search

Jun-Jie-Shi / M2FTrans

[IEEE-JBHI'2023] M2FTrans: Modality-Masked Fusion Transformer for Incomplete Multi-Modality Brain Tumor Segmentation
https://doi.org/10.1109/JBHI.2023.3326151
MIT License
15 stars 2 forks source link

visualization problem #5

Open Deer231 opened 3 months ago

Deer231 commented 3 months ago

Hello, I would like to ask you a question, how can I visualize the predicted results and labels?

Jun-Jie-Shi commented 3 months ago

@Deer231 Hi, I use a relatively simple and crude method, i.e. after finding the required samples, I load the pretrained model and save the samples' prediction results as npy files, and then subsequently visualize them according to the corresponding labels as well as the raw images using the cv2 library.

Deer231 commented 3 months ago

Okay, thank you very much for your answer!

 

YYL @.***

 

------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2024年6月11日(星期二) 晚上11:00 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [Jun-Jie-Shi/M2FTrans] visualization problem (Issue #5)

@Deer231 Hi, I use a relatively simple and crude method, i.e. after finding the required samples, I load the pretrained model and save the samples' prediction results as npy files, and then subsequently visualize them according to the corresponding labels as well as the raw images using the cv2 library. 嗨，我使用了一种相对简单粗暴的方法，即在找到所需样本后，加载预训练模型并将样本的预测结果保存为 npy 文件，然后根据相应的标签以及使用 cv2 库的原始图像对其进行可视化。

— Reply to this email directly, view it on GitHub, or unsubscribe. 直接回复此电子邮件，在 GitHub 上查看或取消订阅。 You are receiving this because you were mentioned. 你收到这个是因为你被提到了。 Message ID: @.***>