train.py

import warnings
warnings.simplefilter(action='ignore', category=FutureWarning)
import itertools
import os
import time
import argparse
import json
import torch
import torch.nn.functional as F
from torch.utils.tensorboard import SummaryWriter
from torch.utils.data import DistributedSampler, DataLoader
import torch.multiprocessing as mp
from torch.distributed import init_process_group
from torch.nn.parallel import DistributedDataParallel
from env import AttrDict, build_env
from meldataset import MelDataset, mel_spectrogram, get_dataset_filelist
from models import Generator, MultiPeriodDiscriminator, MultiScaleDiscriminator, MultiResSpecDiscriminator,\
    CoMBD, SBD, discriminator_loss, feature_loss, generator_loss
from pqmf import PQMF

from utils import plot_spectrogram, scan_checkpoint, load_checkpoint, save_checkpoint
from stft import TorchSTFT

torch.backends.cudnn.benchmark = True

def get_n_params(model):
    pp=0
    for p in list(model.parameters()):
        nn=1
        for s in list(p.size()):
            nn = nn*s
        pp += nn
    return pp

def train(rank, a, h):
    if h.num_gpus > 1:
        init_process_group(backend=h.dist_config['dist_backend'], init_method=h.dist_config['dist_url'],
                           world_size=h.dist_config['world_size'] * h.num_gpus, rank=rank)

    torch.cuda.manual_seed(h.seed)
    device = torch.device('cuda:{:d}'.format(rank))

    subbands2, taps2, cutoff_ratio2, beta2 = h.pqmf_config["lv2"]
    subbands1, taps1, cutoff_ratio1, beta1 = h.pqmf_config["lv2"]
    pqmf_lv2 = PQMF(subbands2, taps2, cutoff_ratio2, beta2)
    pqmf_lv1 = PQMF(subbands1, taps1, cutoff_ratio1, beta1)

    generator = Generator(h).to(device)

    combd = CoMBD(h.combd, [pqmf_lv2, pqmf_lv1]).to(device)
    sbd = SBD(h["sbd"]).to(device)

    if h.use_mrsd:
        print("Use mrsd...")
        msd = MultiResSpecDiscriminator().to(device)
    else:
        msd = MultiScaleDiscriminator().to(device)
    stft = TorchSTFT(filter_length=h.gen_istft_n_fft, hop_length=h.gen_istft_hop_size, win_length=h.gen_istft_n_fft, device=device).to(device)

    if rank == 0:
        print(generator)
        print("Size")
        print(get_n_params(generator))

        os.makedirs(a.checkpoint_path, exist_ok=True)
        print("checkpoints directory : ", a.checkpoint_path)

    if os.path.isdir(a.checkpoint_path):
        cp_g = scan_checkpoint(a.checkpoint_path, 'g_')
        cp_do = scan_checkpoint(a.checkpoint_path, 'do_')

    steps = 0
    if cp_g is None or cp_do is None:
        state_dict_do = None
        last_epoch = -1
    else:
        state_dict_g = load_checkpoint(cp_g, device)
        state_dict_do = load_checkpoint(cp_do, device)
        generator.load_state_dict(state_dict_g['generator'])
        combd.load_state_dict(state_dict_do['combd'])
        sbd.load_state_dict(state_dict_do['sbd'])
        msd.load_state_dict(state_dict_do['msd'])
        steps = state_dict_do['steps'] + 1
        last_epoch = state_dict_do['epoch']

    if h.num_gpus > 1:
        generator = DistributedDataParallel(generator, device_ids=[rank]).to(device)
        combd = DistributedDataParallel(combd, device_ids=[rank]).to(device)
        sbd = DistributedDataParallel(sbd, device_ids=[rank]).to(device)
        msd = DistributedDataParallel(msd, device_ids=[rank]).to(device)

    optim_g = torch.optim.AdamW(generator.parameters(), h.learning_rate, betas=[h.adam_b1, h.adam_b2])
    optim_d = torch.optim.AdamW(itertools.chain(msd.parameters(), sbd.parameters(), combd.parameters()),
                                h.learning_rate, betas=[h.adam_b1, h.adam_b2])

    if state_dict_do is not None:
        optim_g.load_state_dict(state_dict_do['optim_g'])
        optim_d.load_state_dict(state_dict_do['optim_d'])

    scheduler_g = torch.optim.lr_scheduler.ExponentialLR(optim_g, gamma=h.lr_decay, last_epoch=last_epoch)
    scheduler_d = torch.optim.lr_scheduler.ExponentialLR(optim_d, gamma=h.lr_decay, last_epoch=last_epoch)

    training_filelist, validation_filelist = get_dataset_filelist(a)

    trainset = MelDataset(training_filelist, h.segment_size, h.n_fft, h.num_mels,
                          h.hop_size, h.win_size, h.sampling_rate, h.fmin, h.fmax, n_cache_reuse=0,
                          shuffle=False if h.num_gpus > 1 else True, fmax_loss=h.fmax_for_loss, device=device,
                          fine_tuning=a.fine_tuning, base_mels_path=a.input_mels_dir)

    train_sampler = DistributedSampler(trainset) if h.num_gpus > 1 else None

    train_loader = DataLoader(trainset, num_workers=h.num_workers, shuffle=False,
                              sampler=train_sampler,
                              batch_size=h.batch_size,
                              pin_memory=True,
                              drop_last=True)

    if rank == 0:
        validset = MelDataset(validation_filelist, h.segment_size, h.n_fft, h.num_mels,
                              h.hop_size, h.win_size, h.sampling_rate, h.fmin, h.fmax, False, False, n_cache_reuse=0,
                              fmax_loss=h.fmax_for_loss, device=device, fine_tuning=a.fine_tuning,
                              base_mels_path=a.input_mels_dir)
        validation_loader = DataLoader(validset, num_workers=1, shuffle=False,
                                       sampler=None,
                                       batch_size=1,
                                       pin_memory=True,
                                       drop_last=True)

        sw = SummaryWriter(os.path.join(a.checkpoint_path, 'logs'))

    generator.train()
    combd.train()
    sbd.train()
    msd.train()
    for epoch in range(max(0, last_epoch), a.training_epochs):
        if rank == 0:
            start = time.time()
            print("Epoch: {}".format(epoch+1))

        if h.num_gpus > 1:
            train_sampler.set_epoch(epoch)

        for i, batch in enumerate(train_loader):
            if rank == 0:
                start_b = time.time()
            x, y, length, _, y_mel = batch
            x = torch.autograd.Variable(x.to(device, non_blocking=True))
            y = torch.autograd.Variable(y.to(device, non_blocking=True))
            length = torch.autograd.Variable(length.to(device, non_blocking=True))
            y_mel = torch.autograd.Variable(y_mel.to(device, non_blocking=True))
            y = y.unsqueeze(1)

            #if use this, need to add some conv in generator, follow: https://github.com/ncsoft/avocodo/blob/2999557bbd040a6f3eb6f7006a317d89537b78cd/avocodo/models/generator.py#L109
            #in this implement, only last conv is used

            # ys = [
            #     pqmf_lv2.analysis(
            #         y
            #     )[:, :h.projection_filters[1]],
            #     pqmf_lv1.analysis(
            #         y
            #     )[:, :h.projection_filters[2]],
            #     y
            # ]
            ys = [y]

            spec, phase = generator(x, length)

            y_g_hat = stft.inverse(spec, phase)
            
            y_g_hat_mel = mel_spectrogram(y_g_hat.squeeze(1), h.n_fft, h.num_mels, h.sampling_rate, h.hop_size, h.win_size,
                                          h.fmin, h.fmax_for_loss)

            ys_hat = [y_g_hat.detach()]

            optim_d.zero_grad()
            
            # combd
            #ys list length 1 contain y shape batch x 1 x segment_size
            #y_g_hat shape batch x 1 x segment_size
            y_du_hat_r, y_du_hat_g, _, _ = combd(ys, ys_hat)
            loss_disc_u, losses_disc_u_r, losses_disc_u_g = discriminator_loss(y_du_hat_r, y_du_hat_g)

            # sbd
            y_dp_hat_r, y_dp_hat_g, _, _ = sbd(y, y_g_hat.detach())
            loss_disc_p, losses_disc_p_r, losses_disc_p_g = discriminator_loss(y_dp_hat_r, y_dp_hat_g)
            
            # MSD
            y_ds_hat_r, y_ds_hat_g, _, _ = msd(y, y_g_hat.detach())
            loss_disc_s, losses_disc_s_r, losses_disc_s_g = discriminator_loss(y_ds_hat_r, y_ds_hat_g)

            loss_disc_all = loss_disc_s + loss_disc_p + loss_disc_u

            loss_disc_all.backward()
            optim_d.step()

            # Generator
            optim_g.zero_grad()

            # L1 Mel-Spectrogram Loss
            loss_mel = F.l1_loss(y_mel, y_g_hat_mel) * 45
            y_du_hat_r, y_du_hat_g, fmap_u_r, fmap_u_g = combd(ys, ys_hat)
            y_dp_hat_r, y_dp_hat_g, fmap_p_r, fmap_p_g = sbd(y, y_g_hat)            
            y_ds_hat_r, y_ds_hat_g, fmap_s_r, fmap_s_g = msd(y, y_g_hat)

            loss_fm_u = feature_loss(fmap_u_r, fmap_u_g)
            loss_fm_p = feature_loss(fmap_p_r, fmap_p_g)
            loss_fm_s = feature_loss(fmap_s_r, fmap_s_g)

            loss_gen_u, losses_gen_u = generator_loss(y_du_hat_g)
            loss_gen_p, losses_gen_p = generator_loss(y_dp_hat_g)
            loss_gen_s, losses_gen_s = generator_loss(y_ds_hat_g)

            loss_gen_all = loss_gen_s + loss_gen_p + loss_gen_u + loss_fm_s + loss_fm_u + loss_fm_p + loss_mel

            loss_gen_all.backward()
            optim_g.step()

            if rank == 0:
                # STDOUT logging
                if steps % a.stdout_interval == 0:
                    with torch.no_grad():
                        mel_error = F.l1_loss(y_mel, y_g_hat_mel).item()

                    print('Steps : {:d}, Gen Loss Total : {:4.3f}, Mel-Spec. Error : {:4.3f}, s/b : {:4.3f}'.
                          format(steps, loss_gen_all, mel_error, time.time() - start_b))

                # checkpointing
                if steps % a.checkpoint_interval == 0 and steps != 0:
                    checkpoint_path = "{}/g_{:08d}".format(a.checkpoint_path, steps)
                    save_checkpoint(checkpoint_path,
                                    {'generator': (generator.module if h.num_gpus > 1 else generator).state_dict()})
                    checkpoint_path = "{}/do_{:08d}".format(a.checkpoint_path, steps)
                    try:
                        os.remove("{}/do_{:08d}".format(a.checkpoint_path, steps-10000))
                    except:
                        print("continue_training")
                    save_checkpoint(checkpoint_path, 
                                    {'combd': (combd.module if h.num_gpus > 1
                                                         else combd).state_dict(),
                                     'sbd': (sbd.module if h.num_gpus > 1
                                                         else sbd).state_dict(),
                                     'msd': (msd.module if h.num_gpus > 1
                                                         else msd).state_dict(),
                                     'optim_g': optim_g.state_dict(), 'optim_d': optim_d.state_dict(), 'steps': steps,
                                     'epoch': epoch})

                # Tensorboard summary logging
                if steps % a.summary_interval == 0:
                    sw.add_scalar("training/gen_loss_total", loss_gen_all, steps)
                    sw.add_scalar("training/mel_spec_error", mel_error, steps)

                # Validation
                if steps % a.validation_interval == 0:  # and steps != 0:
                    generator.eval()
                    torch.cuda.empty_cache()
                    val_err_tot = 0
                    with torch.no_grad():
                        for j, batch in enumerate(validation_loader):
                            x, y, length, _, y_mel = batch
                            # y_g_hat = generator(x.to(device))
                            length = length.to(device)
                            length = torch.LongTensor([x.size(2)]).to(device)
                            spec, phase = generator(x.to(device), length)

                            y_g_hat = stft.inverse(spec, phase)

                            y_mel = torch.autograd.Variable(y_mel.to(device, non_blocking=True))
                            
                            y_g_hat_mel = mel_spectrogram(y_g_hat.squeeze(1), h.n_fft, h.num_mels, h.sampling_rate,
                                                          h.hop_size, h.win_size,
                                                          h.fmin, h.fmax_for_loss)
                            val_err_tot += F.l1_loss(y_mel, y_g_hat_mel).item()

                            if j <= 4:
                                if steps == 0:
                                    sw.add_audio('gt/y_{}'.format(j), y[0], steps, h.sampling_rate)
                                    sw.add_figure('gt/y_spec_{}'.format(j), plot_spectrogram(x[0]), steps)

                                sw.add_audio('generated/y_hat_{}'.format(j), y_g_hat[0], steps, h.sampling_rate)
                                
                                y_hat_spec = mel_spectrogram(y_g_hat.squeeze(1), h.n_fft, h.num_mels,
                                                             h.sampling_rate, h.hop_size, h.win_size,
                                                             h.fmin, h.fmax)
                                sw.add_figure('generated/y_hat_spec_{}'.format(j),
                                              plot_spectrogram(y_hat_spec.squeeze(0).cpu().numpy()), steps)

                        val_err = val_err_tot / (j+1)
                        sw.add_scalar("validation/mel_spec_error", val_err, steps)

                    generator.train()

            steps += 1

        scheduler_g.step()
        scheduler_d.step()
        
        if rank == 0:
            print('Time taken for epoch {} is {} sec\n'.format(epoch + 1, int(time.time() - start)))


def main():
    print('Initializing Training Process..')

    parser = argparse.ArgumentParser()

    parser.add_argument('--group_name', default=None)
    parser.add_argument('--input_wavs_dir', default='LJSpeech-1.1/wavs')
    parser.add_argument('--input_mels_dir', default='ft_dataset')
    parser.add_argument('--input_training_file', default='LJSpeech-1.1/training.txt')
    parser.add_argument('--input_validation_file', default='LJSpeech-1.1/validation.txt')
    parser.add_argument('--checkpoint_path', default='lightvoc_checkpoint')
    parser.add_argument('--config', default='')
    parser.add_argument('--training_epochs', default=3000, type=int)
    parser.add_argument('--stdout_interval', default=5, type=int)
    parser.add_argument('--checkpoint_interval', default=10000, type=int)
    parser.add_argument('--summary_interval', default=100, type=int)
    parser.add_argument('--validation_interval', default=2000, type=int)
    parser.add_argument('--fine_tuning', default=False, type=bool)

    a = parser.parse_args()

    with open(a.config) as f:
        data = f.read()

    json_config = json.loads(data)
    h = AttrDict(json_config)
    build_env(a.config, 'config.json', a.checkpoint_path)

    torch.manual_seed(h.seed)
    if torch.cuda.is_available():
        torch.cuda.manual_seed(h.seed)
        h.num_gpus = torch.cuda.device_count()
        h.batch_size = int(h.batch_size / h.num_gpus)
        print('Batch size per GPU :', h.batch_size)
    else:
        pass

    if h.num_gpus > 1:
        mp.spawn(train, nprocs=h.num_gpus, args=(a, h,))
    else:
        train(0, a, h)


if __name__ == '__main__':
    main()