lisjin / dcan-tensorflow Goto Github PK
View Code? Open in Web Editor NEWDeep Contour-Aware Networks in TensorFlow
Deep Contour-Aware Networks in TensorFlow
What's ur reason?
thanks
pywrap_tensorflow.TF_GetCode(status))
tensorflow.python.framework.errors_impl.NotFoundError: train.csv
[[Node: ReaderReadV2 = ReaderReadV2[_device="/job:localhost/replica:0/task:0/cpu:0"](TextLineReaderV2, input_producer)]]
It through this error can you help me to solve
Subscript indices must either be real positive integers or logicals.
Error in fmArgmax (line 8)
topIndices(numImgs).ind = 0;
Error in saveImgsAll (line 7)
topIndices = fmArgmax(fmVals, imgNamesAll);
Error in fmArgmax (line 8)
topIndices(numImgs).ind = 0;
Error in saveImgsAll (line 2)
topIndices = fmArgmax(fmVals,imgNamesAll);
hi i trained the network successfully, and i have done test by using "evaluate" but i need to generate output images for test set. i am going to regenrate the images that are shown in "DCAN: Deep Contour-Aware Networks for Accurate Gland Segmentation", is there any method in your implementation to regenerate the images.
i will be appreciate for your response.
in dcan-tensorflow/tf-dcan/bbbc006_input.py
line 71: result.label = tf.concat([contour, segment], 2) and dcan-tensorflow/tf-dcan/traceBounds.m
it seems that two classes can both be 1 in same pixel in label array?
vals = unique(nonzeros(img));
bwImg = false(size(img));
for j = 1:length(vals)
cur = (img == vals(j, 1));
[bounds, ~] = bwboundaries(cur, 'noholes');
indices = bounds{1};
for k = 1:size(indices, 1)
bwImg(indices(k, 1), indices(k, 2)) = 1;
end
end
bwImg = imdilate(bwImg, strel('disk', 3));
bwImg = uint8(bwImg) * 255;
imwrite(bwImg, [OUT_DIR_C '/' imgNames{i}]);
I found the pixel size of contour is 6 when disk equals to 3, and it will cover pixels (==3) inside nucleis and outside nucleis (==3). And if contours are thick,the prediction covers more pixels inside the nucleis. What's more, if they are overlapped, the pixels outside this nucleis may cover some pixels inside other nucleis. It's hard to discriminate them.
Do I misunderstand?
hi,i have trained the model successful,then i want to make a test by a new image and create the contour and segment image or label consistent with this test image,so i want know weather just to run the "images=bbbc006.inputs(eval_data=FLAGS.eval_data)" and "c_fuse, s_fuse = bbbc006.inference(images, train=False)"under the pre-train model or not ,
i will be appreciate for your response.
can you upload your pretained model?
I make a same network, but it hard to detect anything...So I want to make sure if it is:
One is the whole segment label, and the other one is the boundary with dilate-->disk(3)?
thanks
I want to know how to calculate the object-level dice loss
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.