对于模型v1,使用的损失函数是loss = (losses["loss"] * weights + losses['loss_cal'] * 10).mean()？以及在segmentation_sample.py文件中，debug时，输出的第一张图片(ss)是怎么的来的，以及使用的是前面的损失函数吗？ 输出的最后一张图片(c)是怎么的来的 是使用loss_cal的损失函数计算得来的吗？ 希望有人能解答一下

[LiuYM00]

[Issues-translate-bot]

Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑‍🤝‍🧑👫🧑🏿‍🤝‍🧑🏻👩🏾‍🤝‍👨🏿👬🏿

---

Title: For model v1, the loss function used is loss = (losses["loss"] * weights + losses['loss_cal'] * 10).mean()? And in the segmentation_sample.py file, when debugging, how did the first picture (ss) output come from, and is the previous loss function used? How did the last picture (c) output come from? Is it calculated using the loss function of loss_cal? Hope someone can answer it