Use expression in tutorial

[joaquimg]

No description provided.

[matbesancon]

Does this change anything performance-wise?

[codecov]

Codecov Report

Merging #226 (06003d7) into master (7094aed) will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master     #226   +/-   ##
=======================================
  Coverage   91.21%   91.21%           
=======================================
  Files           8        8           
  Lines         842      842           
=======================================
  Hits          768      768           
  Misses         74       74           

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 7094aed...06003d7. Read the comment docs.

[joaquimg]

this example is too small.
so with no @expresion

julia> @benchmark ᾱ, ∂ᾱ, msē = descent(0.10, 500)
BenchmarkTools.Trial: 13 samples with 1 evaluation.
 Range (min … max):  354.719 ms … 513.808 ms  ┊ GC (min … max): 0.00% … 1.83%
 Time  (median):     405.427 ms               ┊ GC (median):    1.98%
 Time  (mean ± σ):   409.797 ms ±  48.666 ms  ┊ GC (mean ± σ):  1.25% ± 0.97%

  ██ ██   █        █ ██  █         █      ██                  █
  ██▁██▁▁▁█▁▁▁▁▁▁▁▁█▁██▁▁█▁▁▁▁▁▁▁▁▁█▁▁▁▁▁▁██▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁█ ▁
  355 ms           Histogram: frequency by time          514 ms <

 Memory estimate: 44.53 MiB, allocs estimate: 490490.

and with @expresion

julia> @benchmark ᾱ, ∂ᾱ, msē = descent(0.10, 500)
BenchmarkTools.Trial: 18 samples with 1 evaluation.
 Range (min … max):  248.046 ms … 353.070 ms  ┊ GC (min … max): 0.00% … 2.31%
 Time  (median):     291.233 ms               ┊ GC (median):    2.38%
 Time  (mean ± σ):   289.772 ms ±  27.178 ms  ┊ GC (mean ± σ):  1.60% ± 1.29%

  ▃                       ▃█      ▃
  █▁▇▁▁▁▁▁▇▁▇▁▁▁▁▁▁▁▁▁▁▁▇▁██▁▁▇▁▁▇█▁▁▁▁▇▁▁▁▁▁▁▁▇▁▁▁▁▁▁▁▁▁▁▁▁▁▁▇ ▁
  248 ms           Histogram: frequency by time          353 ms <

 Memory estimate: 44.53 MiB, allocs estimate: 490490.

But @expression is the correct JuMP way of doing performant thing, so we want to stimulate users to do it.