search

shartoo / luna16_multi_size_3dcnn

An implement of paper "Multi-level Contextual 3D CNNs for False Positive Reduction in Pulmonary Nodule Detection"
111 stars 45 forks source link

FileNotFoundError #6

Open shanpufang opened 6 years ago

shanpufang commented 6 years ago

我在运行data_prepare.py时遇到这个问题： FileNotFoundError：File b' d:/data/luna/CSVFILLES/annotation.csv' dose not exist 后来我自己在相应路径下建了了个文件夹也不可以，请问这是为什么呢？ 作者在主页提供的那个数据集是对应代码的数据集吗？Imagenet32_train.zip

mxl94213 commented 6 years ago

I think you can modify the variable 'base_dir' in file "project_config.py" . Such as ' base_dir = "../data/" ' I think he does not provide the dataset. You should go to the LUNA16 website to download the dataset. Unfortunately, I running the data_prepare.py now , it generates a huge data larger than 100G. I'm considering whether to terminate the running.

shanpufang commented 6 years ago

@mxl94213 这个路径改过了，可是还是不可以，我猜是不是数据集的问题。 这个代码对应的论文中，要求的不是LIDC中获取的数据吗？难道是LUNA16吗？ 谢谢

mxl94213 commented 6 years ago

Yes， you should get the dataset from LUNA16 which has been divided into ten folders.

shanpufang commented 6 years ago

@mxl94213 也就是说，这个代码用的数据集是LUNA16，是下载下来就可以直接输入代码吗？还是的需要处理一下啊？谢谢~

mxl94213 commented 6 years ago

You can use directly.

mxl94213 commented 6 years ago

I think my desktop disk is not enough to store the generated data. So, when you run over the code on dataset, can you share the generated data after processing with me?

shanpufang commented 6 years ago

@mxl94213 对，作者主页上说了的，得到的文件差不多100G，得需要比较大的disk才可以

shanpufang commented 6 years ago

@mxl94213 好的，没有问题，如果我这个跑通了，我在邮件里给你分享结果。

mxl94213 commented 6 years ago

Thank you very much. My email is luminguk34@gmail.com . If you have problems, keeping in touch. @shanpufang.

shanpufang commented 6 years ago

@mxl94213好的，有问题随时联系，我的邮箱1667926207@qq.com