训练过程可视化tensorboard和wandb及np.array和tensor互相转换

tensorboard

from tensorboardX import SummaryWriter
#设置保存日志文件路径
logger_path = os.path.join(path, current_time)
logger = SummaryWriter(log_dir=logger_path, comment=comment)

#要保存的数据
logger.add_scalar("value_loss", value_loss, global_step=(step + 1))

wandb

记录日志

import wandb
# start a new wandb run to track this script
wandb.init(
    # 文件名称
    project="my-awesome-project",
    # 运行超参数的配置
    config={
        "learning_rate": 0.02,
        "architecture": "CNN",
        "dataset": "CIFAR-100",
        "epochs": 10,
    },
)

在要记录的数据后加上

 # log metrics to wandb
    wandb.log({"acc": acc, "loss": loss})

最后

wandb.finish()

具体实例

import wandb
import random

# start a new wandb run to track this script
wandb.init(
    # set the wandb project where this run will be logged
    project="my-awesome-project",
    # track hyperparameters and run metadata
    config={
        "learning_rate": 0.02,
        "architecture": "CNN",
        "dataset": "CIFAR-100",
        "epochs": 10,
    },
)


# simulate training
epochs = 10
offset = random.random() / 5
for epoch in range(2, epochs):
    acc = 1 - 2**-epoch - random.random() / epoch - offset
    loss = 2**-epoch + random.random() / epoch + offset

    # log metrics to wandb
    wandb.log({"acc": acc, "loss": loss})

# [optional] finish the wandb run, necessary in notebooks
wandb.finish()

np.array->tensor

ratio = torch.from_numpy(np.array([1.6, 0.5, 1, 0.6]))

tensor->np.array

x=x.numpy()