# Improving performance of item response model in Turing.jl

**URL:** <https://discourse.julialang.org/t/improving-performance-of-item-response-model-in-turing-jl/86249>\
**Category:** Performance\
**Tags:** turing\
**Created:** [August 24, 2022, 8:52am UTC](https://discourse.julialang.org/t/improving-performance-of-item-response-model-in-turing-jl/86249 "2022-08-24T08:52:12Z")\
**Posts on this page:** 1\
**Showing post:** 7

<div class="post-metadata">

**Author:** ![p-gw](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/p-gw/32/210518_2.png) [@p-gw](https://discourse.julialang.org/u/p-gw)\
**Post date:** [August 25, 2022, 8:42am UTC](https://discourse.julialang.org/t/improving-performance-of-item-response-model-in-turing-jl/86249/7 "2022-08-25T08:42:08Z")

</div>

Yes, this is the same model. I am aware of the LazyArray “trick”, but unfortunately it does not work for this particular model. In the thread I linked in the initial post a possible regression was discussed ([#122](https://discourse.julialang.org/t/making-turing-fast-with-large-numbers-of-parameters/69072/122)) but then this should affect all models using `LazyArray`, no? I then just followed what @dlakelan described and summed everything manually using `@addlogprob!`.

Here is the benchmark for `@addlogprob!` vs `LazyArray` version of the model (I had to reduce to P = 100 because of runtime):

```julia
@model function irt_lazy(y, i, p; max_i=maximum(i), max_p=maximum(p))
    theta ~ MvNormal(I(max_p))
    beta ~ MvNormal(I(max_i))
    y ~ arraydist(LazyArray(@~ BernoulliLogit.(theta[p] - beta[i])))
end

@benchmark sample($m3, $alg, $n_samples, init_params=$zeros(120)) 
BenchmarkTools.Trial: 1 sample with 1 evaluation.
 Single result which took 10.272 s (0.72% GC) to evaluate,
 with a memory estimate of 1.06 GiB, over 4902514 allocations.

@benchmark sample($m_lazy, $alg, $n_samples, init_params=$zeros(120)) 
BenchmarkTools.Trial: 1 sample with 1 evaluation.
 Single result which took 350.161 s (0.02% GC) to evaluate,
 with a memory estimate of 1.24 GiB, over 5181966 allocations.

```

So the `LazyArray` version takes 35 times longer than the `@addlogprob!` version.

---

_[View the full topic](https://discourse.julialang.org/t/improving-performance-of-item-response-model-in-turing-jl/86249)._