change_detection.pytorch

change_detection.pytorch copied to clipboard

Hi! Thank you for this repo! I want to use these models in an App. Have you considered adding exporting functionality to TFLite?

mikel-brostrom

Roadmap of cdp

1

We keep this issue open to collect feature requests from users and hear your voice. Our release plan is also available here. You can either: Suggest a new feature by...

likyoo

缺少单独调用模型用于测试的脚本

3

可以单独提供一个用于测试新的两张输入图片的脚本吗？ 我已经训练好了模型，并且在最后一步infer_vis出来的验证图片效果不错  但是缺乏一个可以单独调用模型用于测试的脚本，请问作者有什么解决办法或者可以提供一下代码吗

zzhujiuxingc

变化检测微信群二维码失效了，可否提供一个新的呢？

xgyyao

Load trained model weigths

1

Hi @likyoo , I study with yoru repo for my project.I have been added to new features to your repo.I'll share it with you when I'm done. But I have...

ozanpkr

swjtulinxi

两张影像的输入顺序不同会得到不同的结果？

7

请问为啥，两张影像输入的先后顺序不同，结果基本不一样，变化检测不是孪生网络么，理论上不应该啊？ 我参考了其他的变化检测项目代码，也是一样的情况

swjtulinxi

Dear @likyoo thanks for your open source project. I have trained models and saved the best model. Now, how can I test model on new images (not validation)

manapshymyr-OB

In the final step, “infer_vis” can't work.

4

In the verification step of the last step, each “tiff” cannot be opened, which causes the program to end directly and cannot view the accuracy. I guess it's the "infer_vis"...

a624090359

This piece of code : ```python for x in np.arange(0.6, 0.9, 0.1): print('Eval with TH:', x) metrics = [ cdp.utils.metrics.Fscore(activation='argmax2d', threshold=x), cdp.utils.metrics.Precision(activation='argmax2d', threshold=x), cdp.utils.metrics.Recall(activation='argmax2d', threshold=x), ] valid_epoch = cdp.utils.train.ValidEpoch( model,...

mikel-brostrom