u-net
u-net copied to clipboard
U-Net: Convolutional Networks for Biomedical Image Segmentation
Traceback (most recent call last): File "data.py", line 189, in detseg() File "data.py", line 79, in detseg imgs_train = np.load('imgs_train.npy') File "/usr/local/lib/python3.6/dist-packages/numpy/lib/npyio.py", line 372, in load fid = open(file, "rb")...
I run files successfully after updating model but I cannot get grey scale images of test data. Please explain how to extract them?
` up6 = merge([Convolution2D(256, 2, 2,activation='relu', border_mode='same')(UpSampling2D(size=(2, 2))(conv5)), conv4], mode='concat', concat_axis=1)` concat_axis=1,why not 0? the input size is [channels height width],so I think axis=1 corresponds to height but not channel....
What is the effect of the function detseg()?
Creating training images... ------------------------------ Done: 0/5635 images Done: 100/5635 images Traceback (most recent call last): File "data.py", line 186, in create_train_data() File "data.py", line 51, in create_train_data imgs_mask[i] = img_mask...
The output should be the mask. Why the channel number is 2?
Have you compared the score between resnet and u-net? Which performs better?
I´m trying to run the Network but always running into the Error below: ``` Caused by op 'mul', defined at: File "train.py", line 206, in train_and_predict() File "train.py", line 153,...