polimi-ispl / deep_prior_interpolation Goto Github PK

As the network size is proportional to the dataset size, and the GPU memory is limited, some parallelism has to be implemented.

A naive approach is to split the dataset into possibly-overlapped windows to be processed separately and then the results are merged. This can be done sequentially (as in this repo) or with a DataParallel strategy.

On the other hand, in order to process all the data at once, one might want to adopt a ModelParallel strategy.

Ok, its fine. Thanks for not helping.

Hii, developers,

hello developer,thank you for the source code,but in the process of my verification,I found that the patch[1] was always the best and the patch[0] and patch[2] parts were poor.my input shape is(64,32,32),other parameters are set as shown in the figure.

Hello, I am chiseo. Really, this is an amazing tool for interpolating the 3d seismic data. Could you please suggest me how to use this tool for 2d seismic data. In the 2d case i have the data dimension of mxn. However according to the instruction provided "If you have 2D native datasets, please add an extra axis". Thus i have added an extra axis by using the command inputdata[:, :, newaxis], in python and in this way i convert the 2d native data to 3d and now the data dimension is mxnx1. Though the data is now in 3d it gives something like "Invalid shape (n,)......So, i need some support from developers rectifying this small error.Thanks and looking forward to hearing from you.

Hii deep_prior_interplation developer, While running the second example i am getting error like T.build_input(imgshape)
TypeError: build_input() takes 1 positional argument but 2 were given, however i get The image shape is (512, 128, 128, 1) .Please suggest a solution.Thanks.