I've built Enzo in ubuntu and I've successfully run the first test problem in terminal mode and also used $ yt plot DD0000/DD0000 in the terminal to generate the output slice plots, as per the enzo guide. I'm trying to replicate the slice plotting in a jupyter notebook. When I try to load the DD data using ds = yt.load('/home/ian/enzo-dev/run/Hydro/Hydro-3D/CollapseTestNonCosmological/DD0000') I get the error YTUnidentifiedDataType: Could not determine input format from /home/ian/enzo-dev/run/Hydro/Hydro-3D/CollapseTestNonCosmological/DD0000' By looking at the CollapseTestNonCosmological.enzo file, I see that the data is an AMR format. I therefore tried to load it using ds = yt.load_amr_grids('/home/ian/enzo-dev/run/Hydro/Hydro-3D/CollapseTestNonCosmological/DD0000') but this failed because it was missing the domain_dimensions argument. I think this means it needs to have grid_data added as the first argument: ds = yt.load_amr_grids(grid_data, '/home/ian/enzo-dev/run/Hydro/Hydro-3D/CollapseTestNonCosmological/DD0000') I can try and work out how to define the grid_data but I don't understand why doing so wasn't necessary when I simply ran $ yt plot DD0000/DD0000 in terminal mode. Does yt plot extract the domain_dimensions automatically from the outputs, and if so is there an easy way of doing this in jupyter's yt.load_amr_grids, without having to manually set up grid_data?