使用SwinUnet在自己的数据集（含有14种器官）上训练多器官分割模型

[SurvivorNo1]

问题确认 Search before asking

• [X] 我已经搜索过问题，但是没有找到解答。I have searched the question and found no related answer.

请提出你的问题 Please ask your question

我需要在自己的数据集上使用SwinUnet训练一个多器官分割模型，但是官方给出的配置文件只有8种器官且数据集处理代码tools/prepare_abdomen.py中也只显示了8种器官，请问一下直接修改配置文件的num_classes大于8，是否可以使用？

if __name__ == "__main__":
    prep = Prep_abdomen()
    prep.generate_dataset_json(
        modalities=('CT', ),
        labels={
            0: 'background',
            1: 'spleen',
            2: 'right kidney',
            3: 'left kidney',
            4: 'gallbladder',
            5: 'liver',
            6: 'stomach',
            7: 'aorta',
            8: 'pancreas'
        },
        dataset_name="Abdomen CT scans",
        dataset_description="Under Institutional Review Board (IRB) supervision, 50 abdomen CT scans of were randomly selected from a combination of an ongoing colorectal cancer chemotherapy trial, and a retrospective ventral hernia study.",
        license_desc="https://creativecommons.org/licenses/by/4.0/legalcode",
        dataset_reference="https://www.synapse.org/#!Synapse:syn3193805/wiki/89480",
    )
    prep.load_save(mode='train')
    prep.load_save(mode='val')
    prep.generate_txt()

[shiyutang]

可以使用，详情可以参考在自己数据上训练的教程：https://github.com/PaddlePaddle/PaddleSeg/blob/develop/contrib/MedicalSeg/README_CN.md#-%E5%9C%A8%E8%87%AA%E5%B7%B1%E7%9A%84%E6%95%B0%E6%8D%AE%E4%B8%8A%E8%AE%AD%E7%BB%83