| import swanlab | |
| import random | |
| # 初始化一个新的swanlab run类来跟踪这个脚本 | |
| swanlab.init( | |
| # 设置将记录此次运行的项目信息 | |
| project="MODNet", | |
| workspace="wudi", | |
| # 跟踪超参数和运行元数据 | |
| config={ | |
| "learning_rate": 0.02, | |
| "architecture": "CNN", | |
| "dataset": "CIFAR-100", | |
| "epochs": 10 | |
| } | |
| ) | |
| # 模拟训练 | |
| 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 | |
| # 向swanlab上传训练指标 | |
| swanlab.log({"acc": acc, "loss": loss}) | |
| # [可选] 完成训练，这在notebook环境中是必要的 | |
| swanlab.finish() |