U-Net-PyTorch icon indicating copy to clipboard operation
U-Net-PyTorch copied to clipboard
verified publisher icon jonnedtc

Running out of memory Bandwidth

Open alizeshana opened this issue 8 years ago • 1 comments

Is there a workaround to train the data on a 2GB GPU?

alizeshana avatar Nov 15 '17 05:11 alizeshana

MemoryError Traceback (most recent call last) in () 20 data= np.transpose(data, (2, 0, 1)) # transpose so the z-axis (slices) are the first dimension 21 #print(type(data)) ---> 22 output = np.zeros((len(data), 512, 512))[0] # save output here 23 #print(data) 24 # print(type(output))

MemoryError:

how to solve these issue??

aarativasane89 avatar Jul 02 '19 09:07 aarativasane89