paulcx

@Zardinality What do you mean by using the recomplied version? g++ is 5.40

@Zardinality Not yet. The solution does not work for g++5.4 at least.

The resnet 101 is definitely a good choice to try for better performance.

Would be possible to export/save the features and nodule detection results (ROIs maybe)?

Hi @lfz Is there a way we can simply export the bboxes coordinates and visualize them while testing the detection?

@lfz Thank you. According to the data_classifier, for idx in idcs: pbb = np.load(os.path.join(bboxpath,idx+'_pbb.npy')) pbb = pbb[pbb[:,0]>config['conf_th']] pbb = nms(pbb, config['nms_th']) print('bboxes: ',pbb.shape) lbb = np.load(os.path.join(bboxpath,idx+'_lbb.npy')) pbb_label = [] for...

@aqibsaeed Thanks. Is this dataset same as the data used in this jupyter notebook example?

@aqibsaeed What is the difference between this cleaned version and the dataset used in this repo. It would be very nice if you can provide some example of it to...

it's inspired. I would double check the custom codes. btw, is rust router always disabled if the custom codes are used? if not, what specifications should be followed?

After conducting some tests, I can verify @Narsil's observations regarding the Fast version of the tokenizer. The issue of encountering a "loading timeout" arises from a failure in the automatic...