config_train_baseline.py

from utils_e2e_clean import get_scope_name_of_train_op

max_epochs = int(12)
steps_per_eval = int(600)

infer_beam_width = 5
infer_max_decoding_length = 50

train = {
    'joint': {
        'optimizer': {
            'type': 'AdamOptimizer',
            'kwargs': {
                'learning_rate': 1e-3
            }
        },
        'gradient_clip': {
            'type': 'clip_by_global_norm',
            'kwargs': {
                'clip_norm': 15
            }
        },
    },
    'disc': {
        'optimizer': {
            'type': 'AdamOptimizer',
            'kwargs': {
                'learning_rate': 1e-3
            }
        },
        'gradient_clip': {
            'type': 'clip_by_global_norm',
            'kwargs': {
                'clip_norm': 15
            }
        },
    },
    'align': {
        'optimizer': {
            'type': 'AdamOptimizer',
            'kwargs': {
                'learning_rate': 1e-3
            }
        },
        'gradient_clip': {
            'type': 'clip_by_global_norm',
            'kwargs': {
                'clip_norm': 15
            }
        },
    },
}

for name, hparams in train.items():
    hparams['name'] = get_scope_name_of_train_op(name)