Website: Q4 improvements and cleanup

[siddharth-krishna]

By end Feb:

• Home: SGM on intersection of benchmarks solved by all #114
• Website: add solvers page & relative performance plot from Matthias #115
• Benchmark summary table from slides (or nice graphs) #119

By Mar 17:

• Chore: add pre-commit formatters and linters for NextJS website #120
• Plots: zoom & pan features like plotly
• Results table: custom sorting for Solved Benchmarks and SGM Runtime #116
• Website: some improvements to the filters #121
• Icons: use same kind of table sorting icons/arrows in all dashboards
• Benchmarks: link to source / download each bench

Week of Apr 7:

• Website: Add benchmark set summary slides and table #128
• Landing: contact form: recaptcha and link to 3rd party service (e.g. Sendgrid)
• Landing: issues/stars etc are hardcoded, query GH instead (See Website: Add Google reCAPTCHA to Contact Form #139)

Week of Apr 14:

• Design: fit more on large screens, check on mobile
• More robust sorting code: Website: Sort in Benchmarks and SGM Runtime in Results table #142
• Add contributors to metadata files and show on benchmark details page #145

Before May 5:

• Website: Responsive UI, updates from Figma #135
• Update home dashboard with new tabbed design and filters on the left (wait for Figma to be ready)
• Update benchmark summary graph with new size categorization and "realistic"
• Once timeout is added to results CSV, use this timeout in SGM calculation
• Website: add filter for Realistic benchmarks #174

Before public launch:

• Fix SGM memory calculation for different machine configurations #180
• Performance history misleading as L instances only run on latest solvers #183
• Add links to logs & results for each benchmark instance to the benchmark details page #163
• Website: tooltips and hover effect for dashboard sidebars #153
• Website: fix benchmark instance table #190
• Relative performance plot: how do we handle TOs and other errors? #188
• Website: Fix filter issue #176
• Benchmark Set dashboard: add filters for sizes and realistic #177
• Website: consistent text and spacing styling everywhere using CSS classes #178
• Use log scale for runtime in Runtime vs Memory and Compare Solvers plots #201
• History dashboard improvements #202
• Filters are jumpy because the "only" appears on hover only #229
• Display hostname & runID information for each benchmark in the tooltip of the results plot on home & benchmark dashboards
• Use new color palette from Figma for all graphs/charts (wait for Figma to be ready)
• Landing page:
  • links on top bar do not match order of sections (benchmarks, contributions, mission, etc)
  • Font of "Read more about our methodology" too small, "here" too big, make consistent with previous paragraph
  • Add footer links

Backlog:

• Full results: select columns that are displayed
• Getting started popup on each page & help button
• Top right links on all dashboards: FAQs, issues, and github repo
• Home: can the tooltips have a different background color to stand out more?
• Refactor code base to have less lines of code for each component

[jacek-oet]

@siddharth-krishna
Can you give me more details about these issues? (what will we show, will we create a new page? ...)

• 115 Benchmark details: add relative performance plot from Matthias
• Benchmark summary table from slides (or nice graphs)

[jacek-oet]

@siddharth-krishna

Question 1:
Here is a data of benchmark in metadata

pglib_opf_case2848:
  Short description:
  Model name: PowerModels
  Version:
  Technique: MILP
  Kind of problem: Steady-state optimal power flow
  Sectors: Power
  Time horizon: N/A
  MILP features: Transmission switching
  Sizes:
  - Name: 2848-NA
    Size: L
    URL: https://raw.githubusercontent.com/jump-dev/open-energy-modeling-benchmarks/main/instances/PowerModelsOTS_pglib_opf_case2848_rte.m-8546f681e53f8663080268242343f948670d9ddcd40ed4b2da6ab94ec1e0fc05.mps.gz
    Spatial resolution: 2848
    Temporal resolution: NA
    N. of constraints: 25505
    N. of variables:
    N. of continuous variables: 25507
    N. of integer variables: 0
    N. of binary variables: 3776

Benchmarks: link to source / download each bench

Does this mean we should add a source column to the Instances table, where we'll store the URL from each size's metadata?

Question 2:

Filters: new filter design, easy ways to select all and select only one option

Should we add a checkbox in the filter header that allows users to select all items?

[siddharth-krishna]

Q1: The results/metadata.yaml file now has the direct URL for each benchmark instance. Perhaps we could move the Download button from the top of the benchmark to the Instances table like you suggest and add a new column for it? The column can be called "LP/MPS FILE"

Q2: Nice idea, let's go with that, thanks. :)
I assume the new check box will appear checked when all options are selected?