Python train_convlstm 예제들

예제 #1

#               filters = [64, 64], kernels = [3, 3],
#               drop_remainder_batch = True, overwrite_results = True)

#train_convlstm(num_vars = 4, seq_length = 6, epochs = 40, batch_size = 5,
#               filters = [32, 32, 32], kernels = [3, 3, 3],
#               drop_remainder_batch = True, overwrite_results = True)

#train_convlstm(num_vars = 4, seq_length = 6, epochs = 40, batch_size = 5,
#               filters = [64, 32], kernels = [5, 5],
#               drop_remainder_batch = True, overwrite_results = True)

# converged models.
train_convlstm(num_vars=4,
               seq_length=24,
               epochs=1,
               batch_size=10,
               filters=[128],
               kernels=[3],
               drop_remainder_batch=True,
               overwrite_results=True)

#train_convlstm(num_vars = 4, seq_length = 24, epochs = 40, batch_size = 10,
#               filters = [128, 32], kernels = [3, 3],
#               drop_remainder_batch = True, overwrite_results = True)

#train_convlstm(num_vars = 4, seq_length = 24, epochs = 40, batch_size = 10,
#               filters = [16, 16], kernels = [3, 3],
#               drop_remainder_batch = True, overwrite_results = True)

#train_convlstm(num_vars = 4, seq_length = 24, epochs = 40, batch_size = 10,
#               filters = [8, 8, 8], kernels = [3, 3, 3],
#               drop_remainder_batch = True, overwrite_results = True)

예제 #2

from base_config import train_convlstm

# NON converging models
#train_convlstm(num_vars = 4, seq_length = 6, epochs = 40, batch_size = 5,
#               filters = [256, 256], kernels = [3, 3],
#               drop_remainder_batch = True, overwrite_results = True)

#train_convlstm(num_vars = 4, seq_length = 6, epochs = 40, batch_size = 5,
#               filters = [64, 64], kernels = [3, 3],
#               drop_remainder_batch = True, overwrite_results = True)

train_convlstm(num_vars=4,
               seq_length=6,
               epochs=1,
               batch_size=5,
               filters=[32, 32, 32],
               kernels=[3, 3, 3],
               drop_remainder_batch=True,
               overwrite_results=True)

#train_convlstm(num_vars = 4, seq_length = 6, epochs = 40, batch_size = 5,
#               filters = [64, 32], kernels = [5, 5],
#               drop_remainder_batch = True, overwrite_results = True)

# converged models.
#train_convlstm(num_vars = 4, seq_length = 24, epochs = 40, batch_size = 10,
#               filters = [128], kernels = [3],
#               drop_remainder_batch = True, overwrite_results = True)

#train_convlstm(num_vars = 4, seq_length = 24, epochs = 40, batch_size = 10,
#               filters = [128, 32], kernels = [3, 3],

예제 #3

파일: config_convlstm_models.py 프로젝트: hannasv/MS

from base_config import train_convlstm

# NON converging models
train_convlstm(num_vars=4,
               seq_length=6,
               epochs=40,
               batch_size=5,
               filters=[256, 256],
               kernels=[3, 3],
               drop_remainder_batch=True,
               overwrite_results=True)

train_convlstm(num_vars=4,
               seq_length=6,
               epochs=40,
               batch_size=5,
               filters=[64, 64],
               kernels=[3, 3],
               drop_remainder_batch=True,
               overwrite_results=True)

train_convlstm(num_vars=4,
               seq_length=6,
               epochs=40,
               batch_size=5,
               filters=[32, 32, 32],
               kernels=[3, 3, 3],
               drop_remainder_batch=True,
               overwrite_results=True)

train_convlstm(num_vars=4,