# Segmented\_sum in Knet

**URL:** https://discourse.julialang.org/t/segmented-sum-in-knet/3396
**Category:** Machine Learning
**Tags:** question, knet
**Created:** [April 27, 2017, 5:55am UTC](https://discourse.julialang.org/t/segmented-sum-in-knet/3396 "2017-04-27T05:55:03Z")
**Posts on this page:** 1
**Page:** 1

<div class="post-metadata">

### Author: ![Tomas\_Pevny](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/tomas_pevny/32/25466_2.png) [@Tomas\_Pevny](https://discourse.julialang.org/u/Tomas_Pevny)
#### Post date: [April 27, 2017, 5:55am UTC](https://discourse.julialang.org/t/segmented-sum-in-knet/3396/1 "2017-04-27T05:55:03Z")

</div>

Hello, I have started to play with Knet, but I am unable to reproduce my work related to multi-instance learning in it, since I need segmented\_sum operation from tensorflow. I have created a small example

> using Knet  
> using MLLabelUtils
> 
> function segmented\_sum(x,bags)  
> xx=zeros(length(bags),size(x,2))  
> for (i,bag) in enumerate(bags)  
> xx[i,:]=sum(x[bag,:],1)  
> end  
> return(xx)  
> end
> 
> predict(w,x,bags) = segmented\_sum(xw[1] .+ w[2],bags)w[3] .+ w[4]  
> function softmaxloss(w,x,bags,ygold)  
> ypred = predict(w,x,bags)  
> ynorm = ypred .- logsumexp(ypred,2)  
> -sum(ygold .\* ynorm) / size(ygold,1)  
> end  
> lossgradient = grad(softmaxloss)  
> x=randn(100,5)  
> bags=[i:i+9 for i in 1:10:100]  
> ygold=convertlabel(LabelEnc.OneOfK{Float64},rand(1:2,10), obsdim = 1)  
> d=size(x,2)  
> k=4  
> w=[randn(d,k),randn(1,k),randn(k,2),randn(1,2)]  
> g = lossgradient(w,x,bags,ygold)

The problematic part is xx[i,:]=sum(x[bag,:],1) where I get an error  
Cannot convert an object of type AutoGrad.Rec{Array{Float64,2}} to an object of type Float64  
Can anyone point me, how to fix this?

I have found that I can implement segmented\_sum as

> function segmented\_sum(x,bags)  
> mapreduce(b-\>sum(x[b,:],1),vcat,bags)  
> end

But this is terribly slow.

Thank for any help.