Ajith

@art-programmer could you please reply for this

@Tao-Zhou i think its from `https://github.com/zhixuhao/unet/blob/master/dataPrepare.ipynb ` ` image_arr,mask_arr = geneTrainNpy("data/membrane/train/aug/","data/membrane/train/aug/") #np.save("data/image_arr.npy",image_arr) #np.save("data/mask_arr.npy",mask_arr) ` The above lines first 60 are results of data augumentation done by zhixuhao if you run...

@PanPan0210 try with trainUnet.ipynb . which file are you running?

@PanPan0210 @Pmac23 i am able to execute . check your path for files and imports . keras version

@sagardatascientists could you use all the merges like below code ` merge6 = concatenate([drop4,up6],axis=3)` coudl you provide your python version ? .

@sagardatascientists I have modified the code for concatenate() and transpose but i didnt check the accuracy rate from up6 layer to conv 10 you need to provide the following code...

it depends on your gpu , batch size .

@huangfuts could you tell how you solved it my mail id is [email protected]

@scuShaker when i used pretrained model its giving me infinity and nan in results , could u tell which one of three pretrained model and cgf files u used ?

please find screen shot, also i used the test set for evaluation