if __name__ == "__main__":
if not os.path.exists('./results'):
os.makedirs('./results')
print("Train MTMC with LS-ABS and hybrid-inv")
model = MTMC(data_folder)
ioa = OptAlg(alg_type='LS-ABS', direction='hybrid-inv')
res = {'time': [], 'LL': [], 'epochs': []}
for i in range(20):
tmp = model.optimize(
ioa, **{
'verbose': False,
'max_epochs': 1000,
'batch': 1000
})
res['time'].append(tmp['opti_time'])
res['LL'].append(tmp['fun'])
res['epochs'].append(tmp['nep'])
with open('results/LS-ABS_hybrid-inv.json', 'w') as outfile:
json.dump(res, outfile)
print("{}/20 done!".format(i + 1))
import sys
sys.path.append("../..")
import os
from algos import OptAlg
from models import MTMC
data_folder = '../../../data/'
if __name__ == "__main__":
if not os.path.exists('./results'):
os.makedirs('./results')
print("Train MTMC_MNL with HAMABS")
model = MTMC(data_folder)
ioa = OptAlg(alg_type='LS-ABS', direction='hybrid-inv')
tmp = model.optimize(
ioa, **{
'verbose': True,
'nbr_epochs': 1000,
'batch': 1000
})