0]
raw_pred_subgraphs.append(pred_subgraph)
global_prec, global_rec, global_fm, global_iou, valid_global_prec, valid_global_rec, valid_global_fm, valid_global_iou, _, _, _, _, _ = evaluate_evo(
instance['true_subgraphs'], raw_pred_subgraphs)
logger.debug('-' * 5 + 'performance in the whole interval' + '-' * 5)
logger.debug('global precision: {:.5f}'.format(global_prec))
logger.debug('global recall : {:.5f}'.format(global_rec))
logger.debug('global f-measure: {:.5f}'.format(global_fm))
logger.debug('global iou : {:.5f}'.format(global_iou))
logger.debug('-' * 5 + 'performance in the interval with signals' +
'-' * 5)
logger.debug('global precision: {:.5f}'.format(valid_global_prec))
logger.debug('global recall : {:.5f}'.format(valid_global_rec))
logger.debug('global f-measure: {:.5f}'.format(valid_global_fm))
logger.debug('global iou : {:.5f}'.format(valid_global_iou))
refined_pred_subgraphs = post_process_evo(instance['graph'],
raw_pred_subgraphs)
global_prec, global_rec, global_fm, global_iou, valid_global_prec, valid_global_rec, valid_global_fm, valid_global_iou, _, _, _, _, _ = evaluate_evo(
instance['true_subgraphs'], refined_pred_subgraphs)
logger.debug('-' * 5 + ' refined performance ' + '-' * 5)
logger.debug('refined global precision: {:.5f}'.format(global_prec))
logger.debug('refined global recall : {:.5f}'.format(global_rec))
logger.debug('refined global f-measure: {:.5f}'.format(global_fm))
logger.debug('refined global iou : {:.5f}'.format(global_iou))
logger.debug('-' * 5 + 'refined performance in the interval with signals' +
'-' * 5)
logger.debug('global precision: {:.5f}'.format(valid_global_prec))
logger.debug('global recall : {:.5f}'.format(valid_global_rec))
logger.debug('global f-measure: {:.5f}'.format(valid_global_fm))
logger.debug('global iou : {:.5f}'.format(valid_global_iou))
def run_dataset(paras):
dataset, sparsity, trade_off, learning_rate, max_iter, epsilon, write_to_dir, data_type = paras
if not write_to_dir:
logger = logging.getLogger('fei')
else:
log_fn = '{}_sparsity_{:d}_trade_{}_lr_{}_{}.txt'.format(
DATASET, sparsity, trade_off, learning_rate, data_type)
if os.path.isfile(os.path.join(write_to_dir, log_fn)):
print('file exist !!!')
return
logger = logging.getLogger(log_fn)
formatter = logging.Formatter('')
file_handler = logging.FileHandler(filename=os.path.join(
write_to_dir, log_fn),
mode='w')
file_handler.setFormatter(formatter)
logger.addHandler(file_handler)
all_performance = []
logger.debug('-' * 5 + ' setting ' + '-' * 5)
logger.debug('sparsity: {:d}'.format(sparsity))
logger.debug('learning rate: {:.5f}'.format(learning_rate))
logger.debug('trade off: {:.5f}'.format(trade_off))
for i, instance in enumerate(dataset):
logger.debug('instance: {:d}'.format(i))
opt_x_array, run_time = optimize(instance,
sparsity,
trade_off,
learning_rate,
max_iter,
epsilon,
logger=None)
logger.debug('run time: {:.5f}'.format(run_time))
raw_pred_subgraphs = []
for opt_x in opt_x_array:
pred_subgraph = np.nonzero(opt_x)[0]
print(sorted(pred_subgraph))
raw_pred_subgraphs.append(pred_subgraph)
global_prec, global_rec, global_fm, global_iou, valid_global_prec, valid_global_rec, valid_global_fm, valid_global_iou, _, _, _, _, _ = evaluate_evo(
instance['subgraphs'], raw_pred_subgraphs)
logger.debug('-' * 5 + ' performance in the whole interval ' + '-' * 5)
logger.debug('global precision: {:.5f}'.format(global_prec))
logger.debug('global recall : {:.5f}'.format(global_rec))
logger.debug('global f-measure: {:.5f}'.format(global_fm))
logger.debug('global iou : {:.5f}'.format(global_iou))
logger.debug('-' * 5 + ' performance in the interval with signals ' +
'-' * 5)
logger.debug('global precision: {:.5f}'.format(valid_global_prec))
logger.debug('global recall : {:.5f}'.format(valid_global_rec))
logger.debug('global f-measure: {:.5f}'.format(valid_global_fm))
logger.debug('global iou : {:.5f}'.format(valid_global_iou))
refined_pred_subgraphs = post_process_evo(instance['graph'],
raw_pred_subgraphs,
dataset=DATASET)
refined_global_prec, refined_global_rec, refined_global_fm, refined_global_iou, _, _, _, _, _, _, _, _, _ = evaluate_evo(
instance['subgraphs'], refined_pred_subgraphs)
logger.debug('-' * 5 + ' refined performance ' + '-' * 5)
logger.debug(
'refined global precision: {:.5f}'.format(refined_global_prec))
logger.debug(
'refined global recall : {:.5f}'.format(refined_global_rec))
logger.debug(
'refined global f-measure: {:.5f}'.format(refined_global_fm))
logger.debug(
'refined global iou : {:.5f}'.format(refined_global_iou))
all_performance.append(
(global_prec, global_rec, global_fm, global_iou,
refined_global_prec, refined_global_rec, refined_global_fm,
refined_global_iou, run_time))
all_performance = np.array(all_performance)
avg_performance = np.mean(all_performance, axis=0)
# print(all_performance.shape)
logger.debug('-' * 5 + ' average performance ' + '-' * 5)
logger.debug('average presision: {:.5f}'.format(avg_performance[0]))
logger.debug('average recall : {:.5f}'.format(avg_performance[1]))
logger.debug('average f-measure: {:.5f}'.format(avg_performance[2]))
logger.debug('average iou : {:.5f}'.format(avg_performance[3]))
logger.debug('avg refined prec : {:.5f}'.format(avg_performance[4]))
logger.debug('avg refined rec : {:.5f}'.format(avg_performance[5]))
logger.debug('avg refined fm : {:.5f}'.format(avg_performance[6]))
logger.debug('avg refined iou : {:.5f}'.format(avg_performance[7]))
logger.debug('average run time : {:.5f}'.format(avg_performance[8]))