|
- #2022 Huawei Technologies Co., Ltd
- #
- # Licensed under the Apache License, Version 2.0 (the "License");
- # you may not use this file except in compliance with the License.
- # You may obtain a copy of the License at
- #
- # http://www.apache.org/licenses/LICENSE-2.0
- #
- # Unless required by applicable law or agreed to in writing, software
- # distributed under the License is distributed on an "AS IS" BASIS,
- # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- # See the License for the specific language governing permissions and
- # limitations under the License.
- # ============================================================================
-
- import argparse
- import os
- from preprocess import preprocess
- from data_loader import DatasetGenerator
- from lr_sch import dynamic_lr
- from network_define import WithLossCell
- from model import DPTNet_base
- from loss import Loss
- from mindspore import Model
- from mindspore import context
- from mindspore import load_checkpoint, load_param_into_net
- from mindspore import nn
- from mindspore.communication.management import init, get_rank, get_group_size
- from mindspore.context import ParallelMode
- from mindspore.train.callback import LossMonitor, TimeMonitor, ModelCheckpoint, CheckpointConfig
- import mindspore.dataset as ds
-
-
-
- parser = argparse.ArgumentParser(
- "Dual-path transformer"
- "with Permutation Invariant Training")
- # General config
- # Task related
- parser.add_argument('--train_dir', type=str, default="/home/work/user-job-dir/inputs/data_json/tr",
- help='directory including mix.json, s1.json and s2.json')
- parser.add_argument('--valid_dir', type=str, default='/mass_data/dataset/LS-2mix/Libri2Mix/cv',
- help='directory including mix.json, s1.json and s2.json')
- parser.add_argument('--sample_rate', default=8000, type=int,
- help='Sample rate')
- parser.add_argument('--segment', default=4, type=float,
- help='Segment length (seconds)')
- # Network architecture
- parser.add_argument('--enc_dim', default=256, type=int,
- help='...')
- parser.add_argument('--feature_dim', default=64, type=int,
- help='Number of filters in autoencoder')
- parser.add_argument('--hidden_dim', default=128, type=int,
- help='...')
- parser.add_argument('--layer', default=6, type=int,
- help='Number of repeats')
- parser.add_argument('--segment_size', default=250, type=int,
- help='segment size')
- parser.add_argument('--nspk', default=2, type=int,
- help='Maximum number of speakers')
- parser.add_argument('--win_len', default=2, type=int,
- help='...')
- # Training config
- parser.add_argument('--epochs', default=100, type=int,
- help='Number of maximum epochs')
- parser.add_argument('--device_num', default=2, type=int,
- help='device num')
- parser.add_argument('--device_id', default=2, type=int,
- help='device id')
- # minibatch
- parser.add_argument('--batch_size', default=3, type=int,
- help='Batch size')
- # optimizer
- parser.add_argument('--lr', default=5e-6, type=float,
- help='Init learning rate')
- parser.add_argument('--l2', default=0.0, type=float,
- help='weight decay (L2 penalty)')
- # save and load model
- parser.add_argument('--save_folder', default='exp/temp',
- help='Location to save epoch models')
- parser.add_argument('--continue_train', default=0, type=int,
- help='Continue from checkpoint model')
- parser.add_argument('--step_per_epoch', default=7120, type=int,
- help='...')
- parser.add_argument('--ckpt_path', type=str, default="DPTNet-10_890.ckpt",
- help='Path to model file created by training')
- parser.add_argument('--device_target', type=str, default="Ascend", choices=['Ascend', 'GPU', 'CPU'],
- help='device where the code will be implemented (default: Ascend)')
- parser.add_argument('--modelArts', default=0, type=int,
- help='Continue from checkpoint model')
- # modelarts
- parser.add_argument('--data_url', default='/home/work/user-job-dir/inputs/data/',
- help='path to training/inference dataset folder')
- parser.add_argument('--train_url', default='/home/work/user-job-dir/model/',
- help='model folder to save/load')
- parser.add_argument('--in_dir', type=str, default=r"/home/work/user-job-dir/inputs/data/",
- help='Directory path of wsj0 including tr, cv and tt')
- parser.add_argument('--out_dir', type=str, default=r"/home/work/user-job-dir/inputs/data_json",
- help='Directory path to put output files')
-
- def main(args):
- context.set_context(mode=context.GRAPH_MODE, device_target=args.device_target)
-
- device_num = int(os.environ.get("RANK_SIZE", 1))
- if device_num == 1:
- is_distributed = 'False'
- elif device_num > 1:
- is_distributed = 'True'
-
- if is_distributed == 'True':
- print("parallel init", flush=True)
- init()
- rank_id = get_rank()
- context.reset_auto_parallel_context()
- parallel_mode = ParallelMode.DATA_PARALLEL
- rank_size = get_group_size()
- context.set_auto_parallel_context(parallel_mode=parallel_mode, gradients_mean=True, device_num=args.device_num)
- context.set_auto_parallel_context(parameter_broadcast=True)
- print("Starting traning on multiple devices...")
- else:
- if args.modelArts:
- init()
- rank_id = get_rank()
- rank_size = get_group_size()
- else:
- context.set_context(device_id=args.device_id)
-
- if args.modelArts:
- import moxing as mox
- obs_data_url = args.data_url
- args.data_url = '/home/work/user-job-dir/inputs/data/'
- obs_train_url = args.train_url
-
- home = os.path.dirname(os.path.realpath(__file__))
- train_dir = os.path.join(home, 'checkpoints') + str(rank_id)
- if not os.path.exists(train_dir):
- os.mkdir(train_dir)
-
- save_checkpoint_path = train_dir + '/device_' + os.getenv('DEVICE_ID') + '/'
- if not os.path.exists(save_checkpoint_path):
- os.makedirs(save_checkpoint_path)
- save_ckpt = os.path.join(save_checkpoint_path, 'dptnet.ckpt')
-
- mox.file.copy_parallel(obs_data_url, args.data_url)
- print("Successfully Download {} to {}".format(obs_data_url, args.data_url))
-
- print("start preprocess on modelArts....")
- preprocess(args)
-
- print("Start datasetgenerator")
- tr_dataset = DatasetGenerator(args.train_dir, args.batch_size,
- sample_rate=args.sample_rate, segment=args.segment)
-
- print("start Generatordataset")
- if is_distributed == 'True':
- tr_loader = ds.GeneratorDataset(tr_dataset, ["mixture", "lens", "sources"],
- shuffle=False, num_shards=rank_size, shard_id=rank_id)
- else:
- tr_loader = ds.GeneratorDataset(tr_dataset, ["mixture", "lens", "sources"],
- shuffle=False)
- tr_loader = tr_loader.batch(2)
-
- print("data loading done")
- # model
- net = DPTNet_base(enc_dim=args.enc_dim, feature_dim=args.feature_dim,
- hidden_dim=args.hidden_dim, layer=args.layer, segment_size=args.segment_size,
- nspk=args.nspk, win_len=args.win_len)
-
- if args.continue_train:
- if args.modelArts:
- home = os.path.dirname(os.path.realpath(__file__))
- ckpt = os.path.join(home, args.ckpt_path)
- params = load_checkpoint(ckpt)
- load_param_into_net(net, params)
- else:
- params = load_checkpoint(args.ckpt_path)
- load_param_into_net(net, params)
-
- print(net)
- net.set_train()
-
- lr = dynamic_lr(args.step_per_epoch, args.epochs)
- optimizer = nn.Adam(net.trainable_params(), learning_rate=lr, beta1=0.9, beta2=0.98, eps=1e-9, weight_decay=args.l2)
-
- my_loss = Loss()
- net_with_loss = WithLossCell(net, my_loss)
- model = Model(net_with_loss, optimizer=optimizer)
-
- time_cb = TimeMonitor()
- loss_cb = LossMonitor(1)
- cb = [time_cb, loss_cb]
- config_ck = CheckpointConfig(save_checkpoint_steps=5,
- keep_checkpoint_max=5)
- if args.modelArts:
- ckpt_cb = ModelCheckpoint(prefix="DPTNet", directory=save_ckpt, config=config_ck)
- else:
- ckpt_cb = ModelCheckpoint(prefix="DPTNet", directory=args.save_folder, config=config_ck)
- cb += [ckpt_cb]
-
- model.train(epoch=args.epochs, train_dataset=tr_loader, callbacks=cb, dataset_sink_mode=False)
-
- if args.modelArts:
- import moxing as mox
- mox.file.copy_parallel(train_dir, obs_train_url)
- print("Successfully Upload {} to {}".format(train_dir, obs_train_url))
-
- if __name__ == '__main__':
- arg = parser.parse_args()
- print(arg)
- main(arg)
|