#!/usr/bin/env python
import os, sys
import setlog
conf_file = os.environ['DEV'] + 'dl_management/.log/logging.yaml'
save_file = os.path.abspath(sys.argv[0])[:-len(sys.argv[0])] + 'log/'
setlog.reconfigure(conf_file, save_file)
import system.DescriptorLearning as System
if __name__ == '__main__':
machine = System.MultNet(root=os.path.abspath(sys.argv[0])[:-len(sys.argv[0])],
cnn_type='cnn.yaml',
trainer_file='../trainer_mult_mod_pca.yaml',
dataset_file='../../../SummerTests/datasets/cmu_lt.yaml')
machine.compute_PCA(2048, desc=['desc_no_pca'])
#!/usr/bin/env python
import os, sys
import setlog
conf_file = os.environ['DEV'] + 'dl_management/.log/logging.yaml'
save_file = os.path.abspath(sys.argv[0])[:-len(sys.argv[0])] + 'log/'
setlog.reconfigure(conf_file, save_file)
import system.DescriptorLearning as System
if __name__ == '__main__':
machine = System.MultNet(
root=os.path.abspath(sys.argv[0])[:-len(sys.argv[0])],
cnn_type='cnn.yaml',
dataset_file='../../../../../datasets/default.yaml')
action = input(
'Exec:\n[t]\ttrain\n[e]\ttest\n[p]\tprint (console)\n[P]\tprint (full)\n[ ]\ttrain+test\n'
)
if action == 't':
machine.train()
elif action == 'e':
machine.test()
machine.plot(print_loss=False, print_val=False)
elif action == 'p':
machine.plot(print_loss=False, print_val=False)
elif action == 'P':
machine.plot()
elif action == '':
machine.train()
machine.test()
machine.plot(print_loss=False, print_val=False)
#!/usr/bin/env python
import os, sys
import setlog
conf_file = os.environ['DEV'] + 'dl_management/.log/logging.yaml'
save_file = os.path.abspath(sys.argv[0])[:-len(sys.argv[0])] + 'log/'
setlog.reconfigure(conf_file, save_file)
import system.DescriptorLearning as System
if __name__ == '__main__':
print(os.getcwd())
machine = System.MultNet(
root=os.path.abspath(sys.argv[0])[:-len(sys.argv[0])],
dataset_file='../../../../datasets/cmu_training.yaml',
trainer_file='../trainer.yaml')
action = input(
'Exec:\n[t]\ttrain\n[e]\ttest\n[p]\tprint (console)\n[P]\tprint (full)\n[ ]\ttrain+test\n'
)
if action == 't':
machine.train()
elif action == 'e':
machine.test()
machine.plot(print_loss=False, print_val=False)
elif action == 'p':
machine.plot(print_loss=False, print_val=False)
elif action == 'P':
machine.plot()
elif action == '':
machine.train()
machine.test()
#!/usr/bin/env python
import os, sys
import torch
import setlog
conf_file = os.environ['DEV'] + 'dl_management/.log/logging.yaml'
save_file = os.path.abspath(sys.argv[0])[:-len(sys.argv[0])] + 'log/'
setlog.reconfigure(conf_file, save_file)
import system.DescriptorLearning as System
if __name__ == '__main__':
machine = System.Default(
root=os.path.abspath(sys.argv[0])[:-len(sys.argv[0])])
action = input('''
Exec:
[t]\ttrain
[e]\ttest
[p]\tprint (console)
[P]\tprint (full)
[s]\tserialize net
[c]\tcreat clusters
[ ]\ttrain+test
''')
if action == 't':
machine.train()
elif action == 'e':
machine.test()
machine.plot(print_loss=False, print_val=False)
elif action == 'p':
machine.plot(print_loss=False, print_val=False)
#!/usr/bin/env python
import os, sys
import setlog
conf_file = os.environ['DEV'] + 'dl_management/.log/logging.yaml'
save_file = os.path.abspath(sys.argv[0])[:-len(sys.argv[0])] + 'log/'
setlog.reconfigure(conf_file, save_file)
import system.DescriptorLearning as System
if __name__ == '__main__':
machine = System.Default(root=os.path.abspath(
sys.argv[0])[:-len(sys.argv[0])],
dataset_file='../../../../datasets/default.yaml',
trainer_file='../trainer.yaml')
action = input(
'Exec:\n[t]\ttrain\n[e]\ttest\n[p]\tprint (console)\n[P]\tprint (full)\n[ ]\ttrain+test\n'
)
if action == 't':
machine.train()
elif action == 'e':
machine.test()
machine.plot(print_loss=False, print_val=False)
elif action == 'p':
machine.plot(print_loss=False, print_val=False)
elif action == 'P':
machine.plot()
elif action == '':
machine.train()
machine.test()
machine.plot(print_loss=False, print_val=False)
#!/usr/bin/env python
import os, sys
import setlog
conf_file = os.environ['DEV'] + 'dl_management/.log/logging.yaml'
save_file = os.path.abspath(sys.argv[0])[:-len(sys.argv[0])] + 'log/'
setlog.reconfigure(conf_file, save_file)
import system.DescriptorLearning as System
if __name__ == '__main__':
machine = System.MultNet(root=os.path.abspath(
sys.argv[0])[:-len(sys.argv[0])],
dataset_file='sparse_dataset.yaml')
action = input(
'Exec:\n[t]\ttrain\n[e]\ttest\n[p]\tprint (console)\n[P]\tprint (full)\n[ ]\ttrain+test\n'
)
if action == 't':
machine.train()
elif action == 'e':
machine.test()
machine.plot(print_loss=False, print_val=False)
elif action == 'p':
machine.plot(print_loss=False, print_val=False)
elif action == 'P':
machine.plot()
elif action == '':
machine.train()
machine.test()
machine.plot(print_loss=False, print_val=False)
elif action == 's':
#!/usr/bin/env python
import os, sys
import setlog
conf_file = os.environ['DEV'] + 'dl_management/.log/logging.yaml'
save_file = os.path.abspath(sys.argv[0])[:-len(sys.argv[0])] + 'log/'
setlog.reconfigure(conf_file, save_file)
import system.DescriptorLearning as System
if __name__ == '__main__':
machine = System.MultNet(
root=os.path.abspath(sys.argv[0])[:-len(sys.argv[0])],
cnn_type='cnn.yaml',
trainer_file='../trainer_only_depth.yaml',
dataset_file='../../../SummerTests/datasets/night_ft.yaml')
action = input(
'Exec:\n[t]\ttrain\n[e]\ttest\n[p]\tprint (console)\n[P]\tprint (full)\n[ ]\ttrain+test\n'
)
if action == 't':
machine.train()
elif action == 'e':
machine.test()
machine.plot(print_loss=False, print_val=False)
elif action == 'p':
machine.plot(print_loss=False, print_val=False)
elif action == 'P':
machine.plot()
elif action == '':
machine.train()
machine.test()