Tensorboard + Logger 日志记录

在Pytorch 训练模型的时候，需要日志帮助开发者记录些重要信息和参数，以方便开发者更好的调节模型及参数，常见的日志非 Tensorboard不可，但是Pytorch 对 Tensorboard 的支持不是十分完美，在记录模型重要参数时 Tensorboard 的 writer.add_text() 记录日志不是很方便，Logger 便可替代 Tensorboard 记录文本日志，Tensorboard 专注 模型训练图，两者搭配使用极大提高效率，且讲两者放在 Tensorboard 日志文件夹下，方便调用查看。

• 【解析】TensorBoard 可视化：https://blog.csdn.net/ViatorSun/article/details/116275485

1、train 文件

#  !/usr/bin/env  python
#  -*- coding:utf-8 -*-
# @Time     : 2022.01
# @Author   : 绿色羽毛
# @Email    : lvseyumao@foxmail.com
# @Blog     : https://blog.csdn.net/ViatorSun
# @version  : "1.0" 
# @Note     : Tensorboard + logger 日志使用


import os
import socket
from datetime import datetime
from torch.utils.tensorboard import SummaryWriter





""" ********************************* Tensorboard/日志 配置 ********************************* """
current_time = datetime.now().strftime('%b%d_%H-%M-%S')
log_dir = os.path.join('runs', current_time + '_' + args.run_model + '_' + socket.gethostname())

writer = SummaryWriter(log_dir)
logger = create_logger(log_dir, current_time + args.log_name)
config_from_args = args.__dict__
total_parameters = sum(p.numel() for p in model.parameters() if p.requires_grad )
logger.info(f"=============================== Modle Arguments ===============================")
logger.info(f">>> Model Name        : {args.model_name, args.run_model}")
logger.info(f">>> Training time     : {current_time}")
logger.info(f">>> Computation with  : {current_time}")
logger.info(f"== Model Total params : %.2f MB \n" % (total_parameters / 1e6))
logger.info(f"== ArgumentParser:{json.dumps(config_from_args, indent=20)}")
logger.info(f"=============================== Train Arguments ===============================")
logger.info(f"== criterion     :{criterion}")
logger.info(f"== optimizer     :{optimizer.__class__}")
logger.info(f"                 :{json.dumps(optimizer.defaults, indent=20)}")

logger.info(f"== lr_scheduler  :{lr_scheduler.__class__}")
logger.info(f"                 :{json.dumps(lr_scheduler.state_dict(), indent=20)}")
logger.info(f"==============================================================================")

logger.info(f"Creating model:{args.model_name}")
logger.info(model)

生成的日志

2、创建logger 日志

#  !/usr/bin/env  python
#  -*- coding:utf-8 -*-
# @Time     : 2022.01
# @Author   : 绿色羽毛
# @Email    : lvseyumao@foxmail.com
# @Blog     : https://blog.csdn.net/ViatorSun
# @version  : "1.0" 
# @Note     : Pytorch 文本日志


import os
import logging


def create_logger(log_dir, filename, verbosity=1, name=None):
    level_dict = {0: logging.DEBUG, 1: logging.INFO, 2: logging.WARNING}

    formatter  = logging.Formatter( "[%(asctime)s][%(filename)s][line:%(lineno)d][%(levelname)s] %(message)s" )
    logger     = logging.getLogger(name)
    logger.setLevel(level_dict[verbosity])

    # fh = logging.FileHandler(filename, "w")
    fh = logging.FileHandler(os.path.join(log_dir, filename+'.txt'), mode='a')

    fh.setFormatter(formatter)
    logger.addHandler(fh)

    sh = logging.StreamHandler()
    sh.setFormatter(formatter)
    logger.addHandler(sh)

    return logger

3、Tensorboard 日志

在此截取一部分 Tensorboard 接口代码，可见在没有用户输入 log_dir 时，Tensorboard 会自动创建 runs文件夹，然后以当前时间 current_time + gethostname() + comment 命名Tensorboard日志的文件名，

class SummaryWriter(object):
    """Writes entries directly to event files in the log_dir to be
    consumed by TensorBoard.

    The `SummaryWriter` class provides a high-level API to create an event file
    in a given directory and add summaries and events to it. The class updates the
    file contents asynchronously. This allows a training program to call methods
    to add data to the file directly from the training loop, without slowing down
    training.
    """

    def __init__(self, log_dir=None, comment='', purge_step=None, max_queue=10,
                 flush_secs=120, filename_suffix=''):
        """Creates a `SummaryWriter` that will write out events and summaries
        to the event file.

        Args:
            log_dir (string): Save directory location. Default is
              runs/**CURRENT_DATETIME_HOSTNAME**, which changes after each run.
              Use hierarchical folder structure to compare
              between runs easily. e.g. pass in 'runs/exp1', 'runs/exp2', etc.
              for each new experiment to compare across them.
            comment (string): Comment log_dir suffix appended to the default
              ``log_dir``. If ``log_dir`` is assigned, this argument has no effect.
            purge_step (int):
              When logging crashes at step :math:`T+X` and restarts at step :math:`T`,
              any events whose global_step larger or equal to :math:`T` will be
              purged and hidden from TensorBoard.
              Note that crashed and resumed experiments should have the same ``log_dir``.
            max_queue (int): Size of the queue for pending events and
              summaries before one of the 'add' calls forces a flush to disk.
              Default is ten items.
            flush_secs (int): How often, in seconds, to flush the
              pending events and summaries to disk. Default is every two minutes.
            filename_suffix (string): Suffix added to all event filenames in
              the log_dir directory. More details on filename construction in
              tensorboard.summary.writer.event_file_writer.EventFileWriter.

        Examples::

            from torch.utils.tensorboard import SummaryWriter

            # create a summary writer with automatically generated folder name.
            writer = SummaryWriter()
            # folder location: runs/May04_22-14-54_s-MacBook-Pro.local/

            # create a summary writer using the specified folder name.
            writer = SummaryWriter("my_experiment")
            # folder location: my_experiment

            # create a summary writer with comment appended.
            writer = SummaryWriter(comment="LR_0.1_BATCH_16")
            # folder location: runs/May04_22-14-54_s-MacBook-Pro.localLR_0.1_BATCH_16/

        """
        torch._C._log_api_usage_once("tensorboard.create.summarywriter")
        if not log_dir:
            import socket
            from datetime import datetime
            current_time = datetime.now().strftime('%b%d_%H-%M-%S')
            log_dir = os.path.join(
                'runs', current_time + '_' + socket.gethostname() + comment)
        self.log_dir = log_dir