eachobs
	
function defined in module 
	MLUtils
			eachobs(data; kws...)
			Return an iterator over 
			data.
			Supports the same arguments as 
	
			
			DataLoader. The 
			batchsize default is 
			-1 here while it is 
			1 for 
			DataLoader.
			
			
			
			X
			 
			=
			 
			
			rand
			(
			4
			,
			100
			)
			
			
			
			for
			
			 
			x
			 
			in
			 
			
			eachobs
			(
			X
			)
			
			
    
			# loop entered 100 times
			
    
			
			@
			assert
			
			 
			
			typeof
			(
			x
			)
			 
			<:
			 
			
			Vector
			{
			Float64
			}
			
    
			
			@
			assert
			
			 
			
			size
			(
			x
			)
			 
			==
			 
			
			(
			4
			,
			)
			
			end
			
			
			# mini-batch iterations
			
			
			for
			
			 
			x
			 
			in
			 
			
			eachobs
			(
			X
			,
			 
			
	
			batchsize
			=
			10
			)
			
			
    
			# loop entered 10 times
			
    
			
			@
			assert
			
			 
			
			typeof
			(
			x
			)
			 
			<:
			 
			
			Matrix
			{
			Float64
			}
			
    
			
			@
			assert
			
			 
			
			size
			(
			x
			)
			 
			==
			 
			
			(
			4
			,
			10
			)
			
			end
			
			
			# support for tuples, named tuples, dicts
			
			
			for
			
			 
			
			(
			x
			,
			 
			y
			)
			 
			in
			 
			
			eachobs
			(
			
			(
			X
			,
			 
			Y
			)
			)
			
			
    
			# ...
			
			endThere is
			1
			method for MLUtils.eachobs:
		
The following pages link back here:
Custom learning tasks, Saving and loading models for inference
FastAI.jl , blocks/label.jl , datasets/Datasets.jl , FastTabular.jl , recipes.jl , FastVision.jl , encodings/imagepreprocessing.jl , recipes.jl , MLUtils.jl , deprecations.jl , eachobs.jl