通过FastSpeech2中文合成项目梳理TTS流程2: 数据训练（train.py)

1. 参考github网址：

GitHub - roedoejet/FastSpeech2: An implementation of Microsoft's "FastSpeech 2: Fast and High-Quality End-to-End Text to Speech"

2. 数据训练所用python 命令：

python3 train.py -p config/AISHELL3/preprocess.yaml -m config/AISHELL3/model.yaml -t config/AISHELL3/train.yaml

3. 数据训练代码解析

3.1 代码架构overview：

通过 if __name__ == "__main__"运行整个py文件：

调用 “train.txt"和dataset.py加载数据，

调用utils文件夹下的model.py加载模型，声码器，

调用model文件夹下的loss.py中的FastSpeech2Loss class 设置损失函数，

用前面加载的模型和损失函数开始训练模型，导出结果并记录日志。

3.2 按训练步骤分解代码：

Step 0 : 定义可控训练参数, 调动main函数

if __name__ == "__main__":

    #Define Args
    parser = argparse.ArgumentParser()
    parser.add_argument("--restore_step", type=int, default=0)
    parser.add_argument(
        "-p",
        "--preprocess_config",
        type=str,
        required=True,
        help="path to preprocess.yaml",
    )
    parser.add_argument(
        "-m", "--model_config", type=str, required=True, help="path to model.yaml"
    )
    parser.add_argument(
        "-t", "--train_config", type=str, required=True, help="path to train.yaml"
    )
    args = parser.parse_args() #args为可控训练参数

    # Read Config
    preprocess_config = yaml.load(
        open(args.preprocess_config, "r"), Loader=yaml.FullLoader
    )
    model_config = yaml.load(open(args.model_config, "r"), Loader=yaml.FullLoader)
    train_config = yaml.load(open(args.train_config, "r"), Loader=yaml.FullLoader)
    configs = (preprocess_config, model_config, train_config)

    #Run _main_ function
    main(args, configs)

Step 1 : 启动main函数，加载可控训练参数

def main(args, configs): 
    print("Prepare training ...")

    #加载可控训练参数
    preprocess_config, model_config, train_config = configs

Step 2 : 从train.txt加载数据，并经由dataset.py和torch里的Dataloader处理

def main(args, configs):

    # Get dataset
    dataset = Dataset(
        "train.txt", preprocess_config, train_config, sort=True, drop_last=True
    ) #从 train.txt 中获取dataset
    batch_size = train_config["optimizer"]["batch_size"]
    group_size = 4  # Set this larger than 1 to enable sorting in Dataset,初始值为4

    assert batch_size * group_size < len(dataset)
    loader = DataLoader(
        dataset,
        batch_size=batch_size * group_size,
        shuffle=True,
        collate_fn=dataset.collate_fn,
    )

Step 3 : 定义模型，声码器，损失函数

def main(args, configs):

    # Prepare model
    model, optimizer = get_model(args, configs, device, train=True) #设置优化器

    # 将模型并行训练并移入计算设备中
    model = nn.DataParallel(model) # Model Has Been Defined

    # 计算模型参数量
    num_param = get_param_num(model) # Number of TTS Parameters: num_param
    print("Number of FastSpeech2 Parameters:", num_param)

    # 设置损失函数
    Loss = FastSpeech2Loss(preprocess_config, model_config).to(device)

    # 加载声码器
    vocoder = get_vocoder(model_config, device)

Step 4 : 加载日志，在"./output/log/AISHELL3"目录建立train, val两个文件夹来记录日志

def main(args, configs):

    # Init logger
    for p in train_config["path"].values():
        os.makedirs(p, exist_ok=True)
    train_log_path = os.path.join(train_config["path"]["log_path"], "train")
    val_log_path = os.path.join(train_config["path"]["log_path"], "val")
    os.makedirs(train_log_path, exist_ok=True)
    os.makedirs(val_log_path, exist_ok=True)
    train_logger = SummaryWriter(train_log_path)
    val_logger = SummaryWriter(val_log_path)

Step 5 : 准备训练，加载可控训练参数

def main(args, configs):

    # Training
    step = args.restore_step + 1
    epoch = 1
    grad_acc_step = train_config["optimizer"]["grad_acc_step"]
    grad_clip_thresh = train_config["optimizer"]["grad_clip_thresh"]
    total_step = train_config["step"]["total_step"]
    log_step = train_config["step"]["log_step"]
    save_step = train_config["step"]["save_step"]
    synth_step = train_config["step"]["synth_step"]
    val_step = train_config["step"]["val_step"]

    outer_bar = tqdm(total=total_step, desc="Training", position=0)
    outer_bar.n = args.restore_step
    outer_bar.update()

Step 6 : 准备训练，加载进度条，调动utils文件夹下tools.py中的to_device function来提取数据

    while True:
        inner_bar = tqdm(total=len(loader), desc="Epoch {}".format(epoch), position=1)
        for batchs in loader:
            for batch in batchs:
                batch = to_device(batch, device)

Step 7 :开始训练，前向传播，计算损失，反向传播，梯度剪枝，更新模型权重参数

    #Load Data
            for batch in batchs:
                batch = to_device(batch, device)
                
                # Forward
                output = model(*(batch[2:]))

                # Cal Loss
                losses = Loss(batch, output)
                total_loss = losses[0]

                # Backward
                total_loss = total_loss / grad_acc_step
                total_loss.backward()
                if step % grad_acc_step == 0:
                    # Clipping gradients to avoid gradient explosion
                    nn.utils.clip_grad_norm_(model.parameters(), grad_clip_thresh)

                    # Update weights
                    optimizer.step_and_update_lr()
                    optimizer.zero_grad()

Step 8 : 当训练步数到达预先设定的log_step时，调动utils文件夹下tool.py里的log function，记录loss和step

                if step % log_step == 0:
                    losses = [l.item() for l in losses]
                    message1 = "Step {}/{}, ".format(step, total_step)
                    message2 = "Total Loss: {:.4f}, Mel Loss: {:.4f}, Mel PostNet Loss: {:.4f}, Pitch Loss: {:.4f}, Energy Loss: {:.4f}, Duration Loss: {:.4f}".format(
                        *losses
                    )

                    with open(os.path.join(train_log_path, "log.txt"), "a") as f:
                        f.write(message1 + message2 + "\n")

                    outer_bar.write(message1 + message2)

                    log(train_logger, step, losses=losses)

Step 9 : 当训练步数到达预先设定的synth_step时，调动utils文件夹下tool.py里的log function 和 synth_one_sample function（具体用来干什么没看懂）

                if step % synth_step == 0:
                    fig, wav_reconstruction, wav_prediction, tag = synth_one_sample(
                        batch,
                        output,
                        vocoder,
                        model_config,
                        preprocess_config,
                    )
                    log(
                        train_logger,
                        fig=fig,
                        tag="Training/step_{}_{}".format(step, tag),
                    )
                    sampling_rate = preprocess_config["preprocessing"]["audio"][
                        "sampling_rate"
                    ]
                    log(
                        train_logger,
                        audio=wav_reconstruction,
                        sampling_rate=sampling_rate,
                        tag="Training/step_{}_{}_reconstructed".format(step, tag),
                    )
                    log(
                        train_logger,
                        audio=wav_prediction,
                        sampling_rate=sampling_rate,
                        tag="Training/step_{}_{}_synthesized".format(step, tag),
                    )

Step 10 : 当训练步数到达预先设定的val_step时，调动evaluate.py里的evaluate function来进行evaluation,并记录在log/AISHELL3/val/log.txt

                if step % val_step == 0:
                    model.eval()
                    message = evaluate(model, step, configs, val_logger, vocoder)
                    with open(os.path.join(val_log_path, "log.txt"), "a") as f:
                        f.write(message + "\n")
                    outer_bar.write(message)

                    model.train()

Step 11 : 当训练步数到达预先设定的save_step时，保存训练模型

                if step % save_step == 0:
                    torch.save(
                        {
                            "model": model.module.state_dict(),
                            "optimizer": optimizer._optimizer.state_dict(),
                        },
                        os.path.join(
                            train_config["path"]["ckpt_path"],
                            "{}.pth.tar".format(step),
                        ),
                    )

Step 12 : 当训练步数到达预先设定的total_step时，退出训练

                if step == total_step:
                    quit()
                step += 1
                outer_bar.update(1)

            inner_bar.update(1)
        epoch += 1

4. 数据训练代码的输出

在train_log_path和val_log_path输出日志

在ckpt_path输出训练过程中按照save_step存储的模型