From 8a378594dbb04d641ed3619397d91f6f21e61209 Mon Sep 17 00:00:00 2001 From: Dax Kellie Date: Tue, 26 Apr 2022 14:37:16 +1000 Subject: [PATCH] Fix footer on big screens, add science icon --- _footer.html | 46 +- .../Stevenson_Matilda/Stevenson_Matilda.html | 1048 ++++++++- _people/Westgate_Martin/Westgate_Martin.html | 1062 ++++++++- _site/about.html | 51 +- _site/galah.html | 68 +- _site/images/plotly/taxonomy_navigation.html | 1933 ++++++++++++++++- _site/index.html | 70 +- _site/info_ala-labs-maintenance.html | 207 +- .../Balasubramaniam_Shandiya/index.html | 34 +- _site/people/Buyan_Amanda/index.html | 34 +- _site/people/Fenker_Jess/index.html | 34 +- _site/people/Kellie_Dax/index.html | 34 +- _site/people/Stevenson_Matilda/index.html | 1082 ++++++++- _site/people/Torresan_Olivia/index.html | 34 +- _site/people/Westgate_Martin/index.html | 1096 +++++++++- _site/people/Wraith_Jenna/index.html | 34 +- _site/people/people.json | 44 +- _site/posts.html | 51 +- .../index.html | 34 +- .../index.html | 34 +- .../index.html | 34 +- _site/posts/posts.json | 12 +- _site/search.json | 24 +- .../header-attrs.js | 0 _site/test_style-guide.html | 248 ++- .../figure-html5/unnamed-chunk-10-1.png | Bin 54513 -> 18568 bytes .../figure-html5/unnamed-chunk-8-1.png | Bin 26810 -> 5538 bytes .../figure-html5/unnamed-chunk-8-2.png | Bin 44020 -> 7993 bytes images/plotly/taxonomy_navigation.html | 1933 ++++++++++++++++- index.Rmd | 8 +- 30 files changed, 8692 insertions(+), 597 deletions(-) rename _site/site_libs/{header-attrs-2.13 => header-attrs-2.11}/header-attrs.js (100%) diff --git a/_footer.html b/_footer.html index 70397ead..2c9d7b11 100644 --- a/_footer.html +++ b/_footer.html @@ -7,13 +7,13 @@ .copyright-container { display: flex; width: 100%; - margin-bottom: 30px; + margin-bottom: 30px; margin-top: 30px; } .copyright-row { width: 100%; - height: 60px; + height: 66px; position: relative; margin-bottom: -1.75vw; background-color: #c44d34; @@ -71,14 +71,14 @@ /*-- Logos and acknowledgement --*/ -.footer-container { +.footer-container { display: flex; width: 100%; max-width: 1140px; padding-left: 15px; padding-right: 15px; padding-top: 50px; - padding-bottom: 50px; + padding-bottom: 50px; margin: auto; } @@ -161,7 +161,7 @@ margin: 4px 2px; transition-duration: 0.3s; cursor: pointer; - background-color: white; + background-color: white; border: 2px solid #E06E53; border-radius: 10px; } @@ -184,7 +184,7 @@ margin-top: 20px; margin-bottom: 20px; } - + .footer-container { width: 100%; max-width: 100%; @@ -204,15 +204,15 @@ width: 20em; margin: auto; } - + .copyright-column-left { width: 50%; } - + .copyright-column-right { width: 50%; } - + .copyright-row { height: 100px; margin-bottom: -3.75vw; @@ -243,32 +243,32 @@
- +
-

The ALA is made possible by contributions from its partners, is supported - by - NCRIS, is hosted by +

The ALA is made possible by contributions from its partners, is supported + by + NCRIS, is hosted by CSIRO, and is the Australian node of GBIF.

- NCRIS logo - CSIRO logo - GBIF logo @@ -276,10 +276,10 @@
diff --git a/_people/Stevenson_Matilda/Stevenson_Matilda.html b/_people/Stevenson_Matilda/Stevenson_Matilda.html index 3c7f6dd7..2da6f2c8 100644 --- a/_people/Stevenson_Matilda/Stevenson_Matilda.html +++ b/_people/Stevenson_Matilda/Stevenson_Matilda.html @@ -92,8 +92,8 @@ - - + + @@ -108,7 +108,7 @@ @@ -1434,16 +1434,1031 @@ - - - - - - - - - - + + + + + + + + + + @@ -1455,7 +2470,7 @@ @@ -1476,7 +2491,10 @@

Matilda Stevenson

Posts

-

Hex maps for species occurrence data
Hex maps are a neat way to represent spatial information. Here, we show how to draw one using the most common species in the ALA database

+

Hex +maps for species occurrence data
Hex maps are a neat way to +represent spatial information. Here, we show how to draw one using the +most common species in the ALA database

diff --git a/_people/Westgate_Martin/Westgate_Martin.html b/_people/Westgate_Martin/Westgate_Martin.html index 86ac25e0..c71ae16c 100644 --- a/_people/Westgate_Martin/Westgate_Martin.html +++ b/_people/Westgate_Martin/Westgate_Martin.html @@ -92,8 +92,8 @@ - - + + @@ -108,7 +108,7 @@ @@ -1434,16 +1434,1031 @@ - - - - - - - - - - + + + + + + + + + + @@ -1455,7 +2470,7 @@ @@ -1478,7 +2493,11 @@

Bio

-Martin leads the Science & Decision Support Team. He holds a doctorate in landscape ecology and conservation biology from the Australian National University. His work focuses on conceptual, computational and statistical tools to better understand patterns in nature. +Martin leads the Science & Decision Support Team. He holds a +doctorate in landscape ecology and conservation biology from the +Australian National University. His work focuses on conceptual, +computational and statistical tools to better understand patterns in +nature.
@@ -1515,7 +2534,18 @@

Bio

Posts

-

Sunburst plots for taxonomic data
Since version 1.3.1 of galah, it has been possible to download taxonomic data using a ‘tree’ format from the data.tree package. Here I’ll demonstrate some ideas for plotting these trees using circular diagrams.

Creating a color palette from an image
There are hundreds of color palettes in the R ecosystem, but sometimes we might want to use colors from a specific image. Here I show how to use the paletter package to create a color palette for the 2020 Eucalypt of the Year

+

Sunburst +plots for taxonomic data
Since version 1.3.1 of +galah, it has been possible to download taxonomic data +using a ‘tree’ format from the data.tree package. Here I’ll +demonstrate some ideas for plotting these trees using circular +diagrams.

+Creating +a color palette from an image
There are hundreds of color +palettes in the R ecosystem, but sometimes we might want to use colors +from a specific image. Here I show how to use the paletter +package to create a color palette for the 2020 Eucalypt of the +Year

diff --git a/_site/about.html b/_site/about.html index 5ff33430..238f1090 100644 --- a/_site/about.html +++ b/_site/about.html @@ -2814,7 +2814,7 @@

${suggestion.title}

color: var(--hover-color, white); } - + @@ -3003,7 +3003,8 @@

Shandiya Balasubramaniam

-
+
@@ -3027,13 +3028,13 @@

Shandiya Balasubramaniam

.copyright-container { display: flex; width: 100%; - margin-bottom: 30px; + margin-bottom: 30px; margin-top: 30px; } .copyright-row { width: 100%; - height: 60px; + height: 66px; position: relative; margin-bottom: -1.75vw; background-color: #c44d34; @@ -3091,14 +3092,14 @@

Shandiya Balasubramaniam

/*-- Logos and acknowledgement --*/ -.footer-container { +.footer-container { display: flex; width: 100%; max-width: 1140px; padding-left: 15px; padding-right: 15px; padding-top: 50px; - padding-bottom: 50px; + padding-bottom: 50px; margin: auto; } @@ -3181,7 +3182,7 @@

Shandiya Balasubramaniam

margin: 4px 2px; transition-duration: 0.3s; cursor: pointer; - background-color: white; + background-color: white; border: 2px solid #E06E53; border-radius: 10px; } @@ -3204,7 +3205,7 @@

Shandiya Balasubramaniam

margin-top: 20px; margin-bottom: 20px; } - + .footer-container { width: 100%; max-width: 100%; @@ -3224,15 +3225,15 @@

Shandiya Balasubramaniam

width: 20em; margin: auto; } - + .copyright-column-left { width: 50%; } - + .copyright-column-right { width: 50%; } - + .copyright-row { height: 100px; margin-bottom: -3.75vw; @@ -3261,32 +3262,32 @@

Shandiya Balasubramaniam

- +
-

The ALA is made possible by contributions from its partners, is supported - by - NCRIS, is hosted by +

The ALA is made possible by contributions from its partners, is supported + by + NCRIS, is hosted by CSIRO, and is the Australian node of GBIF.

- NCRIS logo - CSIRO logo - GBIF logo @@ -3294,10 +3295,10 @@

Shandiya Balasubramaniam

diff --git a/_site/galah.html b/_site/galah.html index b91c27cb..b9419b73 100644 --- a/_site/galah.html +++ b/_site/galah.html @@ -2814,7 +2814,7 @@

${suggestion.title}

color: var(--hover-color, white); } - + @@ -2909,9 +2909,20 @@

galah

galah

-

galah is an R interface to biodiversity data hosted by the Atlas of Living Australia (ALA). It enables users to locate and download species occurrence records (observations, specimens, eDNA records, etc.), taxonomic information, or associated media such as images or sounds, and to restrict their queries to particular taxa or locations.

-

Visit the galah R package website to learn more about how to use galah

-

If you have any questions, comments, or spot any bugs, email us or report an issue on our GitHub page

+

galah is an R interface to biodiversity data hosted by +the Atlas of Living Australia (ALA). It enables users to locate and +download species occurrence records (observations, specimens, eDNA +records, etc.), taxonomic information, or associated media such as +images or sounds, and to restrict their queries to particular taxa or +locations.

+

Visit the galah +R package website to learn more about how to use +galah

+

If you have any questions, comments, or spot any bugs, email us or report an +issue on our GitHub page

Get started

@@ -2922,7 +2933,8 @@

Get started

-
+
@@ -2946,13 +2958,13 @@

Get started

.copyright-container { display: flex; width: 100%; - margin-bottom: 30px; + margin-bottom: 30px; margin-top: 30px; } .copyright-row { width: 100%; - height: 60px; + height: 66px; position: relative; margin-bottom: -1.75vw; background-color: #c44d34; @@ -3010,14 +3022,14 @@

Get started

/*-- Logos and acknowledgement --*/ -.footer-container { +.footer-container { display: flex; width: 100%; max-width: 1140px; padding-left: 15px; padding-right: 15px; padding-top: 50px; - padding-bottom: 50px; + padding-bottom: 50px; margin: auto; } @@ -3100,7 +3112,7 @@

Get started

margin: 4px 2px; transition-duration: 0.3s; cursor: pointer; - background-color: white; + background-color: white; border: 2px solid #E06E53; border-radius: 10px; } @@ -3123,7 +3135,7 @@

Get started

margin-top: 20px; margin-bottom: 20px; } - + .footer-container { width: 100%; max-width: 100%; @@ -3143,15 +3155,15 @@

Get started

width: 20em; margin: auto; } - + .copyright-column-left { width: 50%; } - + .copyright-column-right { width: 50%; } - + .copyright-row { height: 100px; margin-bottom: -3.75vw; @@ -3180,32 +3192,32 @@

Get started

- +
-

The ALA is made possible by contributions from its partners, is supported - by - NCRIS, is hosted by +

The ALA is made possible by contributions from its partners, is supported + by + NCRIS, is hosted by CSIRO, and is the Australian node of GBIF.

- NCRIS logo - CSIRO logo - GBIF logo @@ -3213,10 +3225,10 @@

Get started

diff --git a/_site/images/plotly/taxonomy_navigation.html b/_site/images/plotly/taxonomy_navigation.html index b51c3aa8..fb2b3660 100644 --- a/_site/images/plotly/taxonomy_navigation.html +++ b/_site/images/plotly/taxonomy_navigation.html @@ -3,23 +3,1934 @@ plotly - - - - - - - - + + + + + + + +
-
+
- - + + diff --git a/_site/index.html b/_site/index.html index 55f9216b..70905e05 100644 --- a/_site/index.html +++ b/_site/index.html @@ -2814,7 +2814,7 @@

${suggestion.title}

color: var(--hover-color, white); } - + @@ -2923,7 +2923,10 @@

${suggestion.title}

Welcome to ALA Labs

-

This site is a resource for coding projects that use data sourced from the ALA. We hope that users will find interesting content, whether their focus is ecological modelling, data visualisation, or simply investigating the natural world through a digital lens. Enjoy!

+

This site is a resource for coding projects that use data sourced +from the ALA. We hope that users will find interesting content, whether +their focus is ecological modelling, data visualisation, or simply +investigating the natural world through a digital lens. Enjoy!

@@ -2934,7 +2937,8 @@

-Click to see related posts on each topic +Click to see related posts on each +topic
@@ -2978,7 +2982,8 @@

-Click to see related posts on each topic +Click to see related posts on each +topic
@@ -3027,7 +3032,7 @@

- Openness + Openness

Data are most useful when they are widely available and easy to use. We try to encourage the sharing of tools that make data from the Atlas of Living Australia more open and useful for everyone

@@ -3035,8 +3040,8 @@

- - Scientific transparency + + Scientific transparency

Transparency is necessary for reproducible science. We encourage that decisions, methods and deviations in our workflow are clear and transparent from the planning phase, to the retrieval and analysis of data, to the final output

@@ -3045,14 +3050,15 @@

- Robust methods + Robust methods

No method or analytic procedure is perfect. We value consideration for strengths and limitations of each method or analysis to help choose the methods that provide robust, reliable results

-
+
@@ -3076,13 +3082,13 @@

.copyright-container { display: flex; width: 100%; - margin-bottom: 30px; + margin-bottom: 30px; margin-top: 30px; } .copyright-row { width: 100%; - height: 60px; + height: 66px; position: relative; margin-bottom: -1.75vw; background-color: #c44d34; @@ -3140,14 +3146,14 @@

/*-- Logos and acknowledgement --*/ -.footer-container { +.footer-container { display: flex; width: 100%; max-width: 1140px; padding-left: 15px; padding-right: 15px; padding-top: 50px; - padding-bottom: 50px; + padding-bottom: 50px; margin: auto; } @@ -3230,7 +3236,7 @@

margin: 4px 2px; transition-duration: 0.3s; cursor: pointer; - background-color: white; + background-color: white; border: 2px solid #E06E53; border-radius: 10px; } @@ -3253,7 +3259,7 @@

margin-top: 20px; margin-bottom: 20px; } - + .footer-container { width: 100%; max-width: 100%; @@ -3273,15 +3279,15 @@

width: 20em; margin: auto; } - + .copyright-column-left { width: 50%; } - + .copyright-column-right { width: 50%; } - + .copyright-row { height: 100px; margin-bottom: -3.75vw; @@ -3310,32 +3316,32 @@

- +
-

The ALA is made possible by contributions from its partners, is supported - by - NCRIS, is hosted by +

The ALA is made possible by contributions from its partners, is supported + by + NCRIS, is hosted by CSIRO, and is the Australian node of GBIF.

- NCRIS logo - CSIRO logo - GBIF logo @@ -3343,10 +3349,10 @@

diff --git a/_site/info_ala-labs-maintenance.html b/_site/info_ala-labs-maintenance.html index 6fac0122..a731d313 100644 --- a/_site/info_ala-labs-maintenance.html +++ b/_site/info_ala-labs-maintenance.html @@ -1412,13 +1412,13 @@

${suggestion.title}

.distill-site-footer a:hover { color: var(--hover-color, white); } - - + + - - - - + + + + - - - - - + + + + +
- +
-

The ALA is made possible by contributions from its partners, is supported - by - NCRIS, is hosted by +

The ALA is made possible by contributions from its partners, is supported + by + NCRIS, is hosted by CSIRO, and is the Australian node of GBIF.

- NCRIS logo - CSIRO logo - GBIF logo @@ -2162,10 +2229,10 @@

Custom html elements using R functions

diff --git a/_site/people/Balasubramaniam_Shandiya/index.html b/_site/people/Balasubramaniam_Shandiya/index.html index 2e5c4003..3a6cd4e6 100644 --- a/_site/people/Balasubramaniam_Shandiya/index.html +++ b/_site/people/Balasubramaniam_Shandiya/index.html @@ -3978,13 +3978,13 @@

.copyright-container { display: flex; width: 100%; - margin-bottom: 30px; + margin-bottom: 30px; margin-top: 30px; } .copyright-row { width: 100%; - height: 60px; + height: 66px; position: relative; margin-bottom: -1.75vw; background-color: #c44d34; @@ -4042,14 +4042,14 @@

/*-- Logos and acknowledgement --*/ -.footer-container { +.footer-container { display: flex; width: 100%; max-width: 1140px; padding-left: 15px; padding-right: 15px; padding-top: 50px; - padding-bottom: 50px; + padding-bottom: 50px; margin: auto; } @@ -4132,7 +4132,7 @@

margin: 4px 2px; transition-duration: 0.3s; cursor: pointer; - background-color: white; + background-color: white; border: 2px solid #E06E53; border-radius: 10px; } @@ -4155,7 +4155,7 @@

margin-top: 20px; margin-bottom: 20px; } - + .footer-container { width: 100%; max-width: 100%; @@ -4175,15 +4175,15 @@

width: 20em; margin: auto; } - + .copyright-column-left { width: 50%; } - + .copyright-column-right { width: 50%; } - + .copyright-row { height: 100px; margin-bottom: -3.75vw; @@ -4212,12 +4212,12 @@

- +
-

The ALA is made possible by contributions from its partners, is supported - by - NCRIS, is hosted by +

The ALA is made possible by contributions from its partners, is supported + by + NCRIS, is hosted by CSIRO, and is the Australian node of GBIF.

@@ -4230,10 +4230,10 @@

diff --git a/_site/people/Buyan_Amanda/index.html b/_site/people/Buyan_Amanda/index.html index 65ea69c9..bbf0e835 100644 --- a/_site/people/Buyan_Amanda/index.html +++ b/_site/people/Buyan_Amanda/index.html @@ -2946,13 +2946,13 @@

.copyright-container { display: flex; width: 100%; - margin-bottom: 30px; + margin-bottom: 30px; margin-top: 30px; } .copyright-row { width: 100%; - height: 60px; + height: 66px; position: relative; margin-bottom: -1.75vw; background-color: #c44d34; @@ -3010,14 +3010,14 @@

/*-- Logos and acknowledgement --*/ -.footer-container { +.footer-container { display: flex; width: 100%; max-width: 1140px; padding-left: 15px; padding-right: 15px; padding-top: 50px; - padding-bottom: 50px; + padding-bottom: 50px; margin: auto; } @@ -3100,7 +3100,7 @@

margin: 4px 2px; transition-duration: 0.3s; cursor: pointer; - background-color: white; + background-color: white; border: 2px solid #E06E53; border-radius: 10px; } @@ -3123,7 +3123,7 @@

margin-top: 20px; margin-bottom: 20px; } - + .footer-container { width: 100%; max-width: 100%; @@ -3143,15 +3143,15 @@

width: 20em; margin: auto; } - + .copyright-column-left { width: 50%; } - + .copyright-column-right { width: 50%; } - + .copyright-row { height: 100px; margin-bottom: -3.75vw; @@ -3180,12 +3180,12 @@

- +
-

The ALA is made possible by contributions from its partners, is supported - by - NCRIS, is hosted by +

The ALA is made possible by contributions from its partners, is supported + by + NCRIS, is hosted by CSIRO, and is the Australian node of GBIF.

@@ -3198,10 +3198,10 @@

diff --git a/_site/people/Fenker_Jess/index.html b/_site/people/Fenker_Jess/index.html index 885f9baf..cfcd048b 100644 --- a/_site/people/Fenker_Jess/index.html +++ b/_site/people/Fenker_Jess/index.html @@ -2957,13 +2957,13 @@

Bio

.copyright-container { display: flex; width: 100%; - margin-bottom: 30px; + margin-bottom: 30px; margin-top: 30px; } .copyright-row { width: 100%; - height: 60px; + height: 66px; position: relative; margin-bottom: -1.75vw; background-color: #c44d34; @@ -3021,14 +3021,14 @@

Bio

/*-- Logos and acknowledgement --*/ -.footer-container { +.footer-container { display: flex; width: 100%; max-width: 1140px; padding-left: 15px; padding-right: 15px; padding-top: 50px; - padding-bottom: 50px; + padding-bottom: 50px; margin: auto; } @@ -3111,7 +3111,7 @@

Bio

margin: 4px 2px; transition-duration: 0.3s; cursor: pointer; - background-color: white; + background-color: white; border: 2px solid #E06E53; border-radius: 10px; } @@ -3134,7 +3134,7 @@

Bio

margin-top: 20px; margin-bottom: 20px; } - + .footer-container { width: 100%; max-width: 100%; @@ -3154,15 +3154,15 @@

Bio

width: 20em; margin: auto; } - + .copyright-column-left { width: 50%; } - + .copyright-column-right { width: 50%; } - + .copyright-row { height: 100px; margin-bottom: -3.75vw; @@ -3191,12 +3191,12 @@

Bio

- +
-

The ALA is made possible by contributions from its partners, is supported - by - NCRIS, is hosted by +

The ALA is made possible by contributions from its partners, is supported + by + NCRIS, is hosted by CSIRO, and is the Australian node of GBIF.

@@ -3209,10 +3209,10 @@

Bio

diff --git a/_site/people/Kellie_Dax/index.html b/_site/people/Kellie_Dax/index.html index 1731bea8..0b0e3f44 100644 --- a/_site/people/Kellie_Dax/index.html +++ b/_site/people/Kellie_Dax/index.html @@ -3978,13 +3978,13 @@

.copyright-container { display: flex; width: 100%; - margin-bottom: 30px; + margin-bottom: 30px; margin-top: 30px; } .copyright-row { width: 100%; - height: 60px; + height: 66px; position: relative; margin-bottom: -1.75vw; background-color: #c44d34; @@ -4042,14 +4042,14 @@

/*-- Logos and acknowledgement --*/ -.footer-container { +.footer-container { display: flex; width: 100%; max-width: 1140px; padding-left: 15px; padding-right: 15px; padding-top: 50px; - padding-bottom: 50px; + padding-bottom: 50px; margin: auto; } @@ -4132,7 +4132,7 @@

margin: 4px 2px; transition-duration: 0.3s; cursor: pointer; - background-color: white; + background-color: white; border: 2px solid #E06E53; border-radius: 10px; } @@ -4155,7 +4155,7 @@

margin-top: 20px; margin-bottom: 20px; } - + .footer-container { width: 100%; max-width: 100%; @@ -4175,15 +4175,15 @@

width: 20em; margin: auto; } - + .copyright-column-left { width: 50%; } - + .copyright-column-right { width: 50%; } - + .copyright-row { height: 100px; margin-bottom: -3.75vw; @@ -4212,12 +4212,12 @@

- +
-

The ALA is made possible by contributions from its partners, is supported - by - NCRIS, is hosted by +

The ALA is made possible by contributions from its partners, is supported + by + NCRIS, is hosted by CSIRO, and is the Australian node of GBIF.

@@ -4230,10 +4230,10 @@

diff --git a/_site/people/Stevenson_Matilda/index.html b/_site/people/Stevenson_Matilda/index.html index f71a4467..272eec1e 100644 --- a/_site/people/Stevenson_Matilda/index.html +++ b/_site/people/Stevenson_Matilda/index.html @@ -93,8 +93,8 @@ - - + + @@ -110,7 +110,7 @@ @@ -2818,16 +2818,1031 @@

${suggestion.title}

color: var(--hover-color, white); } - - - - - - - - - - + + + + + + + + + + @@ -2903,7 +3918,10 @@

Matilda Stevenson

Posts

-

Hex maps for species occurrence data
Hex maps are a neat way to represent spatial information. Here, we show how to draw one using the most common species in the ALA database

+

Hex +maps for species occurrence data
Hex maps are a neat way to +represent spatial information. Here, we show how to draw one using the +most common species in the ALA database

@@ -2929,13 +3947,13 @@

.copyright-container { display: flex; width: 100%; - margin-bottom: 30px; + margin-bottom: 30px; margin-top: 30px; } .copyright-row { width: 100%; - height: 60px; + height: 66px; position: relative; margin-bottom: -1.75vw; background-color: #c44d34; @@ -2993,14 +4011,14 @@

/*-- Logos and acknowledgement --*/ -.footer-container { +.footer-container { display: flex; width: 100%; max-width: 1140px; padding-left: 15px; padding-right: 15px; padding-top: 50px; - padding-bottom: 50px; + padding-bottom: 50px; margin: auto; } @@ -3083,7 +4101,7 @@

margin: 4px 2px; transition-duration: 0.3s; cursor: pointer; - background-color: white; + background-color: white; border: 2px solid #E06E53; border-radius: 10px; } @@ -3106,7 +4124,7 @@

margin-top: 20px; margin-bottom: 20px; } - + .footer-container { width: 100%; max-width: 100%; @@ -3126,15 +4144,15 @@

width: 20em; margin: auto; } - + .copyright-column-left { width: 50%; } - + .copyright-column-right { width: 50%; } - + .copyright-row { height: 100px; margin-bottom: -3.75vw; @@ -3163,12 +4181,12 @@

- +
-

The ALA is made possible by contributions from its partners, is supported - by - NCRIS, is hosted by +

The ALA is made possible by contributions from its partners, is supported + by + NCRIS, is hosted by CSIRO, and is the Australian node of GBIF.

@@ -3181,10 +4199,10 @@

diff --git a/_site/people/Torresan_Olivia/index.html b/_site/people/Torresan_Olivia/index.html index a66a6180..18a2ba76 100644 --- a/_site/people/Torresan_Olivia/index.html +++ b/_site/people/Torresan_Olivia/index.html @@ -2946,13 +2946,13 @@

.copyright-container { display: flex; width: 100%; - margin-bottom: 30px; + margin-bottom: 30px; margin-top: 30px; } .copyright-row { width: 100%; - height: 60px; + height: 66px; position: relative; margin-bottom: -1.75vw; background-color: #c44d34; @@ -3010,14 +3010,14 @@

/*-- Logos and acknowledgement --*/ -.footer-container { +.footer-container { display: flex; width: 100%; max-width: 1140px; padding-left: 15px; padding-right: 15px; padding-top: 50px; - padding-bottom: 50px; + padding-bottom: 50px; margin: auto; } @@ -3100,7 +3100,7 @@

margin: 4px 2px; transition-duration: 0.3s; cursor: pointer; - background-color: white; + background-color: white; border: 2px solid #E06E53; border-radius: 10px; } @@ -3123,7 +3123,7 @@

margin-top: 20px; margin-bottom: 20px; } - + .footer-container { width: 100%; max-width: 100%; @@ -3143,15 +3143,15 @@

width: 20em; margin: auto; } - + .copyright-column-left { width: 50%; } - + .copyright-column-right { width: 50%; } - + .copyright-row { height: 100px; margin-bottom: -3.75vw; @@ -3180,12 +3180,12 @@

- +
-

The ALA is made possible by contributions from its partners, is supported - by - NCRIS, is hosted by +

The ALA is made possible by contributions from its partners, is supported + by + NCRIS, is hosted by CSIRO, and is the Australian node of GBIF.

@@ -3198,10 +3198,10 @@

diff --git a/_site/people/Westgate_Martin/index.html b/_site/people/Westgate_Martin/index.html index f3b5e8d8..add1427d 100644 --- a/_site/people/Westgate_Martin/index.html +++ b/_site/people/Westgate_Martin/index.html @@ -93,8 +93,8 @@ - - + + @@ -110,7 +110,7 @@ @@ -2818,16 +2818,1031 @@

${suggestion.title}

color: var(--hover-color, white); } - - - - - - - - - - + + + + + + + + + + @@ -2905,7 +3920,11 @@

Bio

-Martin leads the Science & Decision Support Team. He holds a doctorate in landscape ecology and conservation biology from the Australian National University. His work focuses on conceptual, computational and statistical tools to better understand patterns in nature. +Martin leads the Science & Decision Support Team. He holds a +doctorate in landscape ecology and conservation biology from the +Australian National University. His work focuses on conceptual, +computational and statistical tools to better understand patterns in +nature.
@@ -2942,7 +3961,18 @@

Bio

Posts

-

Sunburst plots for taxonomic data
Since version 1.3.1 of galah, it has been possible to download taxonomic data using a ‘tree’ format from the data.tree package. Here I’ll demonstrate some ideas for plotting these trees using circular diagrams.

Creating a color palette from an image
There are hundreds of color palettes in the R ecosystem, but sometimes we might want to use colors from a specific image. Here I show how to use the paletter package to create a color palette for the 2020 Eucalypt of the Year

+

Sunburst +plots for taxonomic data
Since version 1.3.1 of +galah, it has been possible to download taxonomic data +using a ‘tree’ format from the data.tree package. Here I’ll +demonstrate some ideas for plotting these trees using circular +diagrams.

+Creating +a color palette from an image
There are hundreds of color +palettes in the R ecosystem, but sometimes we might want to use colors +from a specific image. Here I show how to use the paletter +package to create a color palette for the 2020 Eucalypt of the +Year

@@ -2968,13 +3998,13 @@

.copyright-container { display: flex; width: 100%; - margin-bottom: 30px; + margin-bottom: 30px; margin-top: 30px; } .copyright-row { width: 100%; - height: 60px; + height: 66px; position: relative; margin-bottom: -1.75vw; background-color: #c44d34; @@ -3032,14 +4062,14 @@

/*-- Logos and acknowledgement --*/ -.footer-container { +.footer-container { display: flex; width: 100%; max-width: 1140px; padding-left: 15px; padding-right: 15px; padding-top: 50px; - padding-bottom: 50px; + padding-bottom: 50px; margin: auto; } @@ -3122,7 +4152,7 @@

margin: 4px 2px; transition-duration: 0.3s; cursor: pointer; - background-color: white; + background-color: white; border: 2px solid #E06E53; border-radius: 10px; } @@ -3145,7 +4175,7 @@

margin-top: 20px; margin-bottom: 20px; } - + .footer-container { width: 100%; max-width: 100%; @@ -3165,15 +4195,15 @@

width: 20em; margin: auto; } - + .copyright-column-left { width: 50%; } - + .copyright-column-right { width: 50%; } - + .copyright-row { height: 100px; margin-bottom: -3.75vw; @@ -3202,12 +4232,12 @@

- +
-

The ALA is made possible by contributions from its partners, is supported - by - NCRIS, is hosted by +

The ALA is made possible by contributions from its partners, is supported + by + NCRIS, is hosted by CSIRO, and is the Australian node of GBIF.

@@ -3220,10 +4250,10 @@

diff --git a/_site/people/Wraith_Jenna/index.html b/_site/people/Wraith_Jenna/index.html index 3693bd49..aeb22e62 100644 --- a/_site/people/Wraith_Jenna/index.html +++ b/_site/people/Wraith_Jenna/index.html @@ -3969,13 +3969,13 @@

Bio

.copyright-container { display: flex; width: 100%; - margin-bottom: 30px; + margin-bottom: 30px; margin-top: 30px; } .copyright-row { width: 100%; - height: 60px; + height: 66px; position: relative; margin-bottom: -1.75vw; background-color: #c44d34; @@ -4033,14 +4033,14 @@

Bio

/*-- Logos and acknowledgement --*/ -.footer-container { +.footer-container { display: flex; width: 100%; max-width: 1140px; padding-left: 15px; padding-right: 15px; padding-top: 50px; - padding-bottom: 50px; + padding-bottom: 50px; margin: auto; } @@ -4123,7 +4123,7 @@

Bio

margin: 4px 2px; transition-duration: 0.3s; cursor: pointer; - background-color: white; + background-color: white; border: 2px solid #E06E53; border-radius: 10px; } @@ -4146,7 +4146,7 @@

Bio

margin-top: 20px; margin-bottom: 20px; } - + .footer-container { width: 100%; max-width: 100%; @@ -4166,15 +4166,15 @@

Bio

width: 20em; margin: auto; } - + .copyright-column-left { width: 50%; } - + .copyright-column-right { width: 50%; } - + .copyright-row { height: 100px; margin-bottom: -3.75vw; @@ -4203,12 +4203,12 @@

Bio

- +
-

The ALA is made possible by contributions from its partners, is supported - by - NCRIS, is hosted by +

The ALA is made possible by contributions from its partners, is supported + by + NCRIS, is hosted by CSIRO, and is the Australian node of GBIF.

@@ -4221,10 +4221,10 @@

Bio

diff --git a/_site/people/people.json b/_site/people/people.json index 89521109..a0ebd86d 100644 --- a/_site/people/people.json +++ b/_site/people/people.json @@ -4,11 +4,11 @@ "title": "Matilda Stevenson", "description": {}, "author": [], - "date": "2022-04-20", + "date": "2022-04-26", "categories": [], - "contents": "\n\nPosts\n\nHex maps for species occurrence dataHex maps are a neat way to represent spatial information. Here, we show how to draw one using the most common species in the ALA database\n\n\n\n", + "contents": "\r\n\r\nPosts\r\n\r\nHex\r\nmaps for species occurrence dataHex maps are a neat way to\r\nrepresent spatial information. Here, we show how to draw one using the\r\nmost common species in the ALA database\r\n\r\n\r\n\r\n", "preview": {}, - "last_modified": "2022-04-20T15:05:33+10:00", + "last_modified": "2022-04-26T13:05:39+10:00", "input_file": {} }, { @@ -16,11 +16,11 @@ "title": "Martin Westgate", "description": {}, "author": [], - "date": "2022-04-20", + "date": "2022-04-26", "categories": [], - "contents": "\nBio\n\n\n\n\nMartin leads the Science & Decision Support Team. He holds a doctorate in landscape ecology and conservation biology from the Australian National University. His work focuses on conceptual, computational and statistical tools to better understand patterns in nature.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nPosts\n\nSunburst plots for taxonomic dataSince version 1.3.1 of galah, it has been possible to download taxonomic data using a ‘tree’ format from the data.tree package. Here I’ll demonstrate some ideas for plotting these trees using circular diagrams.Creating a color palette from an imageThere are hundreds of color palettes in the R ecosystem, but sometimes we might want to use colors from a specific image. Here I show how to use the paletter package to create a color palette for the 2020 Eucalypt of the Year\n\n\n\n", - "preview": "people/Westgate_Martin/distill-preview.png", - "last_modified": "2022-04-20T15:05:32+10:00", + "contents": "\r\nBio\r\n\r\n\r\n\r\n\r\nMartin leads the Science & Decision Support Team. He holds a\r\ndoctorate in landscape ecology and conservation biology from the\r\nAustralian National University. His work focuses on conceptual,\r\ncomputational and statistical tools to better understand patterns in\r\nnature.\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nPosts\r\n\r\nSunburst\r\nplots for taxonomic dataSince version 1.3.1 of\r\ngalah, it has been possible to download taxonomic data\r\nusing a ‘tree’ format from the data.tree package. Here I’ll\r\ndemonstrate some ideas for plotting these trees using circular\r\ndiagrams.Creating\r\na color palette from an imageThere are hundreds of color\r\npalettes in the R ecosystem, but sometimes we might want to use colors\r\nfrom a specific image. Here I show how to use the paletter\r\npackage to create a color palette for the 2020 Eucalypt of the\r\nYear\r\n\r\n\r\n\r\n", + "preview": {}, + "last_modified": "2022-04-26T13:05:36+10:00", "input_file": {} }, { @@ -30,9 +30,9 @@ "author": [], "date": "2022-03-21", "categories": [], - "contents": "\nBio\n\n\n\n\nShandiya is an evolutionary ecologist with interests in conservation\ngenetics and wildlife disease. Her work as a data analyst at the ALA\nfocuses on streamlining reproducible pathways for researchers to access\nand analyse open data.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nPosts\n\n\n\n\n", - "preview": "people/Balasubramaniam_Shandiya/distill-preview.png", - "last_modified": "2022-03-23T13:43:38+11:00", + "contents": "\r\nBio\r\n\r\n\r\n\r\n\r\nShandiya is an evolutionary ecologist with interests in conservation\r\ngenetics and wildlife disease. Her work as a data analyst at the ALA\r\nfocuses on streamlining reproducible pathways for researchers to access\r\nand analyse open data.\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nPosts\r\n\r\n\r\n\r\n\r\n", + "preview": {}, + "last_modified": "2022-03-21T15:17:36+11:00", "input_file": {} }, { @@ -42,9 +42,9 @@ "author": [], "date": "2022-03-21", "categories": [], - "contents": "\nBio\n\n\n\n\nJenna is a conservation scientist with a PhD in Plant Ecology. She is\ncurrently the Training and Outreach Coordinator for the ALA and uses her\nskills in science communication to develop and deliver exciting training\nmaterials and workshops. Jenna is a strong advocate for diversity and\ninclusion in STEMM fields.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", - "preview": "people/Wraith_Jenna/distill-preview.png", - "last_modified": "2022-03-23T13:43:38+11:00", + "contents": "\r\nBio\r\n\r\n\r\n\r\n\r\nJenna is a conservation scientist with a PhD in Plant Ecology. She is\r\ncurrently the Training and Outreach Coordinator for the ALA and uses her\r\nskills in science communication to develop and deliver exciting training\r\nmaterials and workshops. Jenna is a strong advocate for diversity and\r\ninclusion in STEMM fields.\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n", + "preview": {}, + "last_modified": "2022-03-21T15:19:02+11:00", "input_file": {} }, { @@ -54,9 +54,9 @@ "author": [], "date": "2022-03-10", "categories": [], - "contents": "\nBio\n\n\n\n\nDax is an evolutionary biologist, with a PhD in biological sciences and\nsocial psychology. As a data analyst at the ALA, he tries to make data\nin the ALA accessible for scientists and citizen scientists to use and\nfacilitates scientific rigour and transparency.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nPosts\n\n\n\n\n", - "preview": "people/Kellie_Dax/distill-preview.png", - "last_modified": "2022-03-23T13:43:38+11:00", + "contents": "\r\nBio\r\n\r\n\r\n\r\n\r\nDax is an evolutionary biologist, with a PhD in biological sciences and\r\nsocial psychology. As a data analyst at the ALA, he tries to make data\r\nin the ALA accessible for scientists and citizen scientists to use and\r\nfacilitates scientific rigour and transparency.\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nPosts\r\n\r\n\r\n\r\n\r\n", + "preview": {}, + "last_modified": "2022-03-10T15:36:48+11:00", "input_file": {} }, { @@ -66,9 +66,9 @@ "author": [], "date": "2022-02-15", "categories": [], - "contents": "\nBio\n\n\n\n\nAmanda is a Data Analyst for EcoCommons, and is based at the Atlas of Living Australia. She completed her PhD in Structural Biology, and uses her extensive Python skills to manage the integration of data within EcoCommons. She also works to optimize the available scientific workflows within the platform.\n\n\n\n\n\n\n\n\n\n\n\n\n\nPosts\n\n\n\n\n", + "contents": "\r\nBio\r\n\r\n\r\n\r\n\r\nAmanda is a Data Analyst for EcoCommons, and is based at the Atlas of Living Australia. She completed her PhD in Structural Biology, and uses her extensive Python skills to manage the integration of data within EcoCommons. She also works to optimize the available scientific workflows within the platform.\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nPosts\r\n\r\n\r\n\r\n\r\n", "preview": "people/Buyan_Amanda/distill-preview.png", - "last_modified": "2022-02-16T12:00:13+11:00", + "last_modified": "2022-02-15T14:25:39+11:00", "input_file": {} }, { @@ -78,9 +78,9 @@ "author": [], "date": "2022-02-15", "categories": [], - "contents": "\nBio\n\n\n\n\nJess is an evolutionary scientist based at Atlas of Living Australia, with a postdoc on species delimitation methods and PhD on diversification at micro/macro scales. She is also affiliated with the Division of Ecology and Evolution, ANU. Jess brings her experience in R to improve the readability, data standardisation and performance of EcoCommons’ platform ecological modelling algorithms. Jess is a strong advocate for diversity in STEMM.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", + "contents": "\r\nBio\r\n\r\n\r\n\r\n\r\nJess is an evolutionary scientist based at Atlas of Living Australia, with a postdoc on species delimitation methods and PhD on diversification at micro/macro scales. She is also affiliated with the Division of Ecology and Evolution, ANU. Jess brings her experience in R to improve the readability, data standardisation and performance of EcoCommons’ platform ecological modelling algorithms. Jess is a strong advocate for diversity in STEMM.\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n", "preview": "people/Fenker_Jess/distill-preview.png", - "last_modified": "2022-02-16T12:00:13+11:00", + "last_modified": "2022-02-15T14:26:17+11:00", "input_file": {} }, { @@ -90,9 +90,9 @@ "author": [], "date": "2022-02-15", "categories": [], - "contents": "\nBio\n\n\n\n\nOlivia is a graduate from the Australian National University (ANU), acquiring an interdisciplinary double degree in philosophy and natural resource management. At the ALA, she works as a support officer with a primary focus on monitoring the research impact of the Atlas. She is passionate about accessibility, environmental justice and diversity in STEMM.\n\n\n\n\n\n\n\n\n\n\n\n\n\nPosts\n\n\n\n\n", + "contents": "\r\nBio\r\n\r\n\r\n\r\n\r\nOlivia is a graduate from the Australian National University (ANU), acquiring an interdisciplinary double degree in philosophy and natural resource management. At the ALA, she works as a support officer with a primary focus on monitoring the research impact of the Atlas. She is passionate about accessibility, environmental justice and diversity in STEMM.\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nPosts\r\n\r\n\r\n\r\n\r\n", "preview": "people/Torresan_Olivia/distill-preview.png", - "last_modified": "2022-02-16T12:00:13+11:00", + "last_modified": "2022-02-15T14:22:23+11:00", "input_file": {} } ] diff --git a/_site/posts.html b/_site/posts.html index b2111fc0..cf578778 100644 --- a/_site/posts.html +++ b/_site/posts.html @@ -2811,7 +2811,7 @@

${suggestion.title}

color: var(--hover-color, white); } - + @@ -3129,7 +3129,8 @@

Posts

-
+
@@ -3153,13 +3154,13 @@

Posts

.copyright-container { display: flex; width: 100%; - margin-bottom: 30px; + margin-bottom: 30px; margin-top: 30px; } .copyright-row { width: 100%; - height: 60px; + height: 66px; position: relative; margin-bottom: -1.75vw; background-color: #c44d34; @@ -3217,14 +3218,14 @@

Posts

/*-- Logos and acknowledgement --*/ -.footer-container { +.footer-container { display: flex; width: 100%; max-width: 1140px; padding-left: 15px; padding-right: 15px; padding-top: 50px; - padding-bottom: 50px; + padding-bottom: 50px; margin: auto; } @@ -3307,7 +3308,7 @@

Posts

margin: 4px 2px; transition-duration: 0.3s; cursor: pointer; - background-color: white; + background-color: white; border: 2px solid #E06E53; border-radius: 10px; } @@ -3330,7 +3331,7 @@

Posts

margin-top: 20px; margin-bottom: 20px; } - + .footer-container { width: 100%; max-width: 100%; @@ -3350,15 +3351,15 @@

Posts

width: 20em; margin: auto; } - + .copyright-column-left { width: 50%; } - + .copyright-column-right { width: 50%; } - + .copyright-row { height: 100px; margin-bottom: -3.75vw; @@ -3387,32 +3388,32 @@

Posts

- +
-

The ALA is made possible by contributions from its partners, is supported - by - NCRIS, is hosted by +

The ALA is made possible by contributions from its partners, is supported + by + NCRIS, is hosted by CSIRO, and is the Australian node of GBIF.

- NCRIS logo - CSIRO logo - GBIF logo @@ -3420,10 +3421,10 @@

Posts

diff --git a/_site/posts/2021-03-20_creating-a-color-palette-from-an-image/index.html b/_site/posts/2021-03-20_creating-a-color-palette-from-an-image/index.html index 978b2336..529285a0 100644 --- a/_site/posts/2021-03-20_creating-a-color-palette-from-an-image/index.html +++ b/_site/posts/2021-03-20_creating-a-color-palette-from-an-image/index.html @@ -3104,13 +3104,13 @@

Creating a plot

.copyright-container { display: flex; width: 100%; - margin-bottom: 30px; + margin-bottom: 30px; margin-top: 30px; } .copyright-row { width: 100%; - height: 60px; + height: 66px; position: relative; margin-bottom: -1.75vw; background-color: #c44d34; @@ -3168,14 +3168,14 @@

Creating a plot

/*-- Logos and acknowledgement --*/ -.footer-container { +.footer-container { display: flex; width: 100%; max-width: 1140px; padding-left: 15px; padding-right: 15px; padding-top: 50px; - padding-bottom: 50px; + padding-bottom: 50px; margin: auto; } @@ -3258,7 +3258,7 @@

Creating a plot

margin: 4px 2px; transition-duration: 0.3s; cursor: pointer; - background-color: white; + background-color: white; border: 2px solid #E06E53; border-radius: 10px; } @@ -3281,7 +3281,7 @@

Creating a plot

margin-top: 20px; margin-bottom: 20px; } - + .footer-container { width: 100%; max-width: 100%; @@ -3301,15 +3301,15 @@

Creating a plot

width: 20em; margin: auto; } - + .copyright-column-left { width: 50%; } - + .copyright-column-right { width: 50%; } - + .copyright-row { height: 100px; margin-bottom: -3.75vw; @@ -3338,12 +3338,12 @@

Creating a plot

- +
-

The ALA is made possible by contributions from its partners, is supported - by - NCRIS, is hosted by +

The ALA is made possible by contributions from its partners, is supported + by + NCRIS, is hosted by CSIRO, and is the Australian node of GBIF.

@@ -3356,10 +3356,10 @@

Creating a plot

diff --git a/_site/posts/2021-04-14_hex-maps-for-species-occurrence-data/index.html b/_site/posts/2021-04-14_hex-maps-for-species-occurrence-data/index.html index 998bff4b..aa3f9524 100644 --- a/_site/posts/2021-04-14_hex-maps-for-species-occurrence-data/index.html +++ b/_site/posts/2021-04-14_hex-maps-for-species-occurrence-data/index.html @@ -3142,13 +3142,13 @@

Plotting binned data

.copyright-container { display: flex; width: 100%; - margin-bottom: 30px; + margin-bottom: 30px; margin-top: 30px; } .copyright-row { width: 100%; - height: 60px; + height: 66px; position: relative; margin-bottom: -1.75vw; background-color: #c44d34; @@ -3206,14 +3206,14 @@

Plotting binned data

/*-- Logos and acknowledgement --*/ -.footer-container { +.footer-container { display: flex; width: 100%; max-width: 1140px; padding-left: 15px; padding-right: 15px; padding-top: 50px; - padding-bottom: 50px; + padding-bottom: 50px; margin: auto; } @@ -3296,7 +3296,7 @@

Plotting binned data

margin: 4px 2px; transition-duration: 0.3s; cursor: pointer; - background-color: white; + background-color: white; border: 2px solid #E06E53; border-radius: 10px; } @@ -3319,7 +3319,7 @@

Plotting binned data

margin-top: 20px; margin-bottom: 20px; } - + .footer-container { width: 100%; max-width: 100%; @@ -3339,15 +3339,15 @@

Plotting binned data

width: 20em; margin: auto; } - + .copyright-column-left { width: 50%; } - + .copyright-column-right { width: 50%; } - + .copyright-row { height: 100px; margin-bottom: -3.75vw; @@ -3376,12 +3376,12 @@

Plotting binned data

- +
-

The ALA is made possible by contributions from its partners, is supported - by - NCRIS, is hosted by +

The ALA is made possible by contributions from its partners, is supported + by + NCRIS, is hosted by CSIRO, and is the Australian node of GBIF.

@@ -3394,10 +3394,10 @@

Plotting binned data

diff --git a/_site/posts/2022-02-17_sunburst-plots-for-taxonomic-data/index.html b/_site/posts/2022-02-17_sunburst-plots-for-taxonomic-data/index.html index e860ba80..a487cdbc 100644 --- a/_site/posts/2022-02-17_sunburst-plots-for-taxonomic-data/index.html +++ b/_site/posts/2022-02-17_sunburst-plots-for-taxonomic-data/index.html @@ -3245,13 +3245,13 @@

Drawing

.copyright-container { display: flex; width: 100%; - margin-bottom: 30px; + margin-bottom: 30px; margin-top: 30px; } .copyright-row { width: 100%; - height: 60px; + height: 66px; position: relative; margin-bottom: -1.75vw; background-color: #c44d34; @@ -3309,14 +3309,14 @@

Drawing

/*-- Logos and acknowledgement --*/ -.footer-container { +.footer-container { display: flex; width: 100%; max-width: 1140px; padding-left: 15px; padding-right: 15px; padding-top: 50px; - padding-bottom: 50px; + padding-bottom: 50px; margin: auto; } @@ -3399,7 +3399,7 @@

Drawing

margin: 4px 2px; transition-duration: 0.3s; cursor: pointer; - background-color: white; + background-color: white; border: 2px solid #E06E53; border-radius: 10px; } @@ -3422,7 +3422,7 @@

Drawing

margin-top: 20px; margin-bottom: 20px; } - + .footer-container { width: 100%; max-width: 100%; @@ -3442,15 +3442,15 @@

Drawing

width: 20em; margin: auto; } - + .copyright-column-left { width: 50%; } - + .copyright-column-right { width: 50%; } - + .copyright-row { height: 100px; margin-bottom: -3.75vw; @@ -3479,12 +3479,12 @@

Drawing

- +
-

The ALA is made possible by contributions from its partners, is supported - by - NCRIS, is hosted by +

The ALA is made possible by contributions from its partners, is supported + by + NCRIS, is hosted by CSIRO, and is the Australian node of GBIF.

@@ -3497,10 +3497,10 @@

Drawing

diff --git a/_site/posts/posts.json b/_site/posts/posts.json index da4e6244..fdc2166b 100644 --- a/_site/posts/posts.json +++ b/_site/posts/posts.json @@ -17,9 +17,9 @@ "Chordata", "Aves" ], - "contents": "\n\nContents\nGetting species occurrences\nPlotting binned data\n\n\n\n\n/* Remove distill metadata section */\nd-byline {\n display: none;\n }\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nAuthor:\n\nMatilda Stevenson, Martin Westgate, Dax Kellie\n\n\nDate:\n\n25 February, 2022\n\n\n\n\n\nThe Atlas of Living Australia (ALA) holds records of magpie sightings from a number data providers like iNaturalist, eBird and BirdLife Australia. Let’s make a visualisation of Australian Bird of the Year 2018 winner, Magpies, using records held in the ALA.\nGetting species occurrences\nAs with any R project, a good first step is to load the required packages\n\n\n# packages\nlibrary(ggplot2)\nlibrary(tidyr)\nlibrary(dplyr)\nlibrary(ozmaps)\nlibrary(sf)\nlibrary(hexbin)\n\n\n\nWe will use the {galah} package to download records.\nTo download species occurrence records, the {galah} package requires you to add an email registered with the ALA to galah_config(). If running this code yourself, you will need to add an email using the code below, substituting your email with myemail@email.com. This email address should be registered with the ALA, which you can do here\n\n\nlibrary(galah)\ngalah_config(email = \"myemail@email.com\")\n\n\n\nNow we can download magpie occurrence records by using atlas_occurrences(). Note that we also set our data ‘profile’ to ‘ALA’; this means we only download records that meet some basic data quality standards enforced by the atlas. This is optional, but tends to improve the quality of the data returned. (If you wish to see the data quality filters applied in the ALA profile, use search_profile_attributes(\"ALA\"))\n\n\nmagpie_occ <- galah_call() %>%\n galah_identify(\"Cracticus tibicen\") %>%\n galah_filter(profile = \"ALA\") %>%\n atlas_occurrences()\n\n\n\nLet’s have a look at the first few rows of the data we’ve just downloaded:\n\n\nmagpie_occ %>% head()\n\n\n# A tibble: 6 x 7\n decimalLatitude decimalLongitude eventDate scientificName\n \n1 -45.0 169. \"\" Gymnorhina ti~\n2 -44.5 170. \"2018-10-27T21:44:0~ Gymnorhina ti~\n3 -44.1 170. \"2019-05-02T05:18:0~ Gymnorhina ti~\n4 -43.6 147. \"\" Gymnorhina ti~\n5 -43.6 147. \"\" Gymnorhina ti~\n6 -43.6 147. \"\" Gymnorhina ti~\n# ... with 3 more variables: taxonConceptID , recordID ,\n# dataResourceName \n\nFor the purpose of this exercise, we’re going to filter records not on the mainland or Tasmania.\n\n\nfiltered_occ <- magpie_occ %>% filter(decimalLongitude < 155,\n decimalLongitude > 110,\n decimalLatitude > -45,\n decimalLatitude < -10)\n\n\n\nPlotting binned data\nThe easiest way to create a hex map is using the hexbin package\n\n\nggplot(filtered_occ) +\n geom_hex(mapping = aes(x = decimalLongitude, y = decimalLatitude), \n bins = 47, \n colour = \"white\") +\n ylim(-45, -10) +\n xlim(110, 155) +\n scale_fill_gradientn(colours = c(\"#EEECEA\", \"#E06E53\")) +\n coord_fixed() +\n theme_bw() +\n theme(\n panel.border = element_blank(),\n panel.grid.major = element_blank(),\n panel.grid.minor = element_blank(),\n axis.text = element_blank(),\n axis.ticks = element_blank(),\n axis.title = element_blank(),\n legend.position = \"none\")\n\n\n\n\nWhile automated methods are nice, it might sometimes be useful to create our hexagons manually. We can do this by manually creating hexagons, linking them to a grid of Australia, and adding our data of magpie counts to set the fill color of those hexagons. To achieve this, we first convert the map of Australia provided by ozmaps to the same coordinate system as ALA data:\n\n\naus <- st_transform(ozmaps::ozmap_country, 4326)\n\n\n\nWe then transform points from our data frame of magpie occurrences to the correct coordinates\n\n\npoints <- lapply(1:nrow(filtered_occ), function(x) {\n st_point(c(as.numeric(filtered_occ[x,'decimalLongitude']),\n as.numeric(filtered_occ[x,'decimalLatitude'])))\n})\n\n\n\nNow build the hexagonal grid\n\n\ngrid_all <- st_make_grid(aus, cellsize = 1, what = \"polygons\", square = FALSE)[aus]\ngridSF <- st_as_sf(grid_all)\n\n# Find which polygon each point is in\nintersect <- st_intersects(st_as_sfc(points) %>% st_set_crs(4326), gridSF)\n\n# count the number of points in each hexagon\nfreqs <- as.data.frame(table(unlist(intersect)), stringsAsFactors = FALSE)\n\nnames(freqs) <- c(\"hex_id\", \"count\")\n\n\n\nAdd our magpie counts data to the hexagons and fill hexagons without data\n\n\nfreqs$hex_id <- as.integer(freqs$hex_id)\n\n# fill in the values for hexagons with no records\nfilled_freqs <- freqs %>%\n complete(hex_id = full_seq(c(1,925), 1), fill = list(count = 0))\ngridSF$n <- filled_freqs$count\n\n\n\nFinally, build the plot\n\n\nggplot() +\n geom_sf(data = gridSF, aes(fill = n), size = .01) +\n scale_fill_gradientn(colours = c(\"#EEECEA\", \"#E06E53\"), na.value = \"white\", trans = \"log2\") +\n ylim(-45, -10) +\n xlim(110, 155) +\n theme_bw() +\n theme(\n panel.border = element_blank(),\n panel.grid.major = element_blank(),\n panel.grid.minor = element_blank(),\n axis.text = element_blank(),\n axis.ticks = element_blank(),\n axis.title = element_blank(),\n legend.position = \"none\") \n\n\n\n\nThat’s it! All the extra work does make a difference in this case, providing a better representation of the spread of Mapgies across Australia. Manually constructing hex maps can be useful in other circumstances, too. For example, if we wanted to compare the number of magpies to contextual information within each polygon (such as rainfall or human population data), then manually constructing our own hexagons could help us to combine data from different sources.\nA final point is that we could have achieved the same result by creating polygons first, then querying the ALA for the number of magpie records in each polygon using galah_geolocate(). That’s a bit more challenging, and not worthwhile in this case; but it can be an efficient solution where you require information on more species than there are polygons, for example. We’ll come back to this point in a future blog post.\n\n\n\n", + "contents": "\r\n\r\nContents\r\nGetting species occurrences\r\nPlotting binned data\r\n\r\n\r\n\r\n\r\n/* Remove distill metadata section */\r\nd-byline {\r\n display: none;\r\n }\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nAuthor:\r\n\r\nMatilda Stevenson, Martin Westgate, Dax Kellie\r\n\r\n\r\nDate:\r\n\r\n25 February, 2022\r\n\r\n\r\n\r\n\r\n\r\nThe Atlas of Living Australia (ALA) holds records of magpie sightings from a number data providers like iNaturalist, eBird and BirdLife Australia. Let’s make a visualisation of Australian Bird of the Year 2018 winner, Magpies, using records held in the ALA.\r\nGetting species occurrences\r\nAs with any R project, a good first step is to load the required packages\r\n\r\n\r\n# packages\r\nlibrary(ggplot2)\r\nlibrary(tidyr)\r\nlibrary(dplyr)\r\nlibrary(ozmaps)\r\nlibrary(sf)\r\nlibrary(hexbin)\r\n\r\n\r\n\r\nWe will use the {galah} package to download records.\r\nTo download species occurrence records, the {galah} package requires you to add an email registered with the ALA to galah_config(). If running this code yourself, you will need to add an email using the code below, substituting your email with myemail@email.com. This email address should be registered with the ALA, which you can do here\r\n\r\n\r\nlibrary(galah)\r\ngalah_config(email = \"myemail@email.com\")\r\n\r\n\r\n\r\nNow we can download magpie occurrence records by using atlas_occurrences(). Note that we also set our data ‘profile’ to ‘ALA’; this means we only download records that meet some basic data quality standards enforced by the atlas. This is optional, but tends to improve the quality of the data returned. (If you wish to see the data quality filters applied in the ALA profile, use search_profile_attributes(\"ALA\"))\r\n\r\n\r\nmagpie_occ <- galah_call() %>%\r\n galah_identify(\"Cracticus tibicen\") %>%\r\n galah_filter(profile = \"ALA\") %>%\r\n atlas_occurrences()\r\n\r\n\r\n\r\nLet’s have a look at the first few rows of the data we’ve just downloaded:\r\n\r\n\r\nmagpie_occ %>% head()\r\n\r\n\r\n# A tibble: 6 x 7\r\n decimalLatitude decimalLongitude eventDate scientificName\r\n \r\n1 -45.0 169. \"\" Gymnorhina ti~\r\n2 -44.5 170. \"2018-10-27T21:44:0~ Gymnorhina ti~\r\n3 -44.1 170. \"2019-05-02T05:18:0~ Gymnorhina ti~\r\n4 -43.6 147. \"\" Gymnorhina ti~\r\n5 -43.6 147. \"\" Gymnorhina ti~\r\n6 -43.6 147. \"\" Gymnorhina ti~\r\n# ... with 3 more variables: taxonConceptID , recordID ,\r\n# dataResourceName \r\n\r\nFor the purpose of this exercise, we’re going to filter records not on the mainland or Tasmania.\r\n\r\n\r\nfiltered_occ <- magpie_occ %>% filter(decimalLongitude < 155,\r\n decimalLongitude > 110,\r\n decimalLatitude > -45,\r\n decimalLatitude < -10)\r\n\r\n\r\n\r\nPlotting binned data\r\nThe easiest way to create a hex map is using the hexbin package\r\n\r\n\r\nggplot(filtered_occ) +\r\n geom_hex(mapping = aes(x = decimalLongitude, y = decimalLatitude), \r\n bins = 47, \r\n colour = \"white\") +\r\n ylim(-45, -10) +\r\n xlim(110, 155) +\r\n scale_fill_gradientn(colours = c(\"#EEECEA\", \"#E06E53\")) +\r\n coord_fixed() +\r\n theme_bw() +\r\n theme(\r\n panel.border = element_blank(),\r\n panel.grid.major = element_blank(),\r\n panel.grid.minor = element_blank(),\r\n axis.text = element_blank(),\r\n axis.ticks = element_blank(),\r\n axis.title = element_blank(),\r\n legend.position = \"none\")\r\n\r\n\r\n\r\n\r\nWhile automated methods are nice, it might sometimes be useful to create our hexagons manually. We can do this by manually creating hexagons, linking them to a grid of Australia, and adding our data of magpie counts to set the fill color of those hexagons. To achieve this, we first convert the map of Australia provided by ozmaps to the same coordinate system as ALA data:\r\n\r\n\r\naus <- st_transform(ozmaps::ozmap_country, 4326)\r\n\r\n\r\n\r\nWe then transform points from our data frame of magpie occurrences to the correct coordinates\r\n\r\n\r\npoints <- lapply(1:nrow(filtered_occ), function(x) {\r\n st_point(c(as.numeric(filtered_occ[x,'decimalLongitude']),\r\n as.numeric(filtered_occ[x,'decimalLatitude'])))\r\n})\r\n\r\n\r\n\r\nNow build the hexagonal grid\r\n\r\n\r\ngrid_all <- st_make_grid(aus, cellsize = 1, what = \"polygons\", square = FALSE)[aus]\r\ngridSF <- st_as_sf(grid_all)\r\n\r\n# Find which polygon each point is in\r\nintersect <- st_intersects(st_as_sfc(points) %>% st_set_crs(4326), gridSF)\r\n\r\n# count the number of points in each hexagon\r\nfreqs <- as.data.frame(table(unlist(intersect)), stringsAsFactors = FALSE)\r\n\r\nnames(freqs) <- c(\"hex_id\", \"count\")\r\n\r\n\r\n\r\nAdd our magpie counts data to the hexagons and fill hexagons without data\r\n\r\n\r\nfreqs$hex_id <- as.integer(freqs$hex_id)\r\n\r\n# fill in the values for hexagons with no records\r\nfilled_freqs <- freqs %>%\r\n complete(hex_id = full_seq(c(1,925), 1), fill = list(count = 0))\r\ngridSF$n <- filled_freqs$count\r\n\r\n\r\n\r\nFinally, build the plot\r\n\r\n\r\nggplot() +\r\n geom_sf(data = gridSF, aes(fill = n), size = .01) +\r\n scale_fill_gradientn(colours = c(\"#EEECEA\", \"#E06E53\"), na.value = \"white\", trans = \"log2\") +\r\n ylim(-45, -10) +\r\n xlim(110, 155) +\r\n theme_bw() +\r\n theme(\r\n panel.border = element_blank(),\r\n panel.grid.major = element_blank(),\r\n panel.grid.minor = element_blank(),\r\n axis.text = element_blank(),\r\n axis.ticks = element_blank(),\r\n axis.title = element_blank(),\r\n legend.position = \"none\") \r\n\r\n\r\n\r\n\r\nThat’s it! All the extra work does make a difference in this case, providing a better representation of the spread of Mapgies across Australia. Manually constructing hex maps can be useful in other circumstances, too. For example, if we wanted to compare the number of magpies to contextual information within each polygon (such as rainfall or human population data), then manually constructing our own hexagons could help us to combine data from different sources.\r\nA final point is that we could have achieved the same result by creating polygons first, then querying the ALA for the number of magpie records in each polygon using galah_geolocate(). That’s a bit more challenging, and not worthwhile in this case; but it can be an efficient solution where you require information on more species than there are polygons, for example. We’ll come back to this point in a future blog post.\r\n\r\n\r\n\r\n", "preview": "posts/2021-04-14_hex-maps-for-species-occurrence-data/distill-preview.png", - "last_modified": "2022-03-24T12:27:41+11:00", + "last_modified": "2022-04-21T15:32:59+10:00", "input_file": {} }, { @@ -39,9 +39,9 @@ "Animalia", "Chordata" ], - "contents": "\n\nContents\nDownloading taxonomic trees\nGetting plot-ready data\nDrawing\n\n\n\n\n\n\nd-byline {\n display: none;\n }\n\n\n\n\n\n\n\n\n\n\n\nAuthor:\n\nMartin Westgate\n\n\nDate:\n\n17 February, 2022\n\n\n\n\n\nTaxonomy is pretty important at the ALA. Every occurrence record in the atlas is linked to a unique taxonomic identifier. These identifiers are themselves drawn from expertly curated taxonomic datasets. This system of classification is so important to our infrastructure that we have a special name for it; the ‘taxonomic backbone’. But what does it look like?\nVisualising trees is not particularly easy for me; I didn’t train in it, and the data structures involved can be a bit complex. More importantly, until recently it was difficult to download detailed taxonomic information from the ALA. Since version 1.3.1 of galah, however, it has been possible to download taxonomic trees using the atlas_taxonomy() function. Let’s have a go at visualising these trees now.\nDownloading taxonomic trees\nThe first step is to choose a taxonomic group to represent in tree form. I’ve chosen the chordates (Phylum Chordata) because they aren’t too large a group and the names are fairly well-known. We can specify this within galah using the function galah_identify. The second piece of information we need to supply is how far ‘down’ the tree to travel. I’ve chosen the Order level here using galah_down_to(order); while we could have gone to the Family or even Genus, trying to traverse too many levels (i.e. to Genus or Species) would take a very long time. A full list of accepted ranks can be found by calling show_all_ranks().\n\n\nlibrary(galah)\nchordate_orders <- galah_call() |>\n galah_identify(\"chordata\") |>\n galah_down_to(order) |>\n atlas_taxonomy()\n\n\n\nThe object returned by atlas_taxonomy is slightly unusual; it uses the data.tree package, meaning that the dataset is literally structured like a tree. This is notably different from other representations of networks, such as you might find in igraph, for example. To get an idea of what the data look like, we can use the inbuilt print method for this data type:\n\n\nlibrary(data.tree)\nprint(chordate_orders, pruneMethod = \"dist\", limit = 10)\n\n\n levelName\n1 Chordata \n2 ¦--Cephalochordata \n3 ¦ °--Amphioxi \n4 ¦ °--... 1 nodes w/ 0 sub \n5 ¦--Tunicata \n6 ¦ ¦--Appendicularia \n7 ¦ ¦ °--... 1 nodes w/ 0 sub \n8 ¦ ¦--Ascidiacea \n9 ¦ ¦ °--... 5 nodes w/ 0 sub \n10 ¦ °--Thaliacea \n11 ¦ °--... 3 nodes w/ 0 sub \n12 °--Vertebrata \n13 ¦--Agnatha \n14 ¦ °--... 2 nodes w/ 2 sub \n15 °--Gnathostomata \n16 °--... 5 nodes w/ 134 sub\n\nThis shows there are three nodes directly beneath Chordata in the taxonomic hierarchy, of which the largest (by number of sub-nodes) is the vertebrates (Vertebrata). There is a lot we could do with this tree; each node contains a unique taxonomic identifer, for example, meaning that we could use individual nodes to make new queries using galah. However, for now a useful task is simply to visualise the structure of the whole tree.\nGetting plot-ready data\nTaxonomic trees are complex. While all species have a Kingdom, Phylum, Order, Class and Family, there are many intermediate categories that are ‘optional’. In practice, this means that when we convert to a data.frame for plotting, there are a lot of missing values; nodes that apply to some rows but not others.\n\n\ndf_rank <- ToDataFrameTypeCol(chordate_orders, type = \"rank\")\ndf_rank[10:20,]\n\n\n rank_phylum rank_subphylum rank_class rank_informal\n10 Chordata Tunicata Thaliacea \n11 Chordata Vertebrata Myxini, Agnatha\n12 Chordata Vertebrata Petromyzontida, Agnatha\n13 Chordata Vertebrata Amphibia Gnathostomata\n14 Chordata Vertebrata Amphibia Gnathostomata\n15 Chordata Vertebrata Amphibia Gnathostomata\n16 Chordata Vertebrata Aves Gnathostomata\n17 Chordata Vertebrata Aves Gnathostomata\n18 Chordata Vertebrata Aves Gnathostomata\n19 Chordata Vertebrata Aves Gnathostomata\n20 Chordata Vertebrata Aves Gnathostomata\n rank_subclass rank_superorder rank_subinfraclass rank_infraclass\n10 \n11 \n12 \n13 Lissamphibia \n14 Labyrinthodonta \n15 Salientia \n16 Neognathae \n17 Neognathae \n18 Palaeognathae Ratitae \n19 Palaeognathae Ratitae \n20 \n rank_subdivision zoology rank_order\n10 Salpida\n11 Myxiniformes\n12 Petromyzontiformes\n13 Anura\n14 Temnospondyli\n15 Sphenodontia\n16 Accipititrifomes\n17 Phaethontifomes\n18 Casuariifomes\n19 Dinornithiformes\n20 Anseriformes\n\nThese missing values will show up as empty sections in the resulting diagram, which isn’t ideal. Instead, we can build this data.frame so as to place all nodes in order by row, with empty ‘levels’ being placed at the end. This also avoids the problem where ‘unnamed’ ranks are grouped in the same column. To achieve this, we simply choose a different node attribute (level in this case) to supply to the type argument.\n\n\ndf_level <- ToDataFrameTypeCol(chordate_orders, type = \"level\")\ndf_level[10:20, ]\n\n\n level_1 level_2 level_3 level_4\n10 Chordata Tunicata Thaliacea Salpida\n11 Chordata Vertebrata Agnatha Myxini\n12 Chordata Vertebrata Agnatha Petromyzontida\n13 Chordata Vertebrata Gnathostomata Amphibia\n14 Chordata Vertebrata Gnathostomata Amphibia\n15 Chordata Vertebrata Gnathostomata Amphibia\n16 Chordata Vertebrata Gnathostomata Aves\n17 Chordata Vertebrata Gnathostomata Aves\n18 Chordata Vertebrata Gnathostomata Aves\n19 Chordata Vertebrata Gnathostomata Aves\n20 Chordata Vertebrata Gnathostomata Aves\n level_5 level_6 level_7 level_8\n10 \n11 Myxiniformes \n12 Petromyzontiformes \n13 Lissamphibia Anura \n14 Labyrinthodonta Temnospondyli \n15 Salientia Sphenodontia \n16 Neognathae Accipititrifomes \n17 Neognathae Phaethontifomes \n18 Palaeognathae Ratitae Casuariifomes \n19 Palaeognathae Ratitae Dinornithiformes \n20 Anseriformes \n\nAnother problem in this dataset is the existence of duplicated taxonomic names. This happens because different authorities place the same taxon in different parts of the tree, and while the ALA tries to clean up these issues, some disagreements remain. The code below assumes that each name is only present once, so we have to remove duplicates to proceed. Fortunately there is a function in package base that flags duplcated values as TRUE and unique values as FALSE. We can use this function to identify rows where order is not unique.\n\n\nlibrary(dplyr)\nkeep_rows <- !duplicated(df_rank$rank_order)\ndf_rank <- filter(df_rank, keep_rows)\ndf_level <- filter(df_level, keep_rows)\n\n\n\nThe next step is to determine how to represent this structure in a plot. At the moment we can’t do this, because the data are in ‘wide’ format. Instead, we need to reorder our data so that each node/taxon is represented once, and other plotting aesthetics can be added as additional columns. To achieve this, we first convert to ‘long’ format, preserving information like what row and column each taxonomic label was recorded in.\n\n\ndf_long <- tibble(\n row = rep(seq_len(nrow(df_level)), ncol(df_level)),\n level = rep(seq_len(ncol(df_level)), each = nrow(df_level)),\n taxon = do.call(c, df_level)) |> \n filter(!is.na(taxon)) # remove missing values\n\n\n\nThen, we can summarize this plot so that each row is a single taxon, recording some metadata about rows and columns from the original dataset\n\n\ndf_plot <- df_long |>\n group_by(taxon) |>\n summarize(\n xmin = min(row) - 1, \n xmax = max(row), \n ymin = level[1] - 1,\n ymax = level[1])\n \ndf_plot\n\n\n# A tibble: 161 x 5\n taxon xmin xmax ymin ymax\n \n 1 Acanthopterygii 66 79 6 7\n 2 Accipititrifomes 15 16 5 6\n 3 Accipitriformes 21 22 4 5\n 4 Actinopterygii 56 96 4 5\n 5 Agnatha 10 12 2 3\n 6 Albuliformes 62 63 6 7\n 7 Amphibia 12 15 3 4\n 8 Amphioxi 0 1 2 3\n 9 Amphioxiformes 0 1 3 4\n10 Anguilliformes 63 64 6 7\n# ... with 151 more rows\n\nDrawing\nOur dataset now contains all the information we need to plot the structure of our taxonomic tree. As usual, we’re going to plot this with ggplot2.\n\n\nlibrary(ggplot2)\nggplot(df_plot) +\n geom_rect(\n mapping = aes(\n xmin = xmin, xmax = xmax, ymin = ymin, ymax = ymax, \n group = taxon,\n fill = ymax),\n color = \"white\")\n\n\n\n\nWhile this is (probably) accurate, it’s not very informative. The most obvious missing element is labels; to add these, we’ll need to determine which nodes are ‘leaves’, and which are ‘branches’. We’ll also want to restrict labelling to larger branches, to avoid the text looking crowded. Finally, there is no need to label leaves with both a rectangle and text; so we’ll remove the leaf rectangles from the plot.\n\n\ndf_plot <- df_plot |> mutate(\n x_dist = xmax - xmin,\n is_leaf = taxon %in% df_rank$rank_order)\n\np <- ggplot() +\n geom_rect(\n data = filter(df_plot, !is_leaf),\n mapping = aes(\n xmin = xmin, xmax = xmax, ymin = ymin, ymax = ymax, \n group = taxon,\n fill = ymax),\n color = \"white\")\n\np +\n # branch labels\n geom_text(\n data = filter(df_plot, x_dist > 5),\n mapping = aes(\n x = xmin + (x_dist * 0.5), \n y = ymin + 0.5,\n label = taxon),\n color = \"white\",\n size = 3) +\n # leaf labels\n geom_text(\n data = filter(df_plot, is_leaf),\n aes(x = xmin + 0.5, y = ymin, label = taxon),\n angle = 90,\n hjust = 0,\n size = 2.5,\n color = \"grey20\") \n\n\n\n\nThis is better, but not ideal. A much more pleasing look is to use coord_polar() to generate a circular plot; but this leads to linear text on a circular plot, which looks messy. Fortunately, the new package geomtextpath solves this problem. All we have to do is replace geom_text with geom_textpath, leaving all other code the same, and add coord_polar() at the end.\n\n\nlibrary(geomtextpath)\n\np <- p + \n geom_textpath(\n data = filter(df_plot, x_dist > 5),\n mapping = aes(\n x = xmin + (x_dist * 0.5), \n y = ymin + 0.5,\n label = taxon),\n color = \"white\",\n size = 2.7) +\n geom_textpath(\n data = filter(df_plot, is_leaf),\n aes(x = xmin + 0.5, y = ymin, label = taxon),\n angle = 90,\n hjust = 0,\n size = 2.3,\n color = \"grey20\") +\n coord_polar()\np\n\n\n\n\nFinally, we can add some finishing touches by changing the color scheme, hiding the background colors and legend, and resizing the y axis so all the labels are visible.\n\n\nlibrary(viridis)\n\np +\n scale_fill_viridis(begin = 0, end = 0.9, direction = -1) +\n lims(y = c(0, 9)) +\n theme_void() + \n theme(legend.position = \"none\")\n\n\n\n\nDone! This is a fun plot, but there are ways it could be expanded or improved, the most obvious of which is to find ways to add supplementary information. Wouldn’t it be great, for example, to add leaf-level record counts as marginal barplots? Or scale the size of segments to the number of records, rather than the number of clades? While none of these are impossible, I’m going to leave this here for now. I hope you like the result!\n\n\n\n", + "contents": "\r\n\r\nContents\r\nDownloading taxonomic trees\r\nGetting plot-ready data\r\nDrawing\r\n\r\n\r\n\r\n\r\n\r\n\r\nd-byline {\r\n display: none;\r\n }\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nAuthor:\r\n\r\nMartin Westgate\r\n\r\n\r\nDate:\r\n\r\n17 February, 2022\r\n\r\n\r\n\r\n\r\n\r\nTaxonomy is pretty important at the ALA. Every occurrence record in the atlas is linked to a unique taxonomic identifier. These identifiers are themselves drawn from expertly curated taxonomic datasets. This system of classification is so important to our infrastructure that we have a special name for it; the ‘taxonomic backbone’. But what does it look like?\r\nVisualising trees is not particularly easy for me; I didn’t train in it, and the data structures involved can be a bit complex. More importantly, until recently it was difficult to download detailed taxonomic information from the ALA. Since version 1.3.1 of galah, however, it has been possible to download taxonomic trees using the atlas_taxonomy() function. Let’s have a go at visualising these trees now.\r\nDownloading taxonomic trees\r\nThe first step is to choose a taxonomic group to represent in tree form. I’ve chosen the chordates (Phylum Chordata) because they aren’t too large a group and the names are fairly well-known. We can specify this within galah using the function galah_identify. The second piece of information we need to supply is how far ‘down’ the tree to travel. I’ve chosen the Order level here using galah_down_to(order); while we could have gone to the Family or even Genus, trying to traverse too many levels (i.e. to Genus or Species) would take a very long time. A full list of accepted ranks can be found by calling show_all_ranks().\r\n\r\n\r\nlibrary(galah)\r\nchordate_orders <- galah_call() |>\r\n galah_identify(\"chordata\") |>\r\n galah_down_to(order) |>\r\n atlas_taxonomy()\r\n\r\n\r\n\r\nThe object returned by atlas_taxonomy is slightly unusual; it uses the data.tree package, meaning that the dataset is literally structured like a tree. This is notably different from other representations of networks, such as you might find in igraph, for example. To get an idea of what the data look like, we can use the inbuilt print method for this data type:\r\n\r\n\r\nlibrary(data.tree)\r\nprint(chordate_orders, pruneMethod = \"dist\", limit = 10)\r\n\r\n\r\n levelName\r\n1 Chordata \r\n2 ¦--Cephalochordata \r\n3 ¦ °--Amphioxi \r\n4 ¦ °--... 1 nodes w/ 0 sub \r\n5 ¦--Tunicata \r\n6 ¦ ¦--Appendicularia \r\n7 ¦ ¦ °--... 1 nodes w/ 0 sub \r\n8 ¦ ¦--Ascidiacea \r\n9 ¦ ¦ °--... 5 nodes w/ 0 sub \r\n10 ¦ °--Thaliacea \r\n11 ¦ °--... 3 nodes w/ 0 sub \r\n12 °--Vertebrata \r\n13 ¦--Agnatha \r\n14 ¦ °--... 2 nodes w/ 2 sub \r\n15 °--Gnathostomata \r\n16 °--... 5 nodes w/ 134 sub\r\n\r\nThis shows there are three nodes directly beneath Chordata in the taxonomic hierarchy, of which the largest (by number of sub-nodes) is the vertebrates (Vertebrata). There is a lot we could do with this tree; each node contains a unique taxonomic identifer, for example, meaning that we could use individual nodes to make new queries using galah. However, for now a useful task is simply to visualise the structure of the whole tree.\r\nGetting plot-ready data\r\nTaxonomic trees are complex. While all species have a Kingdom, Phylum, Order, Class and Family, there are many intermediate categories that are ‘optional’. In practice, this means that when we convert to a data.frame for plotting, there are a lot of missing values; nodes that apply to some rows but not others.\r\n\r\n\r\ndf_rank <- ToDataFrameTypeCol(chordate_orders, type = \"rank\")\r\ndf_rank[10:20,]\r\n\r\n\r\n rank_phylum rank_subphylum rank_class rank_informal\r\n10 Chordata Tunicata Thaliacea \r\n11 Chordata Vertebrata Myxini, Agnatha\r\n12 Chordata Vertebrata Petromyzontida, Agnatha\r\n13 Chordata Vertebrata Amphibia Gnathostomata\r\n14 Chordata Vertebrata Amphibia Gnathostomata\r\n15 Chordata Vertebrata Amphibia Gnathostomata\r\n16 Chordata Vertebrata Aves Gnathostomata\r\n17 Chordata Vertebrata Aves Gnathostomata\r\n18 Chordata Vertebrata Aves Gnathostomata\r\n19 Chordata Vertebrata Aves Gnathostomata\r\n20 Chordata Vertebrata Aves Gnathostomata\r\n rank_subclass rank_superorder rank_subinfraclass rank_infraclass\r\n10 \r\n11 \r\n12 \r\n13 Lissamphibia \r\n14 Labyrinthodonta \r\n15 Salientia \r\n16 Neognathae \r\n17 Neognathae \r\n18 Palaeognathae Ratitae \r\n19 Palaeognathae Ratitae \r\n20 \r\n rank_subdivision zoology rank_order\r\n10 Salpida\r\n11 Myxiniformes\r\n12 Petromyzontiformes\r\n13 Anura\r\n14 Temnospondyli\r\n15 Sphenodontia\r\n16 Accipititrifomes\r\n17 Phaethontifomes\r\n18 Casuariifomes\r\n19 Dinornithiformes\r\n20 Anseriformes\r\n\r\nThese missing values will show up as empty sections in the resulting diagram, which isn’t ideal. Instead, we can build this data.frame so as to place all nodes in order by row, with empty ‘levels’ being placed at the end. This also avoids the problem where ‘unnamed’ ranks are grouped in the same column. To achieve this, we simply choose a different node attribute (level in this case) to supply to the type argument.\r\n\r\n\r\ndf_level <- ToDataFrameTypeCol(chordate_orders, type = \"level\")\r\ndf_level[10:20, ]\r\n\r\n\r\n level_1 level_2 level_3 level_4\r\n10 Chordata Tunicata Thaliacea Salpida\r\n11 Chordata Vertebrata Agnatha Myxini\r\n12 Chordata Vertebrata Agnatha Petromyzontida\r\n13 Chordata Vertebrata Gnathostomata Amphibia\r\n14 Chordata Vertebrata Gnathostomata Amphibia\r\n15 Chordata Vertebrata Gnathostomata Amphibia\r\n16 Chordata Vertebrata Gnathostomata Aves\r\n17 Chordata Vertebrata Gnathostomata Aves\r\n18 Chordata Vertebrata Gnathostomata Aves\r\n19 Chordata Vertebrata Gnathostomata Aves\r\n20 Chordata Vertebrata Gnathostomata Aves\r\n level_5 level_6 level_7 level_8\r\n10 \r\n11 Myxiniformes \r\n12 Petromyzontiformes \r\n13 Lissamphibia Anura \r\n14 Labyrinthodonta Temnospondyli \r\n15 Salientia Sphenodontia \r\n16 Neognathae Accipititrifomes \r\n17 Neognathae Phaethontifomes \r\n18 Palaeognathae Ratitae Casuariifomes \r\n19 Palaeognathae Ratitae Dinornithiformes \r\n20 Anseriformes \r\n\r\nAnother problem in this dataset is the existence of duplicated taxonomic names. This happens because different authorities place the same taxon in different parts of the tree, and while the ALA tries to clean up these issues, some disagreements remain. The code below assumes that each name is only present once, so we have to remove duplicates to proceed. Fortunately there is a function in package base that flags duplcated values as TRUE and unique values as FALSE. We can use this function to identify rows where order is not unique.\r\n\r\n\r\nlibrary(dplyr)\r\nkeep_rows <- !duplicated(df_rank$rank_order)\r\ndf_rank <- filter(df_rank, keep_rows)\r\ndf_level <- filter(df_level, keep_rows)\r\n\r\n\r\n\r\nThe next step is to determine how to represent this structure in a plot. At the moment we can’t do this, because the data are in ‘wide’ format. Instead, we need to reorder our data so that each node/taxon is represented once, and other plotting aesthetics can be added as additional columns. To achieve this, we first convert to ‘long’ format, preserving information like what row and column each taxonomic label was recorded in.\r\n\r\n\r\ndf_long <- tibble(\r\n row = rep(seq_len(nrow(df_level)), ncol(df_level)),\r\n level = rep(seq_len(ncol(df_level)), each = nrow(df_level)),\r\n taxon = do.call(c, df_level)) |> \r\n filter(!is.na(taxon)) # remove missing values\r\n\r\n\r\n\r\nThen, we can summarize this plot so that each row is a single taxon, recording some metadata about rows and columns from the original dataset\r\n\r\n\r\ndf_plot <- df_long |>\r\n group_by(taxon) |>\r\n summarize(\r\n xmin = min(row) - 1, \r\n xmax = max(row), \r\n ymin = level[1] - 1,\r\n ymax = level[1])\r\n \r\ndf_plot\r\n\r\n\r\n# A tibble: 161 x 5\r\n taxon xmin xmax ymin ymax\r\n \r\n 1 Acanthopterygii 66 79 6 7\r\n 2 Accipititrifomes 15 16 5 6\r\n 3 Accipitriformes 21 22 4 5\r\n 4 Actinopterygii 56 96 4 5\r\n 5 Agnatha 10 12 2 3\r\n 6 Albuliformes 62 63 6 7\r\n 7 Amphibia 12 15 3 4\r\n 8 Amphioxi 0 1 2 3\r\n 9 Amphioxiformes 0 1 3 4\r\n10 Anguilliformes 63 64 6 7\r\n# ... with 151 more rows\r\n\r\nDrawing\r\nOur dataset now contains all the information we need to plot the structure of our taxonomic tree. As usual, we’re going to plot this with ggplot2.\r\n\r\n\r\nlibrary(ggplot2)\r\nggplot(df_plot) +\r\n geom_rect(\r\n mapping = aes(\r\n xmin = xmin, xmax = xmax, ymin = ymin, ymax = ymax, \r\n group = taxon,\r\n fill = ymax),\r\n color = \"white\")\r\n\r\n\r\n\r\n\r\nWhile this is (probably) accurate, it’s not very informative. The most obvious missing element is labels; to add these, we’ll need to determine which nodes are ‘leaves’, and which are ‘branches’. We’ll also want to restrict labelling to larger branches, to avoid the text looking crowded. Finally, there is no need to label leaves with both a rectangle and text; so we’ll remove the leaf rectangles from the plot.\r\n\r\n\r\ndf_plot <- df_plot |> mutate(\r\n x_dist = xmax - xmin,\r\n is_leaf = taxon %in% df_rank$rank_order)\r\n\r\np <- ggplot() +\r\n geom_rect(\r\n data = filter(df_plot, !is_leaf),\r\n mapping = aes(\r\n xmin = xmin, xmax = xmax, ymin = ymin, ymax = ymax, \r\n group = taxon,\r\n fill = ymax),\r\n color = \"white\")\r\n\r\np +\r\n # branch labels\r\n geom_text(\r\n data = filter(df_plot, x_dist > 5),\r\n mapping = aes(\r\n x = xmin + (x_dist * 0.5), \r\n y = ymin + 0.5,\r\n label = taxon),\r\n color = \"white\",\r\n size = 3) +\r\n # leaf labels\r\n geom_text(\r\n data = filter(df_plot, is_leaf),\r\n aes(x = xmin + 0.5, y = ymin, label = taxon),\r\n angle = 90,\r\n hjust = 0,\r\n size = 2.5,\r\n color = \"grey20\") \r\n\r\n\r\n\r\n\r\nThis is better, but not ideal. A much more pleasing look is to use coord_polar() to generate a circular plot; but this leads to linear text on a circular plot, which looks messy. Fortunately, the new package geomtextpath solves this problem. All we have to do is replace geom_text with geom_textpath, leaving all other code the same, and add coord_polar() at the end.\r\n\r\n\r\nlibrary(geomtextpath)\r\n\r\np <- p + \r\n geom_textpath(\r\n data = filter(df_plot, x_dist > 5),\r\n mapping = aes(\r\n x = xmin + (x_dist * 0.5), \r\n y = ymin + 0.5,\r\n label = taxon),\r\n color = \"white\",\r\n size = 2.7) +\r\n geom_textpath(\r\n data = filter(df_plot, is_leaf),\r\n aes(x = xmin + 0.5, y = ymin, label = taxon),\r\n angle = 90,\r\n hjust = 0,\r\n size = 2.3,\r\n color = \"grey20\") +\r\n coord_polar()\r\np\r\n\r\n\r\n\r\n\r\nFinally, we can add some finishing touches by changing the color scheme, hiding the background colors and legend, and resizing the y axis so all the labels are visible.\r\n\r\n\r\nlibrary(viridis)\r\n\r\np +\r\n scale_fill_viridis(begin = 0, end = 0.9, direction = -1) +\r\n lims(y = c(0, 9)) +\r\n theme_void() + \r\n theme(legend.position = \"none\")\r\n\r\n\r\n\r\n\r\nDone! This is a fun plot, but there are ways it could be expanded or improved, the most obvious of which is to find ways to add supplementary information. Wouldn’t it be great, for example, to add leaf-level record counts as marginal barplots? Or scale the size of segments to the number of records, rather than the number of clades? While none of these are impossible, I’m going to leave this here for now. I hope you like the result!\r\n\r\n\r\n\r\n", "preview": "posts/2022-02-17_sunburst-plots-for-taxonomic-data/distill-preview.png", - "last_modified": "2022-03-24T12:28:09+11:00", + "last_modified": "2022-04-21T15:32:59+10:00", "input_file": {} }, { @@ -60,9 +60,9 @@ "Eukaryota", "Plantae" ], - "contents": "\n\nContents\nColors in R\nExtracting colors\nCreating a plot\n\n\n\n\n\nd-byline {\n display: none;\n }\n\n\n\n\n\n\n\n\n\nAuthor:\n\nMartin Westgate\n\nDate:\n\nMarch 2021\n\n\n\nColors in R\nColor palettes are important to people, and the R ecosystem includes literally hundreds of possible palettes. If you want a “complete” list, go and check out Emil Hvitfeldt’s list of palettes here; but in practice there are only a few that we use routinely. Our default at ALA labs is to use viridis for continuous scales, because (to quote their CRAN page) it’s color-blind friendly, perceptually uniform, and pretty. The default purple-green-yellow color scheme is lovely, but I’m a big fan of ‘magma’, which has a black-purple-orange-yellow scheme\n\n\nlibrary(galah)\nlibrary(ggplot2)\nlibrary(viridis)\n\nrecords <- ala_counts(group_by = \"cl22\")\nrecords$State <- factor(seq_len(nrow(records)), labels = records$cl22) \n\nggplot(records, aes(x = State, y = log10(count), fill = count)) + \n geom_bar(stat = \"identity\") +\n coord_flip() +\n scale_fill_viridis(option = \"magma\", begin = 0.10, end = 0.95) +\n theme_bw() +\n theme(legend.position = \"none\")\n\n\n\n\nMy default for categorical color schemes is the ‘dark2’ palette from RColorBrewer; but given the subject matter of our work, it’s worth mentioning the wonderful feather package by Shandiya Balasubramaniam, which gives colors based on Australian bird plumage\n\n\n# remotes::install_github(repo = \"shandiya/feathers\")\nlibrary(feathers)\n\nrcfd <- ala_counts(\n taxa = select_taxa(\"Rose-crowned Fruit-Dove\"),\n group_by = \"cl22\")\nrcfd$State <- factor(seq_len(nrow(rcfd)), labels = rcfd$cl22) \n\nggplot(rcfd, aes(x = State, y = log10(count), fill = State)) + \n geom_bar(stat = \"identity\") +\n coord_flip() +\n scale_fill_manual(values = get_pal(\"rose_crowned_fruit_dove\")) +\n theme_bw() +\n theme(legend.position = \"none\")\n\n\n\n\nAll of this is fine, but what if you have a specific image that you want to take colors from? A logical choice is to pick the colors you want using an image editting program, but if we want to try something automated, there are options in R as well.\nExtracting colors\nNational Eucalypt Day aims to raise awareness about Eucalypts and celebrate their influence on the lives of Australians. In honour of National Eucalypt day, we wanted to created a plot based on occurrences data held in the Atlas of Living Australia, themed using colours from actual Eucalypts.\nWe used this image from a tweet by Dean Nicolle:\n\nHappy 'National Eucalypt Day'!The Western Australian gimlet (Eucalyptus salubris) has just been announced as Eucalypt of the Year for 2021. Renowned for its fluted, smooth, shiny, and colourful trunk & branches. pic.twitter.com/pOsufQtxWS— Dean Nicolle (@DeanNicolle1) March 22, 2021\n\n\nImage of Eucalyptus salubris by Dean NicolleFirst, get observations of the Eucalypt of the Year 2021 from ALA using the galah package. Specifically, we use ala_counts() to determine how many records of Eucalyptus salubris are held by the ALA:\n\n\nn_records <- ala_counts(taxa = select_taxa(\"Eucalyptus salubris\"))\n\n\n\nThen get a color scheme from images of the species in question using the paletter package (which needs to be installed from GitHub) \n\n\n# remotes::install_github(\"AndreaCirilloAC/paletter\")\nlibrary(paletter)\n\nimage_pal <- create_palette(\n image_path = \"./data/Dean_Nicolle_Esalubris_image_small.jpeg\",\n type_of_variable = \"categorical\",\n number_of_colors = 15)\n\n\n\n\n\n\nNote that we downsized the image before running the paletter code, as large images take much longer to process.\nCreating a plot\nOnce we have this palette, the obvious question is what kind of plot to draw. We could have done a map, but those can be a bit boring. We decided to try something that represented the number of observations we had of this species at ALA, and included color, but was otherwise just a pretty picture that didn’t need to contain any further information. Rather than have a traditional x and y axis, therefore, we decided to try out the igraph package to plot the points in an interesting way.\nFirst, we create a vector containing as many points as we want to display, and distribute our colors among them as evenly as possible\n\n\n# create a vector to index colours\ncolour_index <- rep(seq_along(image_pal),\n each = floor(n_records / length(image_pal)))\n# correct rounding errors\ncolour_index <- c(colour_index,\n rep(length(image_pal), n_records - length(colour_index)))\n\n\n\nThen we can create a network using igraph, and use it to create a layout for our points\n\n\nlibrary(igraph)\n\ngraph_list <- lapply(c(1:15), function(a){\n lookup <- which(colour_index == a)\n return(data.frame(\n from = lookup[c(1:(length(lookup)-1))],\n to = lookup[c(2:length(lookup))]))\n })\ngraph_df <- as.matrix(do.call(rbind, graph_list)) # build matrix\ncolour_graph <- graph_from_edgelist(graph_df) # create network graph\n\n# convert to a set of point locations\ntest_layout <- as.data.frame(layout_nicely(colour_graph)) # convert to df\ncolnames(test_layout) <- c(\"x\", \"y\") # change colnames\ntest_layout$colour_index <- factor(colour_index) # add colour_index col\n\n\n\nFinally, we draw the plot with ggplot2, removing axes with theme_void()\n\n\nggplot(test_layout, aes(x = x, y = y, colour = colour_index)) +\n geom_point(size = 3, alpha = 0.9) +\n scale_color_manual(values = image_pal) +\n coord_fixed() +\n theme_void() +\n theme(legend.position = \"none\")\n\n\n\n\n\n\n\nThat’s it! While I like the effect here, I think the paletter package is best suited to cases where there are large areas of strongly contrasting colors; it’s less ideal for images with subtle color differences. It also doesn’t appear to have been updated lately, which may mean it’s not being supported any more. But I’m happy with this plot, and would definitely consider using it again.\n\n\n\n", + "contents": "\r\n\r\nContents\r\nColors in R\r\nExtracting colors\r\nCreating a plot\r\n\r\n\r\n\r\n\r\n\r\nd-byline {\r\n display: none;\r\n }\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nAuthor:\r\n\r\nMartin Westgate\r\n\r\nDate:\r\n\r\nMarch 2021\r\n\r\n\r\n\r\nColors in R\r\nColor palettes are important to people, and the R ecosystem includes literally hundreds of possible palettes. If you want a “complete” list, go and check out Emil Hvitfeldt’s list of palettes here; but in practice there are only a few that we use routinely. Our default at ALA labs is to use viridis for continuous scales, because (to quote their CRAN page) it’s color-blind friendly, perceptually uniform, and pretty. The default purple-green-yellow color scheme is lovely, but I’m a big fan of ‘magma’, which has a black-purple-orange-yellow scheme\r\n\r\n\r\nlibrary(galah)\r\nlibrary(ggplot2)\r\nlibrary(viridis)\r\n\r\nrecords <- ala_counts(group_by = \"cl22\")\r\nrecords$State <- factor(seq_len(nrow(records)), labels = records$cl22) \r\n\r\nggplot(records, aes(x = State, y = log10(count), fill = count)) + \r\n geom_bar(stat = \"identity\") +\r\n coord_flip() +\r\n scale_fill_viridis(option = \"magma\", begin = 0.10, end = 0.95) +\r\n theme_bw() +\r\n theme(legend.position = \"none\")\r\n\r\n\r\n\r\n\r\nMy default for categorical color schemes is the ‘dark2’ palette from RColorBrewer; but given the subject matter of our work, it’s worth mentioning the wonderful feather package by Shandiya Balasubramaniam, which gives colors based on Australian bird plumage\r\n\r\n\r\n# remotes::install_github(repo = \"shandiya/feathers\")\r\nlibrary(feathers)\r\n\r\nrcfd <- ala_counts(\r\n taxa = select_taxa(\"Rose-crowned Fruit-Dove\"),\r\n group_by = \"cl22\")\r\nrcfd$State <- factor(seq_len(nrow(rcfd)), labels = rcfd$cl22) \r\n\r\nggplot(rcfd, aes(x = State, y = log10(count), fill = State)) + \r\n geom_bar(stat = \"identity\") +\r\n coord_flip() +\r\n scale_fill_manual(values = get_pal(\"rose_crowned_fruit_dove\")) +\r\n theme_bw() +\r\n theme(legend.position = \"none\")\r\n\r\n\r\n\r\n\r\nAll of this is fine, but what if you have a specific image that you want to take colors from? A logical choice is to pick the colors you want using an image editting program, but if we want to try something automated, there are options in R as well.\r\nExtracting colors\r\nNational Eucalypt Day aims to raise awareness about Eucalypts and celebrate their influence on the lives of Australians. In honour of National Eucalypt day, we wanted to created a plot based on occurrences data held in the Atlas of Living Australia, themed using colours from actual Eucalypts.\r\nWe used this image from a tweet by Dean Nicolle:\r\n\r\nHappy 'National Eucalypt Day'!The Western Australian gimlet (Eucalyptus salubris) has just been announced as Eucalypt of the Year for 2021. Renowned for its fluted, smooth, shiny, and colourful trunk & branches. pic.twitter.com/pOsufQtxWS— Dean Nicolle (@DeanNicolle1) March 22, 2021\r\n\r\n\r\nImage of Eucalyptus salubris by Dean NicolleFirst, get observations of the Eucalypt of the Year 2021 from ALA using the galah package. Specifically, we use ala_counts() to determine how many records of Eucalyptus salubris are held by the ALA:\r\n\r\n\r\nn_records <- ala_counts(taxa = select_taxa(\"Eucalyptus salubris\"))\r\n\r\n\r\n\r\nThen get a color scheme from images of the species in question using the paletter package (which needs to be installed from GitHub) \r\n\r\n\r\n# remotes::install_github(\"AndreaCirilloAC/paletter\")\r\nlibrary(paletter)\r\n\r\nimage_pal <- create_palette(\r\n image_path = \"./data/Dean_Nicolle_Esalubris_image_small.jpeg\",\r\n type_of_variable = \"categorical\",\r\n number_of_colors = 15)\r\n\r\n\r\n\r\n\r\n\r\n\r\nNote that we downsized the image before running the paletter code, as large images take much longer to process.\r\nCreating a plot\r\nOnce we have this palette, the obvious question is what kind of plot to draw. We could have done a map, but those can be a bit boring. We decided to try something that represented the number of observations we had of this species at ALA, and included color, but was otherwise just a pretty picture that didn’t need to contain any further information. Rather than have a traditional x and y axis, therefore, we decided to try out the igraph package to plot the points in an interesting way.\r\nFirst, we create a vector containing as many points as we want to display, and distribute our colors among them as evenly as possible\r\n\r\n\r\n# create a vector to index colours\r\ncolour_index <- rep(seq_along(image_pal),\r\n each = floor(n_records / length(image_pal)))\r\n# correct rounding errors\r\ncolour_index <- c(colour_index,\r\n rep(length(image_pal), n_records - length(colour_index)))\r\n\r\n\r\n\r\nThen we can create a network using igraph, and use it to create a layout for our points\r\n\r\n\r\nlibrary(igraph)\r\n\r\ngraph_list <- lapply(c(1:15), function(a){\r\n lookup <- which(colour_index == a)\r\n return(data.frame(\r\n from = lookup[c(1:(length(lookup)-1))],\r\n to = lookup[c(2:length(lookup))]))\r\n })\r\ngraph_df <- as.matrix(do.call(rbind, graph_list)) # build matrix\r\ncolour_graph <- graph_from_edgelist(graph_df) # create network graph\r\n\r\n# convert to a set of point locations\r\ntest_layout <- as.data.frame(layout_nicely(colour_graph)) # convert to df\r\ncolnames(test_layout) <- c(\"x\", \"y\") # change colnames\r\ntest_layout$colour_index <- factor(colour_index) # add colour_index col\r\n\r\n\r\n\r\nFinally, we draw the plot with ggplot2, removing axes with theme_void()\r\n\r\n\r\nggplot(test_layout, aes(x = x, y = y, colour = colour_index)) +\r\n geom_point(size = 3, alpha = 0.9) +\r\n scale_color_manual(values = image_pal) +\r\n coord_fixed() +\r\n theme_void() +\r\n theme(legend.position = \"none\")\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nThat’s it! While I like the effect here, I think the paletter package is best suited to cases where there are large areas of strongly contrasting colors; it’s less ideal for images with subtle color differences. It also doesn’t appear to have been updated lately, which may mean it’s not being supported any more. But I’m happy with this plot, and would definitely consider using it again.\r\n\r\n\r\n\r\n", "preview": "posts/2021-03-20_creating-a-color-palette-from-an-image/distill-preview.png", - "last_modified": "2022-03-24T12:27:20+11:00", + "last_modified": "2022-04-21T15:32:59+10:00", "input_file": {} } ] diff --git a/_site/search.json b/_site/search.json index 3eafbd00..bb6b7ef9 100644 --- a/_site/search.json +++ b/_site/search.json @@ -5,24 +5,24 @@ "title": "About us", "description": "The ALA Science & Decision Support Team", "author": [], - "contents": "\nWe are responsible for:\nData analysis & reporting on key projects\nR package development & maintenance\nOutreach & engagement with the research community\nMeet the team\nClick on photos to view their articles and contributions\n\n\n\n\n\n\n\n\nMartin Westgate\n\nTeam Leader\n\n\n\n\n\n\n\n\n\nJessica Fenker\n\nData Analyst\n\n\n\n\n\n\n\n\n\nJenna Wraith\n\nTraining & Outreach Coordinator\n\n\n\n\n\n\n\n\n\n\n\nDax Kellie\n\nData Analyst\n\n\n\n\n\n\n\n\n\nOlivia Torresan\n\nSupport Officer\n\n\n\n\n\n\n\n\n\nAmanda Buyan\n\nData Analyst\n\n\n\n\n\n\n\n\n\n\n\nShandiya Balasubramaniam\n\nData Analyst\n\n\n\n\n\n\n\n\n", - "last_modified": "2022-04-26T12:26:31+10:00" + "contents": "\r\nWe are responsible for:\r\nData analysis & reporting on key projects\r\nR package development & maintenance\r\nOutreach & engagement with the research community\r\nMeet the team\r\nClick on photos to view their articles and contributions\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nMartin Westgate\r\n\r\nTeam Leader\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nJessica Fenker\r\n\r\nData Analyst\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nJenna Wraith\r\n\r\nTraining & Outreach Coordinator\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nDax Kellie\r\n\r\nData Analyst\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nOlivia Torresan\r\n\r\nSupport Officer\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nAmanda Buyan\r\n\r\nData Analyst\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nShandiya Balasubramaniam\r\n\r\nData Analyst\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n", + "last_modified": "2022-04-26T13:05:51+10:00" }, { "path": "galah.html", "title": "galah", "author": [], "date": "`r Sys.Date()`", - "contents": "\n\n\nd-title {\n display: none;\n }\n\n\n\n\n\n\n\ngalah\ngalah is an R interface to biodiversity data hosted by the Atlas of Living Australia (ALA). It enables users to locate and download species occurrence records (observations, specimens, eDNA records, etc.), taxonomic information, or associated media such as images or sounds, and to restrict their queries to particular taxa or locations.\nVisit the galah R package website to learn more about how to use galah\nIf you have any questions, comments, or spot any bugs, email us or report an issue on our GitHub page\n\n\nGet started\nInstall from CRAN\n\n\ninstall.packages(\"galah\")\n\n\n\n\n\n\n", - "last_modified": "2022-04-26T12:26:31+10:00" + "contents": "\r\n\r\n\r\nd-title {\r\n display: none;\r\n }\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\ngalah\r\ngalah is an R interface to biodiversity data hosted by\r\nthe Atlas of Living Australia (ALA). It enables users to locate and\r\ndownload species occurrence records (observations, specimens, eDNA\r\nrecords, etc.), taxonomic information, or associated media such as\r\nimages or sounds, and to restrict their queries to particular taxa or\r\nlocations.\r\nVisit the galah\r\nR package website to learn more about how to use\r\ngalah\r\nIf you have any questions, comments, or spot any bugs, email us or report an\r\nissue on our GitHub page\r\n\r\n\r\nGet started\r\nInstall from CRAN\r\n\r\n\r\ninstall.packages(\"galah\")\r\n\r\n\r\n\r\n\r\n\r\n\r\n", + "last_modified": "2022-04-26T13:05:53+10:00" }, { "path": "index.html", "title": "Welcome to ALA Labs", "description": "This site is a resource for coding projects that use data sourced from the ALA. We hope that users will find interesting content, whether their focus is ecological modelling, data visualisation, or simply investigating the natural world through a digital lens. Enjoy!", "author": [], - "contents": "\n\n\nData visualisation\n\n\n\nClick to see related posts on each topic\n\n\n\n\n\n\n\n\n\nSummaries\n\n\n\n\n\n\n\n\nMaps\n\n\n\n\n\n\n\n\nTrees\n\n\n\n\n\n\n\n\nTaxonomy\n\n\n\nClick to see related posts on each topic\n\n\n\n\n\n\n\n\n\nPlants\n\n\n\n\n\n\n\n\nBirds\n\n\n\n\n\n\n\n\nMammals\n\n\n\n\n\n\n\n\nWe value\n\n\n\n\n\n\n\n\n Openness\n\nData are most useful when they are widely available and easy to use. We try to encourage the sharing of tools that make data from the Atlas of Living Australia more open and useful for everyone\n\n\n\n\n\n\n Scientific transparency\n\nTransparency is necessary for reproducible science. We encourage that decisions, methods and deviations in our workflow are clear and transparent from the planning phase, to the retrieval and analysis of data, to the final output\n\n\n\n\n\n\n Robust methods\n\nNo method or analytic procedure is perfect. We value consideration for strengths and limitations of each method or analysis to help choose the methods that provide robust, reliable results\n\n\n\n\n\n\n\n", - "last_modified": "2022-04-26T12:26:32+10:00" + "contents": "\r\n\r\n\r\nData visualisation\r\n\r\n\r\n\r\nClick to see related posts on each\r\ntopic\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nSummaries\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nMaps\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nTrees\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nTaxonomy\r\n\r\n\r\n\r\nClick to see related posts on each\r\ntopic\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nPlants\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nBirds\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nMammals\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nWe value\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n Openness\r\n\r\nData are most useful when they are widely available and easy to use. We try to encourage the sharing of tools that make data from the Atlas of Living Australia more open and useful for everyone\r\n\r\n\r\n\r\n\r\n\r\n\r\n Scientific transparency\r\n\r\nTransparency is necessary for reproducible science. We encourage that decisions, methods and deviations in our workflow are clear and transparent from the planning phase, to the retrieval and analysis of data, to the final output\r\n\r\n\r\n\r\n\r\n\r\n\r\n Robust methods\r\n\r\nNo method or analytic procedure is perfect. We value consideration for strengths and limitations of each method or analysis to help choose the methods that provide robust, reliable results\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n", + "last_modified": "2022-04-26T13:05:56+10:00" }, { "path": "info_ala-labs-maintenance.html", @@ -34,15 +34,15 @@ } ], "date": "27/02/2022", - "contents": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nAbout\nPosts\n\n\nResources\n \n▾\n\n\ngalah\nContact Us\n\n\n\n\n\n\n\n☰\n\n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n About\n\n\n Posts\n\n\n \n Resources\n \n \n \n \n galah\n \n \n Contact Us\n \n \n\n \n\n\n \n \n \n \n\n \n \n\n\n\n\n\n\nManaging the ALA Labs Website\nDax Kellie\n27/02/2022\n\n\n\n\n\nInstallation\nInstall the {distill} package from CRAN:\ninstall.packages(\"distill\")\nClone the ala-labs repository to a local directory.\n\n\nBuild the website\nWe use a custom file to build the website. To build the website, run the code in R/ignition_sequence.R.\nIf successful, to see the built website navigate to your local file explorer and opening _site/index.html.\nWe use a custom file to build the website because we have added several bespoke features to the standard distill website format (like auto-updating lists in _people pages and a custom search toolbar). Adding these features means that using the recommended method by {distill} to use the Build Website tab in the Build pane no longer works. Only using rmarkdown::build_site() will also not work correctly.\n\nWhen you build the website, the following things occur:\nAll of the Rmd and md files in the root website directory will be rendered into HTML. Note, however, that markdown files beginning with “_” are not rendered (this is a convention to designate files that are included by top level documents, or in other words, distill knows not to overwrite these folders because they are important).\nPosts will be copied into the _site/posts sub-directory. Pages with post listings will be re-generated.\nThe generated HTML files and any supporting files (e.g. CSS and JavaScript) are copied into the _site directory.\nThe top 3 most recently written Posts will be added to the author’s _people page. (Note: For the website to build, ensure that all people who have authored a post have their own People page, even if this page is not accessible to the public)\nThe HTML files within the _site directory are built and ready to deploy as a standalone static website.\n\nPublishing the website\nThe _site folder has all of the information needed for the static website. The ALA web hosting service has been told to publish what is in the _site folder on the master branch to .\nWhen you are happy with changes or updates to the website, make sure your changes to the _site folder are pushed to the master branch.\n\n\n\nUpdate the website\n\nWebsite workflow\nUpdates to the website should be added to the dev branch of the GitHub repo. In the dev branch, any changes made will not be published to the existing website. The dev branch effectively works as a test site prior to deployment.\n\n\n\nAdd or edit content\n\nSite Output and Navigation\nThe _site.yml controls the main website information and site navigation.\nElements can be edited to alter navigation paths. Here is one example of the navbar for the ALA Labs website:\n---\nnavbar:\n logo:\n image: images/logos/ALA_Logo_Stacked_REV.png\n href: https://www.ala.org.au\n right:\n - text: \"About\"\n href: about.html\n - text: \"Posts\"\n href: posts.html\n - text: \"Resources\"\n menu:\n - text: \"galah\"\n href: galah.html\n - text: \"ALA Labs Style Guide\"\n href: test_style-guide.html\n - icon: fab fa-github\n href: https://github.com/AtlasOfLivingAustralia/science\n---\n\n\nWeb Pages\n\nAdd top-level pages\nTop-level web pages (the ones people see in the top navigation bar) are saved in the top-level folder by their page name (for example, the content for the About page is in about.Rmd).\nAdditional pages can be added by creating a new R Markdown page in the main website folder with the yaml header:\n---\ntitle: \"\"\ndescription: |\n description-goes-here\noutput: \n distill::distill_article:\n css: theme.css\n---\nPages can then be added to the website by their text and href to the site navigation in _site.yml under navbar (see Site Output and Navigation)\n\n\n\nPosts\n\nWriting new posts\nPosts are written in R markdown files. Post templates can be found in the templates folder.\nTo create a new post:\nCreate a new folder in the _posts folder. The folder name should be the date and title of your article (for example, 2022-02-17_sunburst-plots-for-taxonomic-data\nIn your new post folder, save your Post file. You are welcome to use a simple name for your Rmd file (for example, sunburst-plots.Rmd)\nBe sure that you have filled in all the necessary information about your post in the yaml header and in the author card section\nWrite your post\nRender/Knit your document using the R/render_post.R document. Using the code in R/render_post.R allows you to avoid adding personal information to your Post (like your personal email in galah_config()). The most recently knitted version of your file is what will appear on the website\nWhen you are happy with your post, build the website. After the website it built, your rendered post should appear in the Posts listing page\n\n\n\n\nTheme customisation\nDistill uses a CSS framework that can be fully customised. CSS theme settings are in theme.css. See the Distill website page for instructions on how to edit additional website properties.\nSome properties can be found in labeled sections, like settings for website fonts, header and footer sections. For example, website header settings can be found in the .distill-site-header section:\n/*-- WEBSITE HEADER + FOOTER --*/\n/* These properties only apply to Distill sites and blogs */\n\n.distill-site-header {\n --title-size: 18px; \n --text-color: #ff414b; /* edited */\n --text-size: 15px;\n --hover-color: #dd424c; /* edited */\n --bkgd-color: #ffd8db; /* edited */\n}\nOther properties are edited using custom CSS. These are (mostly) labelled and relate either to built-in distill html elements or html elements built in R/function.R.\n\nCustom html elements using R functions\nR/functions.R contains functions that create html elements within the ALA Labs website. R/functions.R uses the {htmlTools} package to write html code using R syntax.\nFor example, the following R function and html code are equivalent. They both will create a link of class = \"article-link\". This class name can be used with CSS in theme.css to edit the element style.\n#\n#my text<\/a>\n# R function\nadd_link_to_article <- function(title, url){\n tags$a(\n class = \"article-link\",\n href = url,\n \"text\"\n )\n}\n\nadd_link_to_article(text = \"my text\", url = \"https://ala.org.au/\")\nThe benefits of writing R functions to create html elements are:\nRather than trying to edit html elements that the Distill package runs in the back-end to build the website, adding html elements using R functions is easier and less prone to errors\nIt is easier for people familiar in R to reuse and edit the content of existing html elements created by R functions\nTo use functions in the R/functions.R file on a specific website page or post, the following code block must be added below the YAML header\n ```{r, include=FALSE}\n library(htmltools)\n source(\"R/functions.R\")\n```\n\n\n\n\n\n\n\n\n\n\n/*-- Copyright --*/\n\n.copyright-container {\n display: flex;\n width: 100%;\n margin-bottom: 30px; \n margin-top: 30px;\n}\n\n.copyright-row {\n width: 100%;\n height: 60px;\n position: relative;\n margin-bottom: -1.75vw;\n background-color: #c44d34;\n background-position: bottom;\n background-size: cover;\n display: flex;\n}\n\n.copyright-width {\n display: flex;\n width: 100%;\n max-width: 1140px;\n padding-left: 15px;\n padding-right: 15px;\n padding-bottom: 30px;\n margin: auto;\n}\n\n.copyright-column-left {\n float: left;\n width: 50%;\n max-width: 50%;\n padding-left: 15px;\n padding-right: 15px;\n position: relative;\n min-height: 1px;\n display: flex;\n flex-direction:column;\n display:inline-block;\n}\n\n.copyright-column-right {\n float: right;\n text-align: right;\n width: 50%;\n max-width: 50%;\n padding-left: 15px;\n padding-right: 15px;\n position: relative;\n min-height: 1px;\n display: flex;\n flex-direction: column;\n display:inline-block;\n}\n\n.copyright-text {\n color: white;\n font-size: 12px;\n padding-top: 10px}\n\n.copyright-text a:hover {\n color: #222322;\n}\n\n\n/*-- Logos and acknowledgement --*/\n\n.footer-container { \n display: flex;\n width: 100%;\n max-width: 1140px;\n padding-left: 15px;\n padding-right: 15px;\n padding-top: 50px;\n padding-bottom: 50px; \n margin: auto;\n}\n\n.distill-site-footer {\n padding-top: 0px;\n padding-bottom: 10px;\n text-align: left;\n}\n\n.row-footer { /* Do not change the name of this column - it causes issues */\n display: -ms-flexbox;\n display: flex;\n -ms-flex-wrap: wrap;\n flex-wrap: wrap;\n}\n\n.column-footer { /* Do not change the name of this column - it causes issues */\n float: left;\n width: 50%;\n max-width: 50%;\n padding-left: 15px;\n padding-right: 15px;\n position: relative;\n min-height: 1px;\n display: flex;\n flex-direction:column;\n display:inline-block;\n}\n\n\n.footer-logo {\n text-align: left;\n margin:5px;\n}\n\n.footer-text {\n font-family: \"Lato\", sans-serif;\n font-weight: 700;\n font-size: 1.1rem;\n line-height: 1.2;\n color: #565656;\n text-align: left;\n}\n\n.footer-text-small {\n font-family: Roboto,sans-serif;\n font-size: .82rem;\n font-weight: 400;\n line-height: 1.5;\n font-family: \"Roboto\", sans-serif;\n font-weight: 400;\n color: #565656;\n text-align: left;\n margin-top: 10px;\n}\n\nd-article a:hover {\n color: #E06E53;\n}\n\n/*-- Back to ALA button --*/\n\n.link-container {\n margin-top: 30px;\n margin-bottom: 30px;\n display: block;\n width: 30em;\n margin-left: auto;\n margin-right: auto;\n}\n\n.link-button {\n border: none;\n color: #E06E53;\n padding: 10px 15px;\n text-align: center;\n text-decoration: none;\n display: inline-block;\n font-size: 18px;\n margin: 4px 2px;\n transition-duration: 0.3s;\n cursor: pointer;\n background-color: white; \n border: 2px solid #E06E53;\n border-radius: 10px;\n}\n\n.link-button:hover {\n background-color: #222322;\n border: 2px solid #222322;\n color: #E06E53;\n}\n\n.ala-icon {\n height: 20px;\n width: auto;\n /* Other styles here */\n}\n\n@media screen and (max-width: 952px) {\n .link-container {\n margin: auto;\n margin-top: 20px;\n margin-bottom: 20px;\n }\n \n .footer-container {\n width: 100%;\n max-width: 100%;\n display: flex;\n flex-wrap: wrap;\n }\n .column-footer {\n float:left;\n width: 100%;\n max-width: 100%;\n display: block;\n margin-top: 20px;\n margin-bottom: 20px;\n }\n .link-container {\n display: block;\n width: 20em;\n margin: auto;\n }\n \n .copyright-column-left {\n width: 50%;\n }\n \n .copyright-column-right {\n width: 50%;\n }\n \n .copyright-row {\n height: 100px;\n margin-bottom: -3.75vw;\n }\n}\n\n.logo-ncris {\n width: 150px;\n height: 109px;\n margin-right: 15px;\n margin-top: 20px;\n}\n\n.logo-csiro {\n width: 109px;\n height: 109px;\n margin-right: 15px;\n margin-top: 20px;\n}\n\n.logo-gbif {\n width: 207px;\n height: 109px;\n}\n\n\n \n \n \n The ALA is made possible by contributions from its partners, is supported \n by \n NCRIS, is hosted by \n CSIRO, and is the Australian node of\n GBIF.\n \n \n \n \n \n \n \n \n \n Acknowledgement of Traditional Owners and Country\n The Atlas of Living Australia acknowledges Australia's Traditional Owners \n and pays respect to the past and present Elders of the nation's Aboriginal \n and Torres Strait Islander communities. We honour and celebrate the \n spiritual, cultural and customary connections of Traditional Owners to \n country and the biodiversity that forms part of that country.\n \n\n\n\n \n \n \n \n This work is licensed under a Creative Commons Attribution 3.0 Australia License\n \n \n \n \n \n Copyright\n   \n Terms of Use\n   \n System Status\n \n \n \n \n\n\n\n\n\n\n\n\n\n\n\n// add bootstrap table styles to pandoc tables\nfunction bootstrapStylePandocTables() {\n $('tr.odd').parent('tbody').parent('table').addClass('table table-condensed');\n}\n$(document).ready(function () {\n bootstrapStylePandocTables();\n});\n\n\n\n$(document).ready(function () {\n window.buildTabsets(\"TOC\");\n});\n\n$(document).ready(function () {\n $('.tabset-dropdown > .nav-tabs > li').click(function () {\n $(this).parent().toggleClass('nav-tabs-open');\n });\n});\n\n (function () {\n var script = document.createElement(\"script\");\n script.type = \"text/javascript\";\n script.src = \"https://mathjax.rstudio.com/latest/MathJax.js?config=TeX-AMS-MML_HTMLorMML\";\n document.getElementsByTagName(\"head\")[0].appendChild(script);\n })();\n", - "last_modified": "2022-04-26T12:26:32+10:00" + "contents": "\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nAbout\r\nPosts\r\n\r\n\r\nResources\r\n \r\n▾\r\n\r\n\r\ngalah\r\nContact Us\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n☰\r\n\r\n\r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n\r\n\r\n About\r\n\r\n\r\n Posts\r\n\r\n\r\n \r\n Resources\r\n \r\n \r\n \r\n \r\n galah\r\n \r\n \r\n Contact Us\r\n \r\n \r\n\r\n \r\n\r\n\r\n \r\n \r\n \r\n \r\n\r\n \r\n \r\n\r\n\r\n\r\n\r\n\r\n\r\nManaging the ALA Labs Website\r\nDax Kellie\r\n27/02/2022\r\n\r\n\r\n\r\n\r\n\r\nInstallation\r\nInstall the {distill} package from CRAN:\r\ninstall.packages(\"distill\")\r\nClone the ala-labs repository to a local directory.\r\n\r\n\r\nBuild the website\r\nWe use a custom file to build the website. To build the website, run\r\nthe code in R/ignition_sequence.R.\r\nIf successful, to see the built website navigate to your local file\r\nexplorer and opening _site/index.html.\r\nWe use a custom file to build the website because we have added\r\nseveral bespoke features to the standard distill website format (like\r\nauto-updating lists in _people pages and a custom search\r\ntoolbar). Adding these features means that using the recommended method\r\nby {distill} to\r\nuse the Build Website tab in the Build pane no longer\r\nworks. Only using rmarkdown::build_site() will also not\r\nwork correctly.\r\n\r\nWhen you build the website, the following things occur:\r\nAll of the Rmd and md files in the root\r\nwebsite directory will be rendered into HTML. Note, however, that\r\nmarkdown files beginning with “_” are not rendered (this is a convention\r\nto designate files that are included by top level documents, or in other\r\nwords, distill knows not to overwrite these folders because they are\r\nimportant).\r\nPosts will be copied into the _site/posts\r\nsub-directory. Pages with post listings will be re-generated.\r\nThe generated HTML files and any supporting files (e.g. CSS and\r\nJavaScript) are copied into the _site directory.\r\nThe top 3 most recently written Posts will be added to the\r\nauthor’s _people page. (Note: For the website to build,\r\nensure that all people who have authored a post have their own People\r\npage, even if this page is not accessible to the public)\r\nThe HTML files within the _site directory are built and\r\nready to deploy as a standalone static website.\r\n\r\nPublishing the website\r\nThe _site folder has all of the information needed for\r\nthe static website. The ALA web hosting service has been told to publish\r\nwhat is in the _site folder on the master branch to .\r\nWhen you are happy with changes or updates to the website, make sure\r\nyour changes to the _site folder are pushed to the master\r\nbranch.\r\n\r\n\r\n\r\nUpdate the website\r\n\r\nWebsite workflow\r\nUpdates to the website should be added to the dev branch\r\nof the GitHub repo. In the dev branch, any changes made\r\nwill not be published to the existing website. The dev\r\nbranch effectively works as a test site prior to deployment.\r\n\r\n\r\n\r\nAdd or edit content\r\n\r\nSite Output and Navigation\r\nThe _site.yml controls the main website information and\r\nsite navigation.\r\nElements can be edited to alter navigation paths. Here is one example\r\nof the navbar for the ALA Labs website:\r\n---\r\nnavbar:\r\n logo:\r\n image: images/logos/ALA_Logo_Stacked_REV.png\r\n href: https://www.ala.org.au\r\n right:\r\n - text: \"About\"\r\n href: about.html\r\n - text: \"Posts\"\r\n href: posts.html\r\n - text: \"Resources\"\r\n menu:\r\n - text: \"galah\"\r\n href: galah.html\r\n - text: \"ALA Labs Style Guide\"\r\n href: test_style-guide.html\r\n - icon: fab fa-github\r\n href: https://github.com/AtlasOfLivingAustralia/science\r\n---\r\n\r\n\r\nWeb Pages\r\n\r\nAdd top-level pages\r\nTop-level web pages (the ones people see in the top navigation bar)\r\nare saved in the top-level folder by their page name (for example, the\r\ncontent for the About page is in about.Rmd).\r\nAdditional pages can be added by creating a new R Markdown page in\r\nthe main website folder with the yaml header:\r\n---\r\ntitle: \"\"\r\ndescription: |\r\n description-goes-here\r\noutput: \r\n distill::distill_article:\r\n css: theme.css\r\n---\r\nPages can then be added to the website by their text and\r\nhref to the site navigation in _site.yml under\r\nnavbar (see Site\r\nOutput and Navigation)\r\n\r\n\r\n\r\nPosts\r\n\r\nWriting new posts\r\nPosts are written in R markdown files. Post templates can be found in\r\nthe templates folder.\r\nTo create a new post:\r\nCreate a new folder in the _posts folder. The folder\r\nname should be the date and title of your article (for example,\r\n2022-02-17_sunburst-plots-for-taxonomic-data\r\nIn your new post folder, save your Post file. You are welcome to\r\nuse a simple name for your Rmd file (for example,\r\nsunburst-plots.Rmd)\r\nBe sure that you have filled in all the necessary information\r\nabout your post in the yaml header and in the author card\r\nsection\r\nWrite your post\r\nRender/Knit your document using the\r\nR/render_post.R document. Using the code in\r\nR/render_post.R allows you to avoid adding personal\r\ninformation to your Post (like your personal email in\r\ngalah_config()). The most recently knitted version of your\r\nfile is what will appear on the website\r\nWhen you are happy with your post, build the website. After the\r\nwebsite it built, your rendered post should appear in the Posts listing\r\npage\r\n\r\n\r\n\r\n\r\nTheme customisation\r\nDistill uses a CSS framework that can be fully customised. CSS theme\r\nsettings are in theme.css. See the Distill\r\nwebsite page for instructions on how to edit additional website\r\nproperties.\r\nSome properties can be found in labeled sections, like settings for\r\nwebsite fonts, header and footer sections. For example, website header\r\nsettings can be found in the .distill-site-header\r\nsection:\r\n/*-- WEBSITE HEADER + FOOTER --*/\r\n/* These properties only apply to Distill sites and blogs */\r\n\r\n.distill-site-header {\r\n --title-size: 18px; \r\n --text-color: #ff414b; /* edited */\r\n --text-size: 15px;\r\n --hover-color: #dd424c; /* edited */\r\n --bkgd-color: #ffd8db; /* edited */\r\n}\r\nOther properties are edited using custom CSS. These are (mostly)\r\nlabelled and relate either to built-in distill html elements or html\r\nelements built in R/function.R.\r\n\r\nCustom html elements using R functions\r\nR/functions.R contains functions that create html\r\nelements within the ALA Labs website. R/functions.R uses\r\nthe {htmlTools}\r\npackage to write html code using R syntax.\r\nFor example, the following R function and html code are equivalent.\r\nThey both will create a link of class = \"article-link\".\r\nThis class name can be used with CSS in theme.css to edit\r\nthe element style.\r\n#\r\n#my text<\/a>\r\n# R function\r\nadd_link_to_article <- function(title, url){\r\n tags$a(\r\n class = \"article-link\",\r\n href = url,\r\n \"text\"\r\n )\r\n}\r\n\r\nadd_link_to_article(text = \"my text\", url = \"https://ala.org.au/\")\r\nThe benefits of writing R functions to create html elements are:\r\nRather than trying to edit html elements that the Distill package\r\nruns in the back-end to build the website, adding html elements using R\r\nfunctions is easier and less prone to errors\r\nIt is easier for people familiar in R to reuse and edit the\r\ncontent of existing html elements created by R functions\r\nTo use functions in the R/functions.R file on a specific\r\nwebsite page or post, the following code block must be added below the\r\nYAML header\r\n ```{r, include=FALSE}\r\n library(htmltools)\r\n source(\"R/functions.R\")\r\n```\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n/*-- Copyright --*/\r\n\r\n.copyright-container {\r\n display: flex;\r\n width: 100%;\r\n margin-bottom: 30px;\r\n margin-top: 30px;\r\n}\r\n\r\n.copyright-row {\r\n width: 100%;\r\n height: 66px;\r\n position: relative;\r\n margin-bottom: -1.75vw;\r\n background-color: #c44d34;\r\n background-position: bottom;\r\n background-size: cover;\r\n display: flex;\r\n}\r\n\r\n.copyright-width {\r\n display: flex;\r\n width: 100%;\r\n max-width: 1140px;\r\n padding-left: 15px;\r\n padding-right: 15px;\r\n padding-bottom: 30px;\r\n margin: auto;\r\n}\r\n\r\n.copyright-column-left {\r\n float: left;\r\n width: 50%;\r\n max-width: 50%;\r\n padding-left: 15px;\r\n padding-right: 15px;\r\n position: relative;\r\n min-height: 1px;\r\n display: flex;\r\n flex-direction:column;\r\n display:inline-block;\r\n}\r\n\r\n.copyright-column-right {\r\n float: right;\r\n text-align: right;\r\n width: 50%;\r\n max-width: 50%;\r\n padding-left: 15px;\r\n padding-right: 15px;\r\n position: relative;\r\n min-height: 1px;\r\n display: flex;\r\n flex-direction: column;\r\n display:inline-block;\r\n}\r\n\r\n.copyright-text {\r\n color: white;\r\n font-size: 12px;\r\n padding-top: 10px}\r\n\r\n.copyright-text a:hover {\r\n color: #222322;\r\n}\r\n\r\n\r\n/*-- Logos and acknowledgement --*/\r\n\r\n.footer-container {\r\n display: flex;\r\n width: 100%;\r\n max-width: 1140px;\r\n padding-left: 15px;\r\n padding-right: 15px;\r\n padding-top: 50px;\r\n padding-bottom: 50px;\r\n margin: auto;\r\n}\r\n\r\n.distill-site-footer {\r\n padding-top: 0px;\r\n padding-bottom: 10px;\r\n text-align: left;\r\n}\r\n\r\n.row-footer { /* Do not change the name of this column - it causes issues */\r\n display: -ms-flexbox;\r\n display: flex;\r\n -ms-flex-wrap: wrap;\r\n flex-wrap: wrap;\r\n}\r\n\r\n.column-footer { /* Do not change the name of this column - it causes issues */\r\n float: left;\r\n width: 50%;\r\n max-width: 50%;\r\n padding-left: 15px;\r\n padding-right: 15px;\r\n position: relative;\r\n min-height: 1px;\r\n display: flex;\r\n flex-direction:column;\r\n display:inline-block;\r\n}\r\n\r\n\r\n.footer-logo {\r\n text-align: left;\r\n margin:5px;\r\n}\r\n\r\n.footer-text {\r\n font-family: \"Lato\", sans-serif;\r\n font-weight: 700;\r\n font-size: 1.1rem;\r\n line-height: 1.2;\r\n color: #565656;\r\n text-align: left;\r\n}\r\n\r\n.footer-text-small {\r\n font-family: Roboto,sans-serif;\r\n font-size: .82rem;\r\n font-weight: 400;\r\n line-height: 1.5;\r\n font-family: \"Roboto\", sans-serif;\r\n font-weight: 400;\r\n color: #565656;\r\n text-align: left;\r\n margin-top: 10px;\r\n}\r\n\r\nd-article a:hover {\r\n color: #E06E53;\r\n}\r\n\r\n/*-- Back to ALA button --*/\r\n\r\n.link-container {\r\n margin-top: 30px;\r\n margin-bottom: 30px;\r\n display: block;\r\n width: 30em;\r\n margin-left: auto;\r\n margin-right: auto;\r\n}\r\n\r\n.link-button {\r\n border: none;\r\n color: #E06E53;\r\n padding: 10px 15px;\r\n text-align: center;\r\n text-decoration: none;\r\n display: inline-block;\r\n font-size: 18px;\r\n margin: 4px 2px;\r\n transition-duration: 0.3s;\r\n cursor: pointer;\r\n background-color: white;\r\n border: 2px solid #E06E53;\r\n border-radius: 10px;\r\n}\r\n\r\n.link-button:hover {\r\n background-color: #222322;\r\n border: 2px solid #222322;\r\n color: #E06E53;\r\n}\r\n\r\n.ala-icon {\r\n height: 20px;\r\n width: auto;\r\n /* Other styles here */\r\n}\r\n\r\n@media screen and (max-width: 952px) {\r\n .link-container {\r\n margin: auto;\r\n margin-top: 20px;\r\n margin-bottom: 20px;\r\n }\r\n\r\n .footer-container {\r\n width: 100%;\r\n max-width: 100%;\r\n display: flex;\r\n flex-wrap: wrap;\r\n }\r\n .column-footer {\r\n float:left;\r\n width: 100%;\r\n max-width: 100%;\r\n display: block;\r\n margin-top: 20px;\r\n margin-bottom: 20px;\r\n }\r\n .link-container {\r\n display: block;\r\n width: 20em;\r\n margin: auto;\r\n }\r\n\r\n .copyright-column-left {\r\n width: 50%;\r\n }\r\n\r\n .copyright-column-right {\r\n width: 50%;\r\n }\r\n\r\n .copyright-row {\r\n height: 100px;\r\n margin-bottom: -3.75vw;\r\n }\r\n}\r\n\r\n.logo-ncris {\r\n width: 150px;\r\n height: 109px;\r\n margin-right: 15px;\r\n margin-top: 20px;\r\n}\r\n\r\n.logo-csiro {\r\n width: 109px;\r\n height: 109px;\r\n margin-right: 15px;\r\n margin-top: 20px;\r\n}\r\n\r\n.logo-gbif {\r\n width: 207px;\r\n height: 109px;\r\n}\r\n\r\n\r\n\r\n \r\n \r\n The ALA is made possible by contributions from its partners, is supported\r\n by\r\n NCRIS, is hosted by\r\n CSIRO, and is the Australian node of\r\n GBIF.\r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n Acknowledgement of Traditional Owners and Country\r\n The Atlas of Living Australia acknowledges Australia's Traditional Owners\r\n and pays respect to the past and present Elders of the nation's Aboriginal\r\n and Torres Strait Islander communities. We honour and celebrate the\r\n spiritual, cultural and customary connections of Traditional Owners to\r\n country and the biodiversity that forms part of that country.\r\n \r\n\r\n\r\n\r\n \r\n \r\n \r\n \r\n This work is licensed under a Creative Commons Attribution 3.0 Australia License\r\n \r\n \r\n \r\n \r\n \r\n Copyright\r\n   \r\n Terms of Use\r\n   \r\n System Status\r\n \r\n \r\n \r\n \r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n// add bootstrap table styles to pandoc tables\r\nfunction bootstrapStylePandocTables() {\r\n $('tr.odd').parent('tbody').parent('table').addClass('table table-condensed');\r\n}\r\n$(document).ready(function () {\r\n bootstrapStylePandocTables();\r\n});\r\n\r\n\r\n\r\n$(document).ready(function () {\r\n window.buildTabsets(\"TOC\");\r\n});\r\n\r\n$(document).ready(function () {\r\n $('.tabset-dropdown > .nav-tabs > li').click(function () {\r\n $(this).parent().toggleClass('nav-tabs-open');\r\n });\r\n});\r\n\r\n (function () {\r\n var script = document.createElement(\"script\");\r\n script.type = \"text/javascript\";\r\n script.src = \"https://mathjax.rstudio.com/latest/MathJax.js?config=TeX-AMS-MML_HTMLorMML\";\r\n document.getElementsByTagName(\"head\")[0].appendChild(script);\r\n })();\r\n", + "last_modified": "2022-04-26T13:05:57+10:00" }, { "path": "posts.html", "title": "Posts", "author": [], - "contents": "\n\n\n\n", - "last_modified": "2022-04-26T12:26:33+10:00" + "contents": "\r\n\r\n\r\n\r\n", + "last_modified": "2022-04-26T13:06:02+10:00" }, { "path": "test_style-guide.html", @@ -50,8 +50,8 @@ "description": "Style guide for R script and R Markdown documents\n", "author": [], "date": "`r Sys.Date()`", - "contents": "\n\nContents\nGetting started\nCreate an ALA R Markdown file\nKniting to HTML safely\n\nPush to Github\nReproducible workflows\nSafe paths\n\nWriting understandable code\nSummarise your code\nStyle code using styler\n\nStyling your document\nTables\nFigures\nModel output\nCode chunk options\nExample: echo and eval\n\n\nCool Distill Things\nCode folding\n\n\nThis document details the Atlas of Living Australia style guide for creating html files from R Markdown\nGetting started\nCreate an ALA R Markdown file\nTo create new R Markdown file that follows the ALA style guide, install the alatheme package.\n\n\ndevtools::install_github(\"AtlasOfLivingAustralia/alatheme\")\n\n\n\nThe {alatheme} package contains ALA templates for R Markdown files. After installing, create a new template R markdown file by selecting File –> New File –> New R Markdown which opens the “New R Markdown” pane\nIn the “New R Markdown” pane, select From Template and choose from the list of {alatheme} templates.\nFill in your information where necessary in the yaml header and in the required code and text chunks.\nKniting to HTML safely\nIn R Studio, clicking the Knit button in the left upper menu above your script renders your R Markdown file into an HTML file.\nHowever, if your code requires a config setting with personal or sensitive information, . However, packages like galah require users to enter an email address prior to downloading data using galah_config(email = \"your-email@email.com\"). Knitting without this information in the document will cause an error.\nA better workflow for rendering R Markdown files is by using a separate script combined with the here package. Running the following script outside of your R Markdown file will render your R Markdown file in the file’s directory:\n\n\nlibrary(galah)\ngalah_config(email = \"your-email@email.com\")\n\nlibrary(here)\npath <- here(\"folder\", \"subfolder\", \"subfolder\", \"your-code.Rmd\")\nrmarkdown::render(path)\n\n\n\n\nPush to Github\n\nReproducible workflows\nSafe paths\nUsing setwd() to set a working directory can cause issues when setting folder paths, and can easily prevent a script or workflow from being reproducible. \nOne alternative suggested by Jenny Bryan is to use here::here() from the here package to build your directory paths. here::here() automatically creates paths relative to the top level directory.\nRead a full description of how to use the here() function to create safe paths in Chapter 3 of What They Forgot to Teach You About R\nFor example:\n\n\nlibrary(tidyverse)\nlibrary(here)\n\ndf <- read.delim(here(\"projects\", , \"style_guide\", \"data.csv\"))\np <- ggplot(df, aes(x, y)) + geom_point()\nggsave(here(\"plots\", \"foofy_scatterplot.png\"))\n\n\n\nWriting understandable code\nSummarise your code\nPeople who read your code should be able to understand what it does and why you made the choices you did. It is helpful to include brief summaries of what your code is meant to do, and the logic of any decision you might make after seeing a result or output. For example, if your code used a statistical model, include your basic interpretation and subsequent decision process after seeing the results. \nUse sentences, paragraphs and code chunks to add structure to your script. Try to create a flowing, logical story of how to run an analysis or make a data visualisation. \nCode chunks should be brief. They should be supplemented with notes, descriptions or visual output that provides context to what a given chunk of code is meant to do. Users should be able to follow each transformation that is made to your data, each output that results from a transformation, and any final output. \nDo not expect your reader to know what they should see in your output. If there is something in your code output that you wish to highlight, it is best to tell your reader what they should see in your output and how you have interpreted it.\nThere is no single correct code chunk size; you must use your best judgement. But if it seems that the result of several lines of code is unclear, you may need to split the code chunks to make the results easier to follow. \nStyle code using styler\nUse the styler package to ensure your code is formatted correctly and/or consistently, The styler package formats your code according to the tidyverse style guide (or another custome style) prior to uploading a finished file document. \nSee the styler github and tidyverse pages for more information \nInstall the styler package to add style buttons to your R Markdown Addins dropdown menu. Clicking Addins -> style active file will reformat the code in your active file. Clicking ** Addins** -> style active section will reformat code in your current section. \nAlternatively, you style_file() or style_text() can be run in the console. \nStyling your document\nSee the R Markdown cheat sheet for more information on R Markdown formatting \nFor tips to make your R Markdown output pretty, see Pimp my RMD by Yan Holtz \nTables\nSeveral packages can create tables with nice formatting.\nOne example is kableExtra:\n\n\nlibrary(tidyverse)\nlibrary(kableExtra)\nkable(\n head(mtcars, n=5)) %>% kable_styling()\n\n\n\n\n\nmpg\n\n\ncyl\n\n\ndisp\n\n\nhp\n\n\ndrat\n\n\nwt\n\n\nqsec\n\n\nvs\n\n\nam\n\n\ngear\n\n\ncarb\n\n\nMazda RX4\n\n\n21.0\n\n\n6\n\n\n160\n\n\n110\n\n\n3.90\n\n\n2.620\n\n\n16.46\n\n\n0\n\n\n1\n\n\n4\n\n\n4\n\n\nMazda RX4 Wag\n\n\n21.0\n\n\n6\n\n\n160\n\n\n110\n\n\n3.90\n\n\n2.875\n\n\n17.02\n\n\n0\n\n\n1\n\n\n4\n\n\n4\n\n\nDatsun 710\n\n\n22.8\n\n\n4\n\n\n108\n\n\n93\n\n\n3.85\n\n\n2.320\n\n\n18.61\n\n\n1\n\n\n1\n\n\n4\n\n\n1\n\n\nHornet 4 Drive\n\n\n21.4\n\n\n6\n\n\n258\n\n\n110\n\n\n3.08\n\n\n3.215\n\n\n19.44\n\n\n1\n\n\n0\n\n\n3\n\n\n1\n\n\nHornet Sportabout\n\n\n18.7\n\n\n8\n\n\n360\n\n\n175\n\n\n3.15\n\n\n3.440\n\n\n17.02\n\n\n0\n\n\n0\n\n\n3\n\n\n2\n\n\nkableExtra is also able to be used using magittr or base piping:\n\n\nmtcars %>% \n group_by(gear) %>% \n dplyr::summarise(cyl = mean(cyl),\n disp = mean(disp)) %>%\n kbl() %>%\n kable_styling()\n\n\n\ngear\n\n\ncyl\n\n\ndisp\n\n\n3\n\n\n7.466667\n\n\n326.3000\n\n\n4\n\n\n4.666667\n\n\n123.0167\n\n\n5\n\n\n6.000000\n\n\n202.4800\n\n\n\n\nhead(mtcars |>\n subset(select = c(\"cyl\", \"disp\", \"gear\")), n = 5) |>\n kbl() |>\n kable_styling()\n\n\n\n\n\ncyl\n\n\ndisp\n\n\ngear\n\n\nMazda RX4\n\n\n6\n\n\n160\n\n\n4\n\n\nMazda RX4 Wag\n\n\n6\n\n\n160\n\n\n4\n\n\nDatsun 710\n\n\n4\n\n\n108\n\n\n4\n\n\nHornet 4 Drive\n\n\n6\n\n\n258\n\n\n3\n\n\nHornet Sportabout\n\n\n8\n\n\n360\n\n\n3\n\n\nFigures\nControlling the size of figures can help when you want people to focus on trends of many plots or on details of a single plot. \nYou can control how many columns your plots created in a chunk are printed into.\nAdd out.width = c(\"50%\", \"50%\"), fig.show = \"hold\":\n\n``{r out.width=c('50%', '50%'), fig.show='hold'}\nboxplot(1:10)\nplot(rnorm(10))\n`\n\n\n\n\nControl other parts of figure output in the chunk header as well. For example:\n\n``{r, fig.align=\"center\", fig.width=6, fig.height=6, fig.cap=\"Figure: Here is a really important caption.\"}\n\n\n\nlibrary(tidyverse)\nmpg %>%\n ggplot( aes(x=reorder(class, hwy), y=hwy, fill=class)) + \n geom_boxplot() +\n xlab(\"class\") +\n theme(legend.position=\"none\")\n\n\n\n\nFigure 1: Figure: Here is a really important caption.\n\n\n\nModel output\nWant to show more readable model output? Use the model_parameters() function from the parameters package to make your output clear and organised. The model_parameters() function can be considered as a lightweight alternative to broom::tidy():\n\n\nlibrary(parameters)\nmodel <- lm(Sepal.Width ~ Petal.Length * Species + Petal.Width, data = iris)\n\n# regular model parameters\nmodel_parameters(model)\n\n\nParameter | Coefficient | SE | 95% CI | t(143) | p\n-------------------------------------------------------------------------------------------\n(Intercept) | 2.89 | 0.36 | [ 2.18, 3.60] | 8.01 | < .001\nPetal Length | 0.26 | 0.25 | [-0.22, 0.75] | 1.07 | 0.287 \nSpecies [versicolor] | -1.66 | 0.53 | [-2.71, -0.62] | -3.14 | 0.002 \nSpecies [virginica] | -1.92 | 0.59 | [-3.08, -0.76] | -3.28 | 0.001 \nPetal Width | 0.62 | 0.14 | [ 0.34, 0.89] | 4.41 | < .001\nPetal Length * Species [versicolor] | -0.09 | 0.26 | [-0.61, 0.42] | -0.36 | 0.721 \nPetal Length * Species [virginica] | -0.13 | 0.26 | [-0.64, 0.38] | -0.50 | 0.618 \n\nmodel_parameters() can also be used for parameter standardization:\n\n\n# standardized parameters\nmodel_parameters(model, standardize = \"refit\")\n\n\nParameter | Coefficient | SE | 95% CI | t(143) | p\n-------------------------------------------------------------------------------------------\n(Intercept) | 3.59 | 1.30 | [ 1.01, 6.17] | 2.75 | 0.007 \nPetal Length | 1.07 | 1.00 | [-0.91, 3.04] | 1.07 | 0.287 \nSpecies [versicolor] | -4.62 | 1.31 | [-7.21, -2.03] | -3.53 | < .001\nSpecies [virginica] | -5.51 | 1.38 | [-8.23, -2.79] | -4.00 | < .001\nPetal Width | 1.08 | 0.24 | [ 0.59, 1.56] | 4.41 | < .001\nPetal Length * Species [versicolor] | -0.38 | 1.06 | [-2.48, 1.72] | -0.36 | 0.721 \nPetal Length * Species [virginica] | -0.52 | 1.04 | [-2.58, 1.54] | -0.50 | 0.618 \n\n\nCode chunk options\nSee the R markdown documentation to view chunk options. \nExample: echo and eval\nReaders should be able to identify where every file in a workflow comes from (little is more frustrating than wondering where a required data file is located). Code used to load or extract data (from galah, for example) should be clearly identified. \nHowever, some code takes a very long time to run and you may have saved it locally or in a repository to save time and/or space. In this case, it is possible to show code without running it. \nAdd eval = FALSE to the chunk header to display the code but prevent the chunk from running:\n\n``{r, eval = FALSE}\nala_counts(group_by = \"phylum\")\n`\n\nYou can then load a local file in the background, without showing the code. \nAdd echo = FALSE to your chunk header to run the code but prevent the chunk from displaying:\n\n``{r, echo = FALSE}\ndata <- readRDS(file = \"local_file.rds\")\n`\n\nCool Distill Things\nCode folding\nDoes a plot require a lot of code that distracts from the plot itself? Distill has a really nice code folding feature:\n\n\nShow code\n\nlibrary(galah)\nlibrary(data.table)\nlibrary(dtplyr) # provides data.table backend for dplyr\nlibrary(tidyverse)\nlibrary(lubridate) # for dates\nlibrary(ggpubr)\n\n# Might need to config email to get ALA data\nala_config(email = \"dax.kellie@csiro.au\")\n\n\n# Get records\noccurrences_Tas <- ala_occurrences(filters = select_filters(stateProvince = \"Tasmania\",\n year = seq(2018, 2020),\n basisOfRecord = \"HumanObservation\"))\noccurrences_Tas <- setDT(occurrences_Tas) # make data.table\n\n\n\n\n\nShow code\n\n# date format\noccurrences_Tas$eventDate <- as_date(occurrences_Tas$eventDate)\n\n# Wrangling\noccurrences_Tas <- occurrences_Tas %>%\n mutate(\n month = month(eventDate, label = TRUE), # add month col\n year = year(eventDate) # add year col\n ) %>%\n filter(!is.na(eventDate)) %>% # remove NAs\n as.data.table()\n# any(is.na(occurrences_Tas$day))\n\n# get month counts\nmonth_counts <- occurrences_Tas[, .N, by = c(\"month\")]\nmonth_counts$month <- ordered(factor(month_counts$month,\n levels = c(\n \"Jan\", \"Feb\", \"Mar\", \"Apr\", \"May\", \"Jun\",\n \"Jul\", \"Aug\", \"Sep\", \"Oct\", \"Nov\", \"Dec\"\n )\n))\n\nggdotchart(month_counts,\n x = \"month\", y = \"N\",\n color = \"month\", # Color by groups\n palette = \"viridis\", # Custom color palette\n sorting = \"none\", # Sort value in descending order\n add = \"segments\", # Add segments from y = 0 to dots\n legend = \"none\",\n rotate = FALSE,\n dot.size = 4,\n add.params = list(color = \"month\", size = 1),\n title = \"Records added to ALA by month (2018-2020)\",\n ylab = \"Number of records\",\n xlab = \"Month\",\n ggtheme = theme_pubr() # ggplot2 theme\n)\n\n\n\n\n\n\n", - "last_modified": "2022-04-26T12:26:35+10:00" + "contents": "\r\n\r\nContents\r\nGetting started\r\nCreate an ALA R Markdown\r\nfile\r\nKniting to HTML safely\r\n\r\nPush to Github\r\nReproducible workflows\r\nSafe paths\r\n\r\nWriting understandable\r\ncode\r\nSummarise your code\r\nStyle code using\r\nstyler\r\n\r\nStyling your document\r\nTables\r\nFigures\r\nModel output\r\nCode chunk options\r\nExample: echo and\r\neval\r\n\r\n\r\nCool Distill Things\r\nCode folding\r\n\r\n\r\nThis document details the Atlas of Living Australia style guide for\r\ncreating html files from R Markdown\r\nGetting started\r\nCreate an ALA R Markdown\r\nfile\r\nTo create new R Markdown file that follows the ALA style guide,\r\ninstall the alatheme\r\npackage.\r\n\r\n\r\ndevtools::install_github(\"AtlasOfLivingAustralia/alatheme\")\r\n\r\n\r\n\r\nThe {alatheme} package contains ALA templates for R Markdown files.\r\nAfter installing, create a new template R markdown file by selecting\r\nFile –> New File –> New\r\nR Markdown which opens the “New R Markdown” pane\r\nIn the “New R Markdown” pane, select From Template\r\nand choose from the list of {alatheme} templates.\r\nFill in your information where necessary in the yaml\r\nheader and in the required code and text chunks.\r\nKniting to HTML safely\r\nIn R Studio, clicking the Knit button in the left\r\nupper menu above your script renders your R Markdown file into an HTML\r\nfile.\r\nHowever, if your code requires a config setting with\r\npersonal or sensitive information, . However, packages like\r\ngalah require users to enter an email address prior to\r\ndownloading data using\r\ngalah_config(email = \"your-email@email.com\"). Knitting\r\nwithout this information in the document will cause an error.\r\nA better workflow for rendering R Markdown files is by using a\r\nseparate script combined with the here package. Running the\r\nfollowing script outside of your R Markdown file will render your R\r\nMarkdown file in the file’s directory:\r\n\r\n\r\nlibrary(galah)\r\ngalah_config(email = \"your-email@email.com\")\r\n\r\nlibrary(here)\r\npath <- here(\"folder\", \"subfolder\", \"subfolder\", \"your-code.Rmd\")\r\nrmarkdown::render(path)\r\n\r\n\r\n\r\n\r\nPush to Github\r\n\r\nReproducible workflows\r\nSafe paths\r\nUsing setwd() to set a working directory can cause\r\nissues when setting folder paths, and can easily prevent a script or\r\nworkflow from being reproducible. \r\nOne alternative suggested by Jenny Bryan\r\nis to use here::here() from the here package to build your directory\r\npaths. here::here() automatically creates paths relative to\r\nthe top level directory.\r\nRead a full description of how to use the here()\r\nfunction to create safe paths in Chapter 3 of What They Forgot\r\nto Teach You About R\r\nFor example:\r\n\r\n\r\nlibrary(tidyverse)\r\nlibrary(here)\r\n\r\ndf <- read.delim(here(\"projects\", , \"style_guide\", \"data.csv\"))\r\np <- ggplot(df, aes(x, y)) + geom_point()\r\nggsave(here(\"plots\", \"foofy_scatterplot.png\"))\r\n\r\n\r\n\r\nWriting understandable code\r\nSummarise your code\r\nPeople who read your code should be able to understand what it does\r\nand why you made the choices you did. It is helpful to include brief\r\nsummaries of what your code is meant to do, and the logic of any\r\ndecision you might make after seeing a result or output. For example, if\r\nyour code used a statistical model, include your basic interpretation\r\nand subsequent decision process after seeing the results. \r\nUse sentences, paragraphs and code chunks to add structure to your\r\nscript. Try to create a flowing, logical story of how to run an analysis\r\nor make a data visualisation. \r\nCode chunks should be brief. They should be supplemented with notes,\r\ndescriptions or visual output that provides context to what a given\r\nchunk of code is meant to do. Users should be able to follow each\r\ntransformation that is made to your data, each output that results from\r\na transformation, and any final output. \r\nDo not expect your reader to know what they should see in your\r\noutput. If there is something in your code output that you wish to\r\nhighlight, it is best to tell your reader what they should see in your\r\noutput and how you have interpreted it.\r\nThere is no single correct code chunk size; you must use your best\r\njudgement. But if it seems that the result of several lines of code is\r\nunclear, you may need to split the code chunks to make the results\r\neasier to follow. \r\nStyle code using\r\nstyler\r\nUse the styler package to ensure your code is formatted\r\ncorrectly and/or consistently, The styler package formats\r\nyour code according to the tidyverse style guide (or another custome\r\nstyle) prior to uploading a finished file document. \r\nSee the styler github and tidyverse\r\npages for more information \r\nInstall the styler package to add style buttons to your\r\nR Markdown Addins dropdown menu. Clicking Addins ->\r\nstyle active file will reformat the code in your active\r\nfile. Clicking ** Addins** -> style active section\r\nwill reformat code in your current section. \r\nAlternatively, you style_file() or\r\nstyle_text() can be run in the console. \r\nStyling your document\r\nSee the R\r\nMarkdown cheat sheet for more information on R Markdown formatting\r\n\r\nFor tips to make your R Markdown output pretty, see Pimp my RMD by Yan\r\nHoltz \r\nTables\r\nSeveral packages can create tables with nice formatting.\r\nOne example is kableExtra:\r\n\r\n\r\nlibrary(tidyverse)\r\nlibrary(kableExtra)\r\nkable(\r\n head(mtcars, n=5)) %>% kable_styling()\r\n\r\n\r\n\r\n\r\n\r\nmpg\r\n\r\n\r\ncyl\r\n\r\n\r\ndisp\r\n\r\n\r\nhp\r\n\r\n\r\ndrat\r\n\r\n\r\nwt\r\n\r\n\r\nqsec\r\n\r\n\r\nvs\r\n\r\n\r\nam\r\n\r\n\r\ngear\r\n\r\n\r\ncarb\r\n\r\n\r\nMazda RX4\r\n\r\n\r\n21.0\r\n\r\n\r\n6\r\n\r\n\r\n160\r\n\r\n\r\n110\r\n\r\n\r\n3.90\r\n\r\n\r\n2.620\r\n\r\n\r\n16.46\r\n\r\n\r\n0\r\n\r\n\r\n1\r\n\r\n\r\n4\r\n\r\n\r\n4\r\n\r\n\r\nMazda RX4 Wag\r\n\r\n\r\n21.0\r\n\r\n\r\n6\r\n\r\n\r\n160\r\n\r\n\r\n110\r\n\r\n\r\n3.90\r\n\r\n\r\n2.875\r\n\r\n\r\n17.02\r\n\r\n\r\n0\r\n\r\n\r\n1\r\n\r\n\r\n4\r\n\r\n\r\n4\r\n\r\n\r\nDatsun 710\r\n\r\n\r\n22.8\r\n\r\n\r\n4\r\n\r\n\r\n108\r\n\r\n\r\n93\r\n\r\n\r\n3.85\r\n\r\n\r\n2.320\r\n\r\n\r\n18.61\r\n\r\n\r\n1\r\n\r\n\r\n1\r\n\r\n\r\n4\r\n\r\n\r\n1\r\n\r\n\r\nHornet 4 Drive\r\n\r\n\r\n21.4\r\n\r\n\r\n6\r\n\r\n\r\n258\r\n\r\n\r\n110\r\n\r\n\r\n3.08\r\n\r\n\r\n3.215\r\n\r\n\r\n19.44\r\n\r\n\r\n1\r\n\r\n\r\n0\r\n\r\n\r\n3\r\n\r\n\r\n1\r\n\r\n\r\nHornet Sportabout\r\n\r\n\r\n18.7\r\n\r\n\r\n8\r\n\r\n\r\n360\r\n\r\n\r\n175\r\n\r\n\r\n3.15\r\n\r\n\r\n3.440\r\n\r\n\r\n17.02\r\n\r\n\r\n0\r\n\r\n\r\n0\r\n\r\n\r\n3\r\n\r\n\r\n2\r\n\r\n\r\nkableExtra is also able to be used using\r\nmagittr or base piping:\r\n\r\n\r\nmtcars %>% \r\n group_by(gear) %>% \r\n dplyr::summarise(cyl = mean(cyl),\r\n disp = mean(disp)) %>%\r\n kbl() %>%\r\n kable_styling()\r\n\r\n\r\n\r\ngear\r\n\r\n\r\ncyl\r\n\r\n\r\ndisp\r\n\r\n\r\n3\r\n\r\n\r\n7.466667\r\n\r\n\r\n326.3000\r\n\r\n\r\n4\r\n\r\n\r\n4.666667\r\n\r\n\r\n123.0167\r\n\r\n\r\n5\r\n\r\n\r\n6.000000\r\n\r\n\r\n202.4800\r\n\r\n\r\n\r\n\r\nhead(mtcars |>\r\n subset(select = c(\"cyl\", \"disp\", \"gear\")), n = 5) |>\r\n kbl() |>\r\n kable_styling()\r\n\r\n\r\n\r\n\r\n\r\ncyl\r\n\r\n\r\ndisp\r\n\r\n\r\ngear\r\n\r\n\r\nMazda RX4\r\n\r\n\r\n6\r\n\r\n\r\n160\r\n\r\n\r\n4\r\n\r\n\r\nMazda RX4 Wag\r\n\r\n\r\n6\r\n\r\n\r\n160\r\n\r\n\r\n4\r\n\r\n\r\nDatsun 710\r\n\r\n\r\n4\r\n\r\n\r\n108\r\n\r\n\r\n4\r\n\r\n\r\nHornet 4 Drive\r\n\r\n\r\n6\r\n\r\n\r\n258\r\n\r\n\r\n3\r\n\r\n\r\nHornet Sportabout\r\n\r\n\r\n8\r\n\r\n\r\n360\r\n\r\n\r\n3\r\n\r\n\r\nFigures\r\nControlling the size of figures can help when you want people to\r\nfocus on trends of many plots or on details of a single plot. \r\nYou can control how many columns your plots created in a chunk are\r\nprinted into.\r\nAdd out.width = c(\"50%\", \"50%\"), fig.show = \"hold\":\r\n\r\n``{r out.width=c('50%', '50%'), fig.show='hold'}\r\nboxplot(1:10)\r\nplot(rnorm(10))\r\n`\r\n\r\n\r\n\r\n\r\nControl other parts of figure output in the chunk header as well. For\r\nexample:\r\n\r\n``{r, fig.align=\"center\", fig.width=6, fig.height=6, fig.cap=\"Figure: Here is a really important caption.\"}\r\n\r\n\r\n\r\nlibrary(tidyverse)\r\nmpg %>%\r\n ggplot( aes(x=reorder(class, hwy), y=hwy, fill=class)) + \r\n geom_boxplot() +\r\n xlab(\"class\") +\r\n theme(legend.position=\"none\")\r\n\r\n\r\n\r\n\r\nFigure 1: Figure: Here is a really important caption.\r\n\r\n\r\n\r\nModel output\r\nWant to show more readable model output? Use the\r\nmodel_parameters() function from the parameters package to\r\nmake your output clear and organised. The\r\nmodel_parameters() function can be considered as a\r\nlightweight alternative to broom::tidy():\r\n\r\n\r\nlibrary(parameters)\r\nmodel <- lm(Sepal.Width ~ Petal.Length * Species + Petal.Width, data = iris)\r\n\r\n# regular model parameters\r\nmodel_parameters(model)\r\n\r\n\r\nParameter | Coefficient | SE | 95% CI | t(143) | p\r\n-------------------------------------------------------------------------------------------\r\n(Intercept) | 2.89 | 0.36 | [ 2.18, 3.60] | 8.01 | < .001\r\nPetal Length | 0.26 | 0.25 | [-0.22, 0.75] | 1.07 | 0.287 \r\nSpecies [versicolor] | -1.66 | 0.53 | [-2.71, -0.62] | -3.14 | 0.002 \r\nSpecies [virginica] | -1.92 | 0.59 | [-3.08, -0.76] | -3.28 | 0.001 \r\nPetal Width | 0.62 | 0.14 | [ 0.34, 0.89] | 4.41 | < .001\r\nPetal Length * Species [versicolor] | -0.09 | 0.26 | [-0.61, 0.42] | -0.36 | 0.721 \r\nPetal Length * Species [virginica] | -0.13 | 0.26 | [-0.64, 0.38] | -0.50 | 0.618 \r\n\r\nmodel_parameters() can also be used for parameter\r\nstandardization:\r\n\r\n\r\n# standardized parameters\r\nmodel_parameters(model, standardize = \"refit\")\r\n\r\n\r\nParameter | Coefficient | SE | 95% CI | t(143) | p\r\n-------------------------------------------------------------------------------------------\r\n(Intercept) | 2.89 | 0.36 | [ 2.18, 3.60] | 8.01 | < .001\r\nPetal Length | 0.26 | 0.25 | [-0.22, 0.75] | 1.07 | 0.287 \r\nSpecies [versicolor] | -1.66 | 0.53 | [-2.71, -0.62] | -3.14 | 0.002 \r\nSpecies [virginica] | -1.92 | 0.59 | [-3.08, -0.76] | -3.28 | 0.001 \r\nPetal Width | 0.62 | 0.14 | [ 0.34, 0.89] | 4.41 | < .001\r\nPetal Length * Species [versicolor] | -0.09 | 0.26 | [-0.61, 0.42] | -0.36 | 0.721 \r\nPetal Length * Species [virginica] | -0.13 | 0.26 | [-0.64, 0.38] | -0.50 | 0.618 \r\n\r\n\r\nCode chunk options\r\nSee the R\r\nmarkdown documentation to view chunk options. \r\nExample: echo and\r\neval\r\nReaders should be able to identify where every file in a workflow\r\ncomes from (little is more frustrating than wondering where a required\r\ndata file is located). Code used to load or extract data (from\r\ngalah, for example) should be clearly identified. \r\nHowever, some code takes a very long time to run and you may have\r\nsaved it locally or in a repository to save time and/or space. In this\r\ncase, it is possible to show code without running it. \r\nAdd eval = FALSE to the chunk header to display the code\r\nbut prevent the chunk from running:\r\n\r\n``{r, eval = FALSE}\r\nala_counts(group_by = \"phylum\")\r\n`\r\n\r\nYou can then load a local file in the background, without showing the\r\ncode. \r\nAdd echo = FALSE to your chunk header to run the code\r\nbut prevent the chunk from displaying:\r\n\r\n``{r, echo = FALSE}\r\ndata <- readRDS(file = \"local_file.rds\")\r\n`\r\n\r\nCool Distill Things\r\nCode folding\r\nDoes a plot require a lot of code that distracts from the plot\r\nitself? Distill has a really nice code folding feature:\r\n\r\n\r\nShow code\r\n\r\nlibrary(galah)\r\nlibrary(data.table)\r\nlibrary(dtplyr) # provides data.table backend for dplyr\r\nlibrary(tidyverse)\r\nlibrary(lubridate) # for dates\r\nlibrary(ggpubr)\r\n\r\n# Might need to config email to get ALA data\r\nala_config(email = \"dax.kellie@csiro.au\")\r\n\r\n\r\n# Get records\r\noccurrences_Tas <- ala_occurrences(filters = select_filters(stateProvince = \"Tasmania\",\r\n year = seq(2018, 2020),\r\n basisOfRecord = \"HumanObservation\"))\r\noccurrences_Tas <- setDT(occurrences_Tas) # make data.table\r\n\r\n\r\n\r\n\r\n\r\nShow code\r\n\r\n# date format\r\noccurrences_Tas$eventDate <- as_date(occurrences_Tas$eventDate)\r\n\r\n# Wrangling\r\noccurrences_Tas <- occurrences_Tas %>%\r\n mutate(\r\n month = month(eventDate, label = TRUE), # add month col\r\n year = year(eventDate) # add year col\r\n ) %>%\r\n filter(!is.na(eventDate)) %>% # remove NAs\r\n as.data.table()\r\n# any(is.na(occurrences_Tas$day))\r\n\r\n# get month counts\r\nmonth_counts <- occurrences_Tas[, .N, by = c(\"month\")]\r\nmonth_counts$month <- ordered(factor(month_counts$month,\r\n levels = c(\r\n \"Jan\", \"Feb\", \"Mar\", \"Apr\", \"May\", \"Jun\",\r\n \"Jul\", \"Aug\", \"Sep\", \"Oct\", \"Nov\", \"Dec\"\r\n )\r\n))\r\n\r\nggdotchart(month_counts,\r\n x = \"month\", y = \"N\",\r\n color = \"month\", # Color by groups\r\n palette = \"viridis\", # Custom color palette\r\n sorting = \"none\", # Sort value in descending order\r\n add = \"segments\", # Add segments from y = 0 to dots\r\n legend = \"none\",\r\n rotate = FALSE,\r\n dot.size = 4,\r\n add.params = list(color = \"month\", size = 1),\r\n title = \"Records added to ALA by month (2018-2020)\",\r\n ylab = \"Number of records\",\r\n xlab = \"Month\",\r\n ggtheme = theme_pubr() # ggplot2 theme\r\n)\r\n\r\n\r\n\r\n\r\n\r\n\r\n", + "last_modified": "2022-04-26T13:06:11+10:00" } ], "collections": ["posts/posts.json", "people/people.json"] diff --git a/_site/site_libs/header-attrs-2.13/header-attrs.js b/_site/site_libs/header-attrs-2.11/header-attrs.js similarity index 100% rename from _site/site_libs/header-attrs-2.13/header-attrs.js rename to _site/site_libs/header-attrs-2.11/header-attrs.js diff --git a/_site/test_style-guide.html b/_site/test_style-guide.html index 2e0b1e52..a23ffefc 100644 --- a/_site/test_style-guide.html +++ b/_site/test_style-guide.html @@ -2817,7 +2817,7 @@

${suggestion.title}

color: var(--hover-color, white); } - + @@ -2906,7 +2906,8 @@

Contents
-

This document details the Atlas of Living Australia style guide for creating html files from R Markdown

+

This document details the Atlas of Living Australia style guide for +creating html files from R Markdown

Getting started

-

Create an ALA R Markdown file

-

To create new R Markdown file that follows the ALA style guide, install the alatheme package.

+

Create an ALA R Markdown +file

+

To create new R Markdown file that follows the ALA style guide, +install the alatheme +package.

devtools::install_github("AtlasOfLivingAustralia/alatheme")
-

The {alatheme} package contains ALA templates for R Markdown files. After installing, create a new template R markdown file by selecting File –> New File –> New R Markdown which opens the “New R Markdown” pane

-

In the “New R Markdown” pane, select From Template and choose from the list of {alatheme} templates.

-

Fill in your information where necessary in the yaml header and in the required code and text chunks.

+

The {alatheme} package contains ALA templates for R Markdown files. +After installing, create a new template R markdown file by selecting +File –> New File –> New +R Markdown which opens the “New R Markdown” pane

+

In the “New R Markdown” pane, select From Template +and choose from the list of {alatheme} templates.

+

Fill in your information where necessary in the yaml +header and in the required code and text chunks.

Kniting to HTML safely

-

In R Studio, clicking the Knit button in the left upper menu above your script renders your R Markdown file into an HTML file.

-

However, if your code requires a config setting with personal or sensitive information, . However, packages like galah require users to enter an email address prior to downloading data using galah_config(email = "your-email@email.com"). Knitting without this information in the document will cause an error.

-

A better workflow for rendering R Markdown files is by using a separate script combined with the here package. Running the following script outside of your R Markdown file will render your R Markdown file in the file’s directory:

+

In R Studio, clicking the Knit button in the left +upper menu above your script renders your R Markdown file into an HTML +file.

+

However, if your code requires a config setting with +personal or sensitive information, . However, packages like +galah require users to enter an email address prior to +downloading data using +galah_config(email = "your-email@email.com"). Knitting +without this information in the document will cause an error.

+

A better workflow for rendering R Markdown files is by using a +separate script combined with the here package. Running the +following script outside of your R Markdown file will render your R +Markdown file in the file’s directory:

library(galah)
@@ -2971,9 +2995,19 @@ 
Push to Github

 

 
Reproducible workflows

 
Safe paths

-
Using setwd() to set a working directory can cause issues when setting folder paths, and can easily prevent a script or workflow from being reproducible. 

-
One alternative suggested by Jenny Bryan is to use here::here() from the here package to build your directory paths. here::here() automatically creates paths relative to the top level directory.

-
Read a full description of how to use the here() function to create safe paths in Chapter 3 of What They Forgot to Teach You About R

+
Using setwd() to set a working directory can cause
+issues when setting folder paths, and can easily prevent a script or
+workflow from being reproducible. 

+
One alternative suggested by Jenny Bryan
+is to use here::here() from the here package to build your directory
+paths. here::here() automatically creates paths relative to
+the top level directory.

+
Read a full description of how to use the here()
+function to create safe paths in Chapter 3 of What They Forgot
+to Teach You About R

 
For example:

 

 

@@ -2988,19 +3022,53 @@ 
Safe paths

 

 
Writing understandable code

 
Summarise your code

-
People who read your code should be able to understand what it does and why you made the choices you did. It is helpful to include brief summaries of what your code is meant to do, and the logic of any decision you might make after seeing a result or output. For example, if your code used a statistical model, include your basic interpretation and subsequent decision process after seeing the results. 

-
Use sentences, paragraphs and code chunks to add structure to your script. Try to create a flowing, logical story of how to run an analysis or make a data visualisation. 

-
Code chunks should be brief. They should be supplemented with notes, descriptions or visual output that provides context to what a given chunk of code is meant to do. Users should be able to follow each transformation that is made to your data, each output that results from a transformation, and any final output. 

-
Do not expect your reader to know what they should see in your output. If there is something in your code output that you wish to highlight, it is best to tell your reader what they should see in your output and how you have interpreted it.

-
There is no single correct code chunk size; you must use your best judgement. But if it seems that the result of several lines of code is unclear, you may need to split the code chunks to make the results easier to follow. 

-
Style code using styler

-
Use the styler package to ensure your code is formatted correctly and/or consistently, The styler package formats your code according to the tidyverse style guide (or another custome style) prior to uploading a finished file document. 

-
See the styler github and tidyverse pages for more information 

-
Install the styler package to add style buttons to your R Markdown Addins dropdown menu. Clicking Addins -> style active file will reformat the code in your active file. Clicking ** Addins** -> style active section will reformat code in your current section. 

-
Alternatively, you style_file() or style_text() can be run in the console. 

+
People who read your code should be able to understand what it does
+and why you made the choices you did. It is helpful to include brief
+summaries of what your code is meant to do, and the logic of any
+decision you might make after seeing a result or output. For example, if
+your code used a statistical model, include your basic interpretation
+and subsequent decision process after seeing the results. 

+
Use sentences, paragraphs and code chunks to add structure to your
+script. Try to create a flowing, logical story of how to run an analysis
+or make a data visualisation. 

+
Code chunks should be brief. They should be supplemented with notes,
+descriptions or visual output that provides context to what a given
+chunk of code is meant to do. Users should be able to follow each
+transformation that is made to your data, each output that results from
+a transformation, and any final output. 

+
Do not expect your reader to know what they should see in your
+output. If there is something in your code output that you wish to
+highlight, it is best to tell your reader what they should see in your
+output and how you have interpreted it.

+
There is no single correct code chunk size; you must use your best
+judgement. But if it seems that the result of several lines of code is
+unclear, you may need to split the code chunks to make the results
+easier to follow. 

+
Style code using
+styler

+
Use the styler package to ensure your code is formatted
+correctly and/or consistently, The styler package formats
+your code according to the tidyverse style guide (or another custome
+style) prior to uploading a finished file document. 

+
See the styler github and tidyverse
+pages for more information 

+
Install the styler package to add style buttons to your
+R Markdown Addins dropdown menu. Clicking Addins ->
+style active file will reformat the code in your active
+file. Clicking ** Addins** -> style active section
+will reformat code in your current section. 

+
Alternatively, you style_file() or
+style_text() can be run in the console. 

 
Styling your document

-
See the R Markdown cheat sheet for more information on R Markdown formatting 

-
For tips to make your R Markdown output pretty, see Pimp my RMD by Yan Holtz 

+
See the R
+Markdown cheat sheet for more information on R Markdown formatting
+

+
For tips to make your R Markdown output pretty, see Pimp my RMD by Yan
+Holtz 

 
Tables

 
Several packages can create tables with nice formatting.

 
One example is kableExtra:

@@ -3246,7 +3314,8 @@ 
Tables

 
 
 

-
kableExtra is also able to be used using magittr or base piping:

+
kableExtra is also able to be used using
+magittr or base piping:

 

 

 
mtcars %>% 
@@ -3407,8 +3476,10 @@ 
Tables

 
 

 
Figures

-
Controlling the size of figures can help when you want people to focus on trends of many plots or on details of a single plot. 

-
You can control how many columns your plots created in a chunk are printed into.

+
Controlling the size of figures can help when you want people to
+focus on trends of many plots or on details of a single plot. 

+
You can control how many columns your plots created in a chunk are
+printed into.

 Add out.width = c("50%", "50%"), fig.show = "hold":

 

 
``{r out.width=c('50%', '50%'), fig.show='hold'}
@@ -3419,7 +3490,8 @@ 
Figures

 

 

 

-
Control other parts of figure output in the chunk header as well. For example:

+
Control other parts of figure output in the chunk header as well. For
+example:

 

 
``{r, fig.align="center", fig.width=6, fig.height=6, fig.cap="Figure: Here is a really important caption."}

 

@@ -3441,7 +3513,12 @@ 
Figures

 

 

 
Model output

-
Want to show more readable model output? Use the model_parameters() function from the parameters package to make your output clear and organised. The model_parameters() function can be considered as a lightweight alternative to broom::tidy():

+
Want to show more readable model output? Use the
+model_parameters() function from the parameters package to
+make your output clear and organised. The
+model_parameters() function can be considered as a
+lightweight alternative to broom::tidy():

 

 

 
library(parameters)
@@ -3461,7 +3538,8 @@ 
Model output

 Petal Length * Species [versicolor] |       -0.09 | 0.26 | [-0.61,  0.42] |  -0.36 | 0.721 
 Petal Length * Species [virginica]  |       -0.13 | 0.26 | [-0.64,  0.38] |  -0.50 | 0.618 

 

-
model_parameters() can also be used for parameter standardization:

+
model_parameters() can also be used for parameter
+standardization:

 

 

 
# standardized parameters
@@ -3470,28 +3548,38 @@ 
Model output

 

 
Parameter                           | Coefficient |   SE |         95% CI | t(143) |      p
 -------------------------------------------------------------------------------------------
-(Intercept)                         |        3.59 | 1.30 | [ 1.01,  6.17] |   2.75 | 0.007 
-Petal Length                        |        1.07 | 1.00 | [-0.91,  3.04] |   1.07 | 0.287 
-Species [versicolor]                |       -4.62 | 1.31 | [-7.21, -2.03] |  -3.53 | < .001
-Species [virginica]                 |       -5.51 | 1.38 | [-8.23, -2.79] |  -4.00 | < .001
-Petal Width                         |        1.08 | 0.24 | [ 0.59,  1.56] |   4.41 | < .001
-Petal Length * Species [versicolor] |       -0.38 | 1.06 | [-2.48,  1.72] |  -0.36 | 0.721 
-Petal Length * Species [virginica]  |       -0.52 | 1.04 | [-2.58,  1.54] |  -0.50 | 0.618 

+(Intercept)                         |        2.89 | 0.36 | [ 2.18,  3.60] |   8.01 | < .001
+Petal Length                        |        0.26 | 0.25 | [-0.22,  0.75] |   1.07 | 0.287 
+Species [versicolor]                |       -1.66 | 0.53 | [-2.71, -0.62] |  -3.14 | 0.002 
+Species [virginica]                 |       -1.92 | 0.59 | [-3.08, -0.76] |  -3.28 | 0.001 
+Petal Width                         |        0.62 | 0.14 | [ 0.34,  0.89] |   4.41 | < .001
+Petal Length * Species [versicolor] |       -0.09 | 0.26 | [-0.61,  0.42] |  -0.36 | 0.721 
+Petal Length * Species [virginica]  |       -0.13 | 0.26 | [-0.64,  0.38] |  -0.50 | 0.618


Code chunk options

-

See the R markdown documentation to view chunk options.

-

Example: echo and eval

-

Readers should be able to identify where every file in a workflow comes from (little is more frustrating than wondering where a required data file is located). Code used to load or extract data (from galah, for example) should be clearly identified.

-

However, some code takes a very long time to run and you may have saved it locally or in a repository to save time and/or space. In this case, it is possible to show code without running it.

-

Add eval = FALSE to the chunk header to display the code but prevent the chunk from running:

+

See the R +markdown documentation to view chunk options.

+

Example: echo and +eval

+

Readers should be able to identify where every file in a workflow +comes from (little is more frustrating than wondering where a required +data file is located). Code used to load or extract data (from +galah, for example) should be clearly identified.

+

However, some code takes a very long time to run and you may have +saved it locally or in a repository to save time and/or space. In this +case, it is possible to show code without running it.

+

Add eval = FALSE to the chunk header to display the code +but prevent the chunk from running:

``{r, eval = FALSE}
 ala_counts(group_by = "phylum")
 `
-

You can then load a local file in the background, without showing the code.

-

Add echo = FALSE to your chunk header to run the code but prevent the chunk from displaying:

+

You can then load a local file in the background, without showing the +code.

+

Add echo = FALSE to your chunk header to run the code +but prevent the chunk from displaying:

``{r, echo = FALSE}
 data <- readRDS(file = "local_file.rds")
@@ -3499,7 +3587,8 @@ 
Example: echo and eval<

Cool Distill Things

Code folding

-

Does a plot require a lot of code that distracts from the plot itself? Distill has a really nice code folding feature:

+

Does a plot require a lot of code that distracts from the plot +itself? Distill has a really nice code folding feature:

@@ -3536,12 +3625,12 @@

Code folding

occurrences_Tas$eventDate <- as_date(occurrences_Tas$eventDate) # Wrangling -occurrences_Tas <- occurrences_Tas %>% +occurrences_Tas <- occurrences_Tas %>% mutate( month = month(eventDate, label = TRUE), # add month col year = year(eventDate) # add year col - ) %>% - filter(!is.na(eventDate)) %>% # remove NAs + ) %>% + filter(!is.na(eventDate)) %>% # remove NAs as.data.table() # any(is.na(occurrences_Tas$day)) @@ -3554,7 +3643,7 @@

Code folding

) )) -ggdotchart(month_counts, +ggdotchart(month_counts, x = "month", y = "N", color = "month", # Color by groups palette = "viridis", # Custom color palette @@ -3567,13 +3656,14 @@

Code folding

title = "Records added to ALA by month (2018-2020)", ylab = "Number of records", xlab = "Month", - ggtheme = theme_pubr() # ggplot2 theme + ggtheme = theme_pubr() # ggplot2 theme )
-
+
@@ -3597,13 +3687,13 @@

Code folding

.copyright-container { display: flex; width: 100%; - margin-bottom: 30px; + margin-bottom: 30px; margin-top: 30px; } .copyright-row { width: 100%; - height: 60px; + height: 66px; position: relative; margin-bottom: -1.75vw; background-color: #c44d34; @@ -3661,14 +3751,14 @@

Code folding

/*-- Logos and acknowledgement --*/ -.footer-container { +.footer-container { display: flex; width: 100%; max-width: 1140px; padding-left: 15px; padding-right: 15px; padding-top: 50px; - padding-bottom: 50px; + padding-bottom: 50px; margin: auto; } @@ -3751,7 +3841,7 @@

Code folding

margin: 4px 2px; transition-duration: 0.3s; cursor: pointer; - background-color: white; + background-color: white; border: 2px solid #E06E53; border-radius: 10px; } @@ -3774,7 +3864,7 @@

Code folding

margin-top: 20px; margin-bottom: 20px; } - + .footer-container { width: 100%; max-width: 100%; @@ -3794,15 +3884,15 @@

Code folding

width: 20em; margin: auto; } - + .copyright-column-left { width: 50%; } - + .copyright-column-right { width: 50%; } - + .copyright-row { height: 100px; margin-bottom: -3.75vw; @@ -3831,32 +3921,32 @@

Code folding

- +
-

The ALA is made possible by contributions from its partners, is supported - by - NCRIS, is hosted by +

The ALA is made possible by contributions from its partners, is supported + by + NCRIS, is hosted by CSIRO, and is the Australian node of GBIF.

- NCRIS logo - CSIRO logo - GBIF logo @@ -3864,10 +3954,10 @@

Code folding

diff --git a/_site/test_style-guide_files/figure-html5/unnamed-chunk-10-1.png b/_site/test_style-guide_files/figure-html5/unnamed-chunk-10-1.png index f63344ea22c6f7a3e65a2cac071d157f5b7d94d9..17c1b692de7afb53602bdfe28d156e4466fa0e2d 100644 GIT binary patch literal 18568 zcmeHP2~<RUGLquZf~z=9kR}z_x}E2|NB4x z;ew;X9`$9q%OD6+|Kwx4FCb_!3He)!h9KzPsVW}`LWMZ)-M zL0&Y7#)W>nVrFIr3Xp8%^0Kl@vhwn>@&dm!tE6^&gWOThUn(Er|y1_M~?B_IEJV&z|b%M@{fU8c}PgHGR}RnC1o1ysT(R z;P#>=DH_ng)lO>%H%x=!$Ei?Ff_+VbUkw%92{oW)4H&JjrVp-x;r1k}cA8f^EvX%} zK^lPzZinGMs$U;9p|1wq)V>-}o!Zw2_rWli%bh;uD?P#n1H&~5pgr79gF%}0WmFS!eVf5ZRdLxB*aAB+5<68i;TLy!^liQR|$L(|4Or7>xPhsc6$ zvH6$r*S0LXI)q_90=rVfs+{Z1?*qmVT)8>KAF17w4Pw}V?WY%VpxXUaZAIzNTXV#`_io2)sg~vC(Q7SqP%*@ za*(n8PiNAiQIf*C@N|r!!&$VzZinOBmr$P?{9T?KY8UmT1*#52;0yHsE^GS#r4P%K z7Kov-?LA39#IEed+pJJGPhl;4zHdIQrbvH2pI)v=fB7oyNm65NIMxe zXS=HTfqs!YR%kq4*?`ZNPk%j)mDvtf%?_&ps&L|VFy^atI;(vOqYSAeza zWhJeDA8%q{pNx`B)(=N|L^Gie&(H4mpMOEh*z1G}w<&{pO?g~B9m<3zpo~9d=Z!Q7 z8&!yO883>mjhg5ZscRhr@SjDmdh_;TpD9#Jt_eucE>QKEms4N&HbKHu34PzQcpAKkr_YKYcb4 z6~O2`bqT^|5X%QYBAudZak%=0ueOE#9|uVx?4ih+7-+s$n)w@>3oEY}FhafI9Je8l7eiZFxSo zA`MoP8Btf>aVA~a?20~x>P=uuCL}y5PG#uC-10WyRoU_Ly!+_%vI}aC;|e~0{rnn- z*F5mA;DOK^wI|#(w$Cf~M;F#zc3<*J+1~(ZzXjMA`!A5cB!vCDUaY6R( zMXk3E?Ou0$^xwrx|Y7;DB(@57ObQYM}RKEh4?R)s)&>E~j5rI1DOC7F4J~TnPoW zeH~}&)lnmI7qJ1$S9vX_w^A5!{zeB$fZ-QP$c-g6CKZVj(zH3Rg4Sky^gBK4 z7s!NQQD5TG?**@_B%w_44GNp?EE?E%Zz<2~ud`oc@=sz26#DsR^KJ?a;F@p@QhF0q zYM|Xp#(tgszrds^tf<xPwyXnNt1ZU11u0eRucZ=|%N*9`y8yuau3^yG-h zg)04*iGV~Vbn)f9f(Ul!33Mnc=J(+&Mjh66;nDH+B<8$nuaWpCBEd|S_TA!!=|?fP zDJ{sX-sPhFnc?ua7i4j{Q=D{vphpzi+LPqG`4+Qi&RwZPxw?EfFlWxMu9DTiQ<8RE z>0rHt`nD*^c2Jkn_m7+UeMGXA4FrYrEo`031wCW2jl*2NuF(^4JgtEqxPD9yT)Ak{ zv$%!Iujfop)>mTYPx!NAMSvr7R4Cut$%$ER8MTGT6->%YoR|joe4;054RUrxKMw%Q zDT^W*?Bg6fC*D&0KbBKE_!tVu^u?0MvV3L*l#|1Yn zwAFy;V6w~5RmUjA*c$<{z2RE?lA%gE{9|lm1ife#rPJiGC|osX&KJ}ddw7%fv%!ul zP`#*IcI;kqsj4T5VFQdwQ@o-eA98bZK$Z#IAoH1}(5BR0r-uq7QCZJLb*S>@SAnZh z#d5hJU300~tJ*HAK2r|Nm~L2_&Qd7Hc`zm?yC*S*Ze^j~Rs_cA%=I~+fs!Z$LkhL7 ze_sX1CGy>=r{wdvg2UdxfuiJ&H&DglQ`v{2b6{gqMj4jN(f@%1|GG^bof`VC93OwN=y7=5tSOt7Idsb;=W4RG3I1NJ(%hA( z40)vag>1va{Xu4UW8`HQFfz@b&Kkqun4s%5D~~{g)sHV@8N=hB5q(K*$a@S$)^gTy zZEc)bNZ*QAN4!aBPty5qONPXT?oUNu(j($k9(34N=RT)y*sg~hK#tg>QpaL)H|tL3 z0WaJXVsRsbH0x4Y<=?haOynu3HZx#8n*v{F%L-I#)cPqWj}+sfA9Gc0-|GyJ>e0W* z&1sn(Mt>EhHCk52G_iflTGC0m+CDh_V|Z}odBqAwvmh+|PATjF@zK3Nee zycP(_+=7V4lN;JSqT#-69CY~bT){<@q=(FtJ)6K;j(;ZDRKOH$5RQDOu2|Z^FG;4& zzh4ak8nL`n^M)?POk=K+ZP=!%PNvGzEwf`~qlya!-}o$8Q0k(fVBy1Y<1L&0B2C+% zF||2B*8prvB@RO`;I(_|#sX!ZmK7hZn8n(F(wYHhR*H5RiZ{in4Beg0KDD1Tfcrs^ zifXf-!+$GE@}eNw@a#-pjds6_AGz0Qq-hlfu@G7JnxUC#RW|OTuI!70f9U@$${_3| zuE_FGgH3i~5Wz*L36{JX_r1irKi*nfb{1 z_oG$;b@q@zj25R*a9i;T4>iJY`YR|@aRDldNwDj{&CUKKs#sz;L4LywA2G+R)@$+b zJ;{-KPUSp{wF&dgw}vuE#U+b-GNX;B4-Nfca_{5oOe#EFr!U6EBBro{=`~mz=>FIE zi%jv}N;qgxl#m9;+*P z?V)0ZSDhbmIZKL&*>YlJ8BTy%Q9WO$q$pVbbW{m;`b(T7I{hp3n&{s;9js7w=7Yk5(tAE{& zet*qNh>WT^j;vehdfG>k!kFPxuI&w7at3Q>h%rLvZ$)1ztX+K2sCk#U2OCNc+GONT zas)L$?bsSJP+JuqH(&;M4GU@=<2}86QK(h#4Y)O2W&O$Ojuc}I_w7_B&_Y5r$ALD& z(?<0vt*8^Ayb6rz%&%LSFr_8_>RzB~P41a~b8Fd6 zFWgue6MAEpo`M)Lm$Q$PMH9r8ZIG|wEPxE$stRaKPm%bVfzy5pjm<*)9&a(xwgc=zd(8Ia&VkSaR8{dw zu;!A>OR^rFXFn-=FSGRm`ndq$8`TB{mnQk`AuvQFzr2;QYRRpfcniY1oe3H_3BZoyEyk@3l_@yg z+N#iMEs(b^lvTav{sNCjpJ2fTuXHFhiyxecDC@;{tJ0K}fv8AP27uU5R1 z8<5Bg0Nr)Lt3yxH7i{6NCFtkhD1x7!xG|Yut)Vzn%)YxTj&5p)f^`)2e(*-*s=eX9 zGWGzGPXH9=Ke;eOp7^W)A>*-Gz(!=p!crH26exdq)|cHdHq-)MDI+A~5zQw)VfX7^ zU68p0{X7(qIYC|V68?f2-u@W&8$4R7pL z!2PYP>m-8r4Q|%fIk@#do7k%L80qU=M!vP4(SY&wBYH-0B1EXL--KA=f!n{)#zu=T zOhB&+f5rn}YFHUp|NLz+E*sD6SW6**Ob3M=#a#$+~_BZMR)CB_}yIg^W#^Z=* zfXViyf-R8O{P1+>J|NyhpCJ2Vr`3Pzpun<2{7U@IrubtDZ)-_#sOAb_{|b0(mGLlz zF0U*7%OKdPI#fZh&lI^6L&Q~zupKsG4P_#6j%;gHhth|Sipp`s-bM>;LZldrUaZ4^ z`w?(Kbh$0oVD(RSh2-R9V4Xi2M4WmP7CgkI*Ne47JXIy}g-D={q*(?tyM!q-aHgiV z4hRA2m+n#-k}aqN@|t2@5;F;|ZrBvtgQWsb`FYgZ@8>4bqvBJxn`B_Z(NB$Xo%#*h zS``t8IL|_~1?c?gENPlK$F;5Mf4HCL|8DBCZQfKOt#*)f6!e+6X{_oE55qXiQ_UO1VKJ&BT|yT zNk~Q(qfE~LY|&S=DAgET$jK29>QmMccEd^(5DIquhB5S{=fq&SyG?3&_!IJyiiKs_ z=zNJlBpx77g`H|6pRfLXS&(JtGFDSwfj3d83z)f@{5(Cv(gwsVDZEv4%TyR+!My&Lvpc6v+T2PMFR|v}L3H2TTK?68_%K_IUFmgM zN)&V`f}1$Zl2_{y+=My9UL=O4VyPE_DK9}BIUxFhHQk_+#RNQsVDP%&&qA_vrBMg@ zu^%ZMiF#XhVLi&qm+0)ySIJ^4pp08dAqG6sp#&nK5eUkKHINW-zN~svhjQ`~Y6NjX zB5{ws;01s;&=`m{T(x@vCF0c_$zcCLz6#LQ)}hz}*C%4enS<>l3h_39m~0Q(o{tDtE^PA9QpZ7lfEwZ>nZ<9& zOpC(fcIf_xi_YwtXP59;uW`TuJy86 z{nNlU+pH9Hp|b_vUU|$om>YbQ7^*0WTWQqh^zckfSqhe~RFngNVFC3S0+9yUb)Jz^ftirITiNMBFM41GG37$t45 z*r0>{EjxE0$!?}d@6PkpD`eKAmT8Cr1~2>#RW5J^{RKCVpkJY{T0H~&P6vjZ3K_#1 zbrIi>)co5r9ypu)D?|pdz-IOO>k#ADIQ#{8fOm?tE-%<}ht0gr9h8X%pNF~gZ9cy~ zYv2xTR`oHxUe*U7#>D`2kEY_^W4@O@sqv2l7rU1Qv=)yAP_YJ@qg}kd!eCzP~ z`R|Rx4dC`?uDcw#{FE}xZ@-f#$q#t}jF7*0Ejp+j@sfaVTbfqsP}-G|#92AOiFQg6 z^3jbC-mNl3Q+`MC%Lda|AC$UXLKTlGb@9k|JZDl`Ao;qE(O^fb7kKo(20my>_f@H} z459roa}>g+l%IT|{e(ZN*$~S~b>NrD{iO7lJ@n|CIrm@k> zxYSxY9aY?GE!s872)ZA^A;>NK%qb$*mj)YfDS;C{NlkKFi3615O8S@#9*OywtAjt{nTM8 z-@}13-B2-19W$&NYpvN86ueEU&x`UFhqwAO;DL=~jN!+AMTzZogo~EL6&~)KjEYVh zxacE zUmNEC;Om1k$uA^sXB1Hoj8zOzKk{ba1SQ2b=^QfI6+R%GtB1w z+->-@1B7HbaKe4On9nS{_yjFJAwL_n( z`K>{&yIY;qQvoK!rS!Q9w^`cibo@jv*4?#xROKhQJeeEoBd+_5A*}X{Iex^o+#+AB z>E$P|_QjgBH6MSMccc)o!F_?qBvJf9+v$t8c7%;fG*6`y?!uLroL<^sb4%%MFelXt zeq>{rIM)|@=9;%2h7<_POWslOIr(!u9EUqpo@^l-l+?Z~I4JZ!< zp5oJwwT$?yAxMRc{9gc4*c9-&@vX@_>e~A&`n~sVH%RR_R&6E|+RR<6PxkW3Yw>l) z0|8nW1s!HDy@g31O zEHiBr-f6<*9s5DUEdYv8m+U^hJ}5 z0b~Ch!_;hV-**47-ih7DOIghkcnqVdB6#LPr9N8_F`V5nZNSQaPGHZJb)l^)!2pk; zO&LX%#9!T~EGfyvRVSuH9P+er9n0QYN>QRzp*GEL|W(%p2?SMR$t{7UhL9sFPI$YEERs0bq#e8 zKHi@AVnF9%?wRy4g}k+M8h@4K#AwEP94x2XL~LBmxpz;JaBefFYy5}$=m{y`J-htN zqmB1udY<0+KvMHh{npG0K7Z&sL&bghq>@r6Rx~WVw(o0#A#AYsrvkQ2OZ|cp(Ilh zmu&^*D{1z&ec;VZb zyu&dQACYJsd;I#E4|ZjU4ipSdGmpA;PQUw7Ifc6h^;3u*i$l_m@jF67V?U9G`KBrPIT7 z%YuiU7j=x1ZP9(s-XvFvTGH`!RKTzqJ||eE<~9>5YJOE%CGU z9%a)hxo~|IE?=H`iW9U7bnvK)evtQBe8B+_2^C8P6@JE%;c*XMRO63nsDJ{_T^w%4 zJN|{NLstJ4$%`(z-3r68xvs8}et>s~(Q-{UEPOM$eE3@iROG_H%s#*$j}-M75Af?Z zqRpjEOkeq!D47^2FBl}=pSVM{MECiX(Mt-aZ#Q|Ajz>7LQO;%edC#P5f=R(#u_&Y~ zwsq!%mhq^xLwDT8J>(qX$w7OkK$V)#aEq9{Xlb>fOhAb0KB|+|KX{8c&Lx(~T&xMj zKbYZp&6wpZkLYg6a<@8Bl6-_pHHp|nA1?=8??6UQRvi+D3rVUfj_XOgi9kj@~w7#g;ASV;;7JH8!OJFuK6Y1R^Z1^w_w_Jig(V z4dl}8nye2ToSf>6#!ABi@{T=4Fpx+|IbZ&|k|Hvp(becOYcR7^FtaIXNph6&^ywtzW}*V)w}=z literal 54513 zcmeFa2T)XL*Dl($pafA=f`SBP1OyR~q{KFXfCPyO0wR)=NY1nYlxP4^kf21#86+zL z3X*eH5Xm{`yEZdAE&ctc>YjV+{B_RptD33)X0y}#uK28Ht^G(rPLiCImK23TkxN~= zpol^dx+DJ(??j8yZ~t19 z(Qz@#vlQ>+zCQc9sqvomh;hQWH|Lv7g*Sxv*p}*6Bwshskc__L&B%Y2Yf@O%wfy7N zr0Zah)kF{3S03%tLv2u%rE9tqo?ZP!(t2}=sbOmG?)3ANUpiPrZzdt$=7e`cR*tk_4jw_!o& z*0nivD9ewDdT71cZT*ScW^I1ORxR`;ao^K3G-R|s$GB9prq6wHQ_EtFqR`H^cJuG+ z`Xt-=tdbzSIGlDs;Thwz@#|tAsb2Vg#=O7pFh<3FRhR8PFPYT7T+j3cXXi`9XWIHC201 zP<{8tF4rAyqn)Zb{G)tdR7qX(EHpk!&Wg(25{u?kG1bY4d~=Oobp_2>g>rdkpHFPC zQ)BJ>;%D)vCZ89dGP1b!_k0)oR$~=@C)~2v{gdX`lXtZ5Xj?K`g-aSYIZX1IOr%7n zk^V3Wy3!SXPQ8s&gzeS?(GR5`Zpp|GQF+`a(cyBm6-+qbOQV^;vm({vR)R~SJI!0G zG4sTXmT%(963=(CI>+=+^6ro`@4wG1C4T7W1G9X``{VO5wyrXw9Vysyt7%gaqMOdz z&U|8&;va6`JJjo4bY*P1>CSL^)t){dqtMK(;1RFV&)@rRlBS$ur`|nK62Uf_HolDX(RxpV|90aYx;%VU>WpL4V;SDp4%P zenHWm_2bQ6^y%h%rSH|`X=@W0N9Q%{&5!k<+MEluHdogjU1oPQ^`dekHg{|uvnMOl zZK(nSaa!-Hl)kJiiXDCvqX;~WP=xS{0REwce^4lN1p4QHA#{(}@$YZs;Hq1OCJKc? zNnJRrY)>%OO%gz_5-L8W%^E6t>j=dM7H;=FOgy_E5}*7?bmameRfXP8`f}fWqDOb> zQ~IDib(EhIFdvrmw8UK0t7qPi+5Pb5UpsPBFYWQ#^B6^tes;lnX0d3qAm)HpaY3tQ zbBb1>yhDeprp;(eVYqgE_p)tae6p`Ka;vyX%*VlP-j(};Tj!i!j$YOCvXSP5knwY? zNyKBh{eBuo^_%)?sCf%rvh5oeL5W63Ca>hY1_iP$FGe8bH2$S)!N=SRRf3`*JSR4^glnJ6*e@QEy}SK!>x$v#Zgd)W?-xx$feL)%!*cegg8Y1lA#Dt*so0kj z@BI{siQT2INIb(+pgArk)JShRHQxIjC(M)K@kYzPNl4)E*vqK(bzk;i-BS}AsiF`8 zvv}8GUMvC0(xf%Ul)ktzRi}27zHD*TPn>?|?z%-0=@b`nTJC82P5F{D-<1zO#PrUV z*JfggJ6K7UUaF4GMvQW8-d^AuJZzF_+#G*U*w)lfvW_2b!HVGZi)tfsPHGp4fM{|L zxmu^bD_NTE*KLZs5*S$ZoYOljEUcPt$qTas-D#e07FB#ZdX_a*k{Vr2YHep%@$u0C zj{Kx+87!*F*V|2LIu69Gy*B*vf`{r66^H7_mHE-P6o%?I8lt2E6Dk^dZG%o+Yna=x zWqO?4^+Lqq4$;l-Je%pUjM{TY>M~1)>LYc#axDYb!<;7GUCp;OYjkOQT~rcqTxmow zE~8naWaR6uG^74<-zP;_^>nIB$<4>s-<-IWV7w%vK(HCQN4dsT1xxvtkbgX_m>{#l z`aOvb)^!mp2L8xcGnI(Ri|aM7M^(Gqc+ zz58MJXAGV&5T;X7p!D9p<-*%S0uXAV6JAH~tl*zB{`X(2p_+7Fm)~|4<3y5p{oSbv zu$oJa;Tx0*s!3jNm14^Yb{(czoyfOvlKp6?!JcYZ|K^3Dg;rFprlw}vY>jY*jwik9 znQ(O>?p>2q6sUPcdt9@Et=a!V*v|aAb^XP(n`K@!m)E}D+v`{B{DYUrs3D5RG~+Ht z&XUAIfgEnBV`H1Bq6X!}tc&d9jARMFpV5N~I^JE{w`My)` z(X=e1nvUjjc^knyl&C}EE|k!tFN1r{s|EI!*X_SO&H=j(4EVq#O^cm&BD0mhm`epw@qsqSO_}WR$0MrO0YF1aY+)P z$EFG=T~fZ02sUPaljAp)26IapRE}$I4c)N4Yd-|Z<&So^4)i!pn&lRHgdTZ(?!EZt zaxJ<1Lec7kH-tR=U0S7Mpr7p_BTsPc{sU^m1+OsuP8!FVok5Q<8G{${v>&fZ>75h* zDo2Rf5tG4;`gnd-54ux@%OW#BSaV4BENkunv3T01)Ib7Up|}tYG*G4v!tJ?p^t^@Y=~~( z$4BH@&2broe&QQT9;|B8(<~Qus=fYDLaya(;P* z*g{%Tc4GuDk)O!Sr_*awr8z>&IYskH)A4GBkbd%gQp2DEQc`+Ia9pOsC9Jhx?I0|s zYs!Sf0EhOWpzC*}d>F-x9`_5zwU9-vuDr4=KlCJ-P}+j3>(21Rgx>+V(?{ga3oN;;;fXR+J1>3F{2(Un-6*WR=tkT4{^ zNOS2(trTl0Wd(ziX_v!bXxi;>A1=MGJDESW|M-P2*VS=vu?K#K!clR$(q?aP-`xIn zNn4sOb73HBA?FEBla|Cs(K10rsjvA7N47$o>7vELltn4+VV>y(wUp!xPucUFQy)w+ zJsb)BByq(0?~Q+);ErNb(+WL+`wEBzse>msb|LrqbH>kTi0RK-{gIMi@zNjt_+uIV zScX3~@jrRay6%-D2MKT^E8?6O)3 zi8J3qZY)b?(1sgm9E>Zuw^!_7>vgMYRXceLl2zoc5Mcqn(Fcf#mo_u=U%xZ}aIYA2 ztG1g<>3X-zVs{#(1s6M4{=xU9qD0McJf7W+p)G6VlIxQB~pUFNtGF{%mG(6H}moN?aIVa363>FqlpK3^)GjC>=hnMy+rW% z0Sdc7LxIY2Bd^BU1_HD!*;r1m>sERLK$%vm1*wM|pjpzF_)K582^66|k)qAm>bgC2 z1#aO++7zEKH*&gdy}eK7u`#%UF_mDFJ52BUF!259XF@LRX#*-BBs#MG58OlTh03z>tO8YUgxk?}mWf>2LSjEQ;4}f&pKrzWMFLgFv(OLfOQ;BFB3}8$!|q zS4hK*IiIo0`&ssvH+S6XW+L<;cM{7VgJdx|Ir;3Iz8V_RV4o69gb{XbJvJ|DQl=&{ z-eU`$>oas57crKQ?~KRv_C9!ZUWF<=WZ%$gAzP??O{D%@W)kU~sQ=pKgm}uq+1cWE z)CQM#Fi4R)6>ZF?RLE$}JY6c;C6?Ok3W(%2BPDXj>@J-T_ela(nr$`AQ(V_K7K_Cj zBEOq(4u;xk9d;T^%M!R2^aMlNeu8i(^JCZ3W*{v0hCdV$vM)Z9Sq#XaIOv0!cfEga zxU?vg^YO;F2?KS5@EzB_xB^1@L<~78wOXL_!b&Qo3SNJn(xA$o0DkW%-;#%1)%X0AmTQwCX!=TnrlEL9G zvE|S5<>vY0*(!t9Z(Ubp%<|L}+z8}t);~RA9$m2e!c&1@tR~qn1|7z6i@bF4f#_b}6NJH3mx>Rlq}=LG)j2<)bGmxrwJ3JZ zB6FlTys{f*g)nXTrJ#zK)z!Jh7mFh+ zN!_J`fcQ=WUq=!(EJjkIqRVPC{luttiY(7P zUmBT|7`8JRtam>y>{u~pF$gDoFvMCf#GbF%g%MwRzn488(aC~wGmFm4xhpbFwXO>p zr&op|nG16^DmhXTGuzcs%Ig<KYWI8D@WD%{yGU*dsgv|Ee6dK40D_C!91B5P@YVb;XU4km~{BXgx=T zA=%_lagZJ0yg6(+p3~!K;yiAmop4`#xH5vvZMllaWo1sa_0u$-Nx!$BZJi2JL2q|G z%HOdHfmV^gh{Ix8+W&#Ppr;&GVg5?+{6dy5uE#+hLBSf^unSX7d80DmIHH-`76yBY z#ClDmq*D2^6`p-;+E0OMR!sXXz}&jzQK%?A@-kw_yO=tJN$Tt^xFOtDP$T^p#6wqj zWbrFwGElXWDHHlFgn`w5jxBD@SHzAb$Dg4D&40qEBoRILPZGLm*w5c1yyttLfy6xiG>{Cnq{uKj3 z+6+jkO?Q5zIdB2$@!l_a4a!MaUr|8eZ(Vh@j>(O{f}pCmZQnM1K29_NwYeovwPl~cZ~G^)v*idXA*|E7S#Ui79eJ7n-G?b=kv zK&oou)qUa1Ukv(Q8icjqd6wdN6@%jY_;t@Ve_6>i6~4I~Zc`dQayW=fn?X&>L2q6r zTm$^H@xFr6-|AE-tS!~m9COWqFmYc$HwdzP1)5K?f>V`C%{%t{V(3vJCD268{OoVT z(FyJip3ye2Xf5r+r+#;UhB8=4cw#Cc29&JAVHoCIy)rK-l4NoEuUP zF-{HvYQ*Ii+^->;_mqH_;;Zr7=DDSrxo>`ah!phb!fNSyraU}4jbHdf*@{uEaN8Rx zk~vS^WAvHsuduN3Lns^tlP;5;#=_(?@F)YL33l67SsF`k+jgV|Qrx^F3XY?#P=}*| z1eQJetT3`Ka?Phyx1GU+9d3->k_t&impT2#C8_td!adV|mDx1KOJzi~mV}HFeyluP+@Ftn;ykk2MjCmiR-!CdjPYntN>$E zPog9siKYhtbtLy`2=_L{tuq3xmrKZ9{$1+`{+xiTv;T*$QeC$tuYk_)RuRKwh3(x~ z$g%8oSu8^cnVaj2{@r%HqO-a~x4(kWU_0BN{hq;E7nInvYSL)0cC_9E*BWxohfeV1 z76PiUt$7MuNQGDIDM$X%N7+ zR4MoCT{2qp1RXJW3Fqtc(!XK(gwh=~OvJ)C zpsr8S04|%;9I)AmF7*%%*OUtrIdhM}rv5W}*tr8Z9>On6x^}VEH8(fs4+1^pCctck z+D>O9u3(GA8(HLD9R#wHv9;Pb@f>Ls>x43mcw_v2p!-dAS=G)9EaqR~dO%}C$lyHn zUP!~lLVA&~!z889*dkNF4h(+HM=$~iGxwE$gF|<$+btD09WC(prd_#dvlL(K>(Ik* z35}#S+h%%)BhCn?-<#W;Ot}_y%%xZx=!lX+FXp}yHZXcONUAr~W?G-Th7fB3_KNdp zaE;+lh;`Y2sn;#uyVn<u@!@)`-S6{4$0Zf4b>OA9iA;@JKwfvB!03#UkUYHWF(?*@9p^fv2>YZ_U3by z&2tvHl&ZoxJBc^k^2u3HjvFp~_vr2wZo^)%d-h#&D`<0?ETjt?uhLpb{~tvaUxN7) z7Dq}Q4~zq*%Udgwp>?3s0|U_t;io;@Ag)T5RD=bGL=wlp4qHg6k!m_mmwo#tx=upg z?MP5pcTITz?{<|fu;l$N#={62hSaE@Oh~F~Hmtkq*StD}0CzV*sgjZx-aSCCS6Fi% z4miJ^gfV`E8$jRFi(OXCqPLC}%2lQ7RlS+4F6-1~8DNO!aIe(V&cK5(?xQ|Q;W&c7 zm6Poqz!%?%1Vp!~x4S#))0sbijcesGf7axWoZv=Xe{|>nyM@5meI7J$#9dXuj7unp zMW9ZBFbXQ4Af;?*GHIYV+Nb%=q+-W*SbCkb%eK>OdfMNuV1~zSW6ia*y^aQ%RO~%G(c=JO(J&!>0rLyET9dCjo~ctGxTl zHb+B%KF`90qCI82YgEU?|$iyGW--r{~Lt?S(iUA@_*@$44GH!;EFFc zk5FunRER5*us-F>KuOaBYH3-VVAz`Irss*u>ONtjh&5p@)M z=;K8ealo^QuhRXEn~tbk#5$0P#Gz19n8Fb0t6FWX+l{B+HgIzr2UA?m)0R2OMEa`MO$>M9ElD2SJWoVmd%8^6 zf%H{v@bm4!TM|vU`Je5;S^7WQ@!zB!)~(&fzboJzDbkiO9hl;LtR-2qGe3DZmm~Zy z7P=hlKJru?Zd|pwrZomlJC-$P20ubNTY2i(EK^+oXvguBK;QeNv_rSB0Lwb1*-!T= z9zcN5=Um!ad1rcnH#j}2y(zRubgr(Uy81pb-%&6^7o_NL)nx4=(*x#ay{3nc>gR(t zuX%B%78%&gnXYZBna(F^r%9Sw6Q%?4PR}A<66Z7_%-0>fgXNzlrs=sG7sC1qh-k_+ z-3k&5S#{xQ<3HG&jOv+lt-1w?`r`% zRSun+(6buesPSIWZL}rHn-YBu$3Fcn)Z^@_4vG!Q8B0~U{?Pn%dUI`-V{xVZ4((p? zwcC0sdF+NW-#-_IOcrl$RF2Lqq^Qe~x;M<^_pT?m=;JdO%#J9;hhQv@hlNedBl>VM zZ|F^B4CVfLP3n`^uP)z#4pSGk#iHcolpdx$yL&WvODM3P%?vd+z~%OCrX^{PI(d$5 z5uy78R^%-vRYS8ihqmgyu%ga}QRwYbR#oXVdsnz(lXo|6<$Kyl-pHW!v*Ad8vAmV& z+B3wau+C@Z;;ZELU$USJ1?Cm2Z2NU93Z%(4YA+e~J@H(jq67DOTTTYi+U2U84FG^X zEutK2RhZU0Yu%Wa_xaQE$LGw>&E9Qk{5cE2Y2T=p-?;*KM(bM9#S5f68yepumAPS8 zaiIAOi4q8!ceT-VM6K>>Rxbzs*tO8)zv6QewF-q+Od5d)4V{MbwmD{x(ayU_4DB|3 zg?63VH&w`XpxkfrqeIhlEQSSD6M6ZdjiMrUuGe!@VuoYldA5dYXz!Exu7?<-9T5mM zC~z~iHAS0o6QqfL6S}gf?7D5#5H?cU=Ekbzinp8Nidx}R4DATCsi!CFN%ODfquGu{ z94Ab_#qqf4iL%8)wV+0)+z4mr`Qhsw(32je(01+y6MC4vF*C}FEmQ{S8eV$YlK5gl zkBz2H1jg=P${}~CNaLi zAg5GIap(9&cc{^(q4@&4>4ra;PR6W|NJR zT*%x1&aruVW}s@dF#aGFRsuHCMA`VpctbwfiE((>+7*kxidR&)-dvKrR7{sv)8hyR z|M2bDXa~}^7D9ljt{J6+t?7wrcB)Xm7uG|*x%g(qyI`^3;&{KwHp>|iN+^8*70uxU zF5;E_%x6}U6*Y^NmhSbni}Fgba!Hv1 zZLPF6W+0yMD4bze_;aNER@kY%xzehp)qAIkvv4$?t3l)3kD5@8Cq0c@53Tj)Dy7asrzrc6>JS>l+x17>+vBxk&uONODNk|b z5tF7Jvl5<&$v?O*=H5LdL6;D16{!GPv*kEI8qur%jJZ}`TR)|?=D~7 z7OZAQV~o%dls}@JP%eYQy^lDObB{J?d_W-ELRUGl3N(eEnsJf74dytVCid^}xaE1z}4xK(m8zw0^j8|6Z%@h<2nm96DPNP8Z+|L^PgkHu%AA~x6HNSVfG;A1^*?9edV>Eo zYZW0}QDpOPr%8kY(qHX0KiW!TeEHgLmTC^p;n_T!(e$Bo1s7z7LaNS&t!CVtgI^+u z5R(t>-G{BkFOD!(zs-wk)mwE9DB2r6DLI~(Ir*tmr7QQ>Myns4iWOf!jx^q0&9gR= z@e`-qB}-5T#UdU`Pud1uA9`T{qz~rwRLKs{x!gSxOk7w1>Uoj2?7?Q2BP;8DG{pyp zBTcJ9pr88m`dnk@9rc@Y*(XBI{t{I_j@w|_0)*C#S|UR`khWddmB#Sw#>c*am|}EQ z5EqxrQsd-znN*#cS7+)MrDJV_pK`c&yj3s%RTtvEfrS|oJuaT|*BML814ff*3ZHS* zd&`dI-z?>2=HMOCDBdqyuPmexG5Jm;-AUI}@Y`~ZtPPW7=VgcKnSE)mV-rsG*t(2{ zq2kQs1b6%@etnX@F|$wR=Gxp~`iPwP8InW?Yh{t?EX_M3nmN#_n>uPc+V?$2tbPT1 z@VzS0#&}0)-}JFf@y9*J{dWbXs8jD!m#5;I(JjroMskY*)ooQPiVRsvERSW}W~>2A zI7(=O*dAVI zyqDNP^+^OJt+wvGKC0~x#G0c;v!Rn8kxf1*=4>A&*^TrX^sG&7_$Fh`M@T0E3DD_~tn`^OuFbHQSh+MAj)eqx08_qq;PCX5IA9TCQ*;d}d{fe355NlCu!nn5TP!e7>f)`!B*!e1HOuOEh55&6rS!M**pZW~t4 zqZ++l94Q5-kcworP#dgs|NR4U5BEBM-GO8>YUtM%&i!5xtZLBf#>Fh=x*4W9t9^~d@<8vd+dM;*i+x#O)p7rqkL>mD!j3^`sDE28$g=3 z9XPHttY3%M8iT@ON%+FQm=2r{U;4<}H&eysZ#__Z<2hu76`s(n^WIuV8(+JM5$!4* zmbF0RPPF)2P8FE2Uv@u9(!iCR+vmk;H;*@{$i0M#$m1{6h;aFw zYHF5AYdJ6~IMjB`cmKfuQ_KSTrT9+Qsw2rrGoWqCJF7v7m0W1AfKgip8TyF2Kt}{0 zLhz|>S;ZR(g}E_>ez^;Sq_wbUYJsV&L}GS54hB&+EhC4je*Ov#_6jGEQXNq-uIRt%9KtTRW>%Tw&g1h z6vbDi>Xh~Y(b6ADP^ZCFWaf1k;BVLw%1>)7b;Hu(*7pW8qqv8V#bz9NNp$16L-IBD?ckB%?1-{xbFxMu&*IO?zwm2!e|6ILv&1 zamGr2pnh`pN3}OOoxtvDQ+hlyT^$x#vCp>wXt0EcxI8*Yi`H$al(7$_P4_y~rz~OY zM_H8AGj_oTMdR6>eb9oED6(z>eHzo*Ifb0?n&CD#NG`wJ9`!y>!JH(hdwzZ+p?D=> z>64_`e(1cF6@f@4F!p{Z?26_0Cz2A5glk?@9DP7`xzgo0nxru?@}Hap>6Vl5F=?}hc7o=W?V-p0m98IwG>AQ=;zi-)a$55pptU_>+WLHm(V?m&Wm0gb2{>$P28`1QjRxOx2y= z>kGkJw^B)ZQcw?`ip!GiHgWV95(e@Rk1I5itEj)aQ+fPDxR9#A_VO#Chcr_oDh@ES zy!_40ix;8ob5^1Hy(Qi?VO|Z~pB{0)rQHA`+w3tpfVW^J3M98^QPF1^ZJMqGrfEBk zXQk+>pVVB)?-RT4_#^ke8RAjA0FTKE+`F6?>@@k$d~bke;UeY8Vyp@0J9dy-`A- zKx#rLpm&<+3l=j-9JTsX^UmjE>vHz%EA!JSn=2_i!2Ys_z3;XVru+9ys7ae6uhL=m zrd-AyV?cHS#_Q?VhILbXAfmZWfkclFy2y=;E`ZfvptPTuFuPeDUYhfW;5c=-T_XHFp&6*n$p2#?8k^l{krT$t@A5v)2v#%IwZ zA{cFPNd`Ldi8EtMf;}rFM7qLs1&z^!mCUI(ZS1}{-42<#{OK|&CzB$VwZm>d6qHM; zPh722R@*DKbVF&J9~bbzJb}XA#bm#MHqlZ^-ic< z!X2ZJZj6yNiBAFY`0KX|w78-b@d+veN7X9Yh$%i@{t8-U7N3GdOZZoVn#0e(yRkAi zyyeM⋙UyW&q4DP~8d`GPY#{#yL{Q>%DcmihM;6N4OZYZ!l|T7~T@GDU%x)-^a^T z=U-ypnexa@n;Nn2L&9Jb~E-W`F#$sr!K89Juy9Y*%xzRQ9JBJk?(8)r`&^Y z$Lu$iNn+}#2ffbjXM!dLXdo7VkyYvTRT@3}0Y%>!Psz_02AiFm`~9g55=0RlOgi#Q z<=_XCFo7CM&)_X9Osc@#2r`O-XRt_D-*@7Jh zU`oW7YAx9IcNo&U0CN2rD!R}!D!}#j8_rfJNZy!T#55Th;u{r=6BK{9kI2}%k0JC` zG;s%d;Hbvd!-MDoxx=8_2QQPVONAPd4m#E>gEYx}orq6LlMzZ`vJ+XAU{rIm!!-X_$!F8QP?#VnlKsF=XY&Pe zG!|3B#>8z}`G$%3X!^t4*SSah$c~;KiHq~8iYwuL^N?+b$*b?~j-%3qlDG9JC@3YG z&I+78i=oCozEAP|@%+YmGYt6)$+un{kB**6PB0dY!kmF8{Bk|#n@D1v1C~)VlZlh7j@bzb z6BI@C7on7B%Me+8tUR&?ihx&DE333OW(0y-mdC)Lztn=tgk8~%WBU#7m3*(X9Q$}c z{xZTt_u@J-96b!9d@dsejg}})+xBDCz|aW_4ic>?m@Hb2DJr@nwuc4Pq!bca4ve(` zOhMB!M1l0i)3md?o(%pRFCc=5DcQpt6YDSPl5ta2E~$kg*R;OWXt^@bS%*eMh5 z5xJAL9Em^gj&UM{qDTH%NFfsb`Ci2wPIMeMFMH?<9nb$*&aZJhT%mdDMg5ec>QG|I z=EkykrUEsJpKZ5B{7&S61$7Ro~auG)$`P;{eHqRP;qqpG(i2+ zRouZYk=kqT2sfczG|&!L>!r@IYf5~E-qcjWes-PMx6tbrN%JtJ$GIgVEU@ssTXVdM5mcTdYnAulT3@y} zcgYgc4xadnzg-3X#4ykurI7H~0A7F(r=S$h3QJ_+TDvV1n*_b=#(`1Jcn64p@5F9c z5l4+@dsWbAOw!YUy?6&A-~iG1!Difb+1k2HX$&e(<>?F_a6_pK2O{Vr)p6esN%^s= zB*dsJX+<%-19wS@QB;|_%HDWhgq?sOLRG5LbNc}W1vqfm*2fqxk5E6xglb}!TK*{# zzZQX%oe8B}=XZtkcSi-*Gl7uMUH#Z`0J5czr{4PfRIx=P8rRa+JuJB|v0nxNCaf04 zQjJ~FGZ>aR84c4%qB>qd4i;UyS6qZ1(932AG2GimsKJFDqgk`=p|JaiR8zDN_rQXTOyo*%bx6Lx_Mnj6xC zVFOP9*iuX{mqRA+gX2HJjyh(L6|;#?Ko5w5cGv880ck<&F#S5gHw!wcvw^zpV3pZ7+(#EF}OR`(CFywV*UFGt2JY; zg{A5KP7U|2PIhh>*|HibK-#Jmdq&Wh+4|Wva111Blg5@GlvX7bCLJ5dxfGo0sm=-< zz&;`DBdYVgoWSnSxzuTP!PGRy*Tq?^%=NRtGn34XSQnR3S>e#gV*i>)8RjWA<3+`V z$37R*qxY4}3O4L%-wZbx@Ab5bTsR{blEE>kKQMCL0>;e`*GrNsr94ZRxCJSo7UKM# zyH59&8R-pJq7Goc5ke9ZIsuH0Z%-4`OrOaLNP(@msvD+Fyqof&um+uJ>REwRKszqW zJU?;h#m7m`{SRFV-xi>G6kElztkx^?3R`M~J4ZVwy81i5^j=yGfQDSHYLOb>t7>P| z7mg$=1&0#42M1;_KyT81WO9n4bE#TkBLaOmK9iPo$9NKztL@a1ee1B6Tw2DFOn|LW z;}=*-$0i>8rz}JZmdTjn@*PNfj;A7?0_jXizJk#(%TjUt!5e<9#v`5|%focLB=Oqt z$E0wvo$BfvEEAjg;!$N1!_SMEv$jl>PjeX4kOIwLt@BY#&GOU~PDBpd(uVu5AgWm?FN)8W4y_>Kb|jNAJznx)0FnYA;4bE$`rIdtkE00b6;U(pC7T2}Uy@g1~ zeIG`&?QJk*+dr>{_}eM$-YBHgnNuS>QIyE~cHWh`CnzCpFc9CU<+ka(|^@9e;%N`2nX0iUo-r=(?3730*I?2QO(9X z@P-&%O*CuMyN!Jbd;sYkKB!is* zJwTZ|mR^;fa;6Y!L+Lx`=I5ggMEVa6rP5A{0TZdT1#E1yX|Zb{Umg6yzVRfwe&LX;e|u4BR8O^fj+RhA+JY!meP1aY&h6 z$mh!_E>``5huNe7I>;0Yt`-Ev*i+tt&u+*gZ%5Xg=)_(_BetaQdL-)9J+c!CR>i_g z9Y6jt3rUDEijFCOSTwN&qRD23Zd%!1g&(@y(}3*-A?wf6HZvpW0PMBlg4r4w#DMD1 z&Q?@}Lm%F3%7eH5UO!!KB@^N?=bj4fXW}--+Sg~ibwb5Zl+v7yo&pEVKlw*?8%)JEoRYWZX#JB+d+p=Vy6d z<+`oqW{&9z`|Dt5_+0un*ZF;m4Rmza3lM9^>_KPFb~#jCn7Q<7d0)w^)#at#uCEj6Ah@c;z$08 z9|Nv}G98-`QDMbIxMHE5G;tzGsQ4SX#U(oPT%GloGPq;?uy;XFs0!{W`=hWeBn7M8 zlX8HVwx%9J(7$2ml4TU|eSj;+Y#|V>R<`Jl0Cd-go@Veurkk&SXBC%%8i)+cEfg=h zXqb?cEse}DY=rq#b=DiIxG)6f3Iz6E9lY?k7!AU5U%DV@r3}*BWRVYh#Q>aEO0FQy zY}0ERmCt@NZ(d9g!rj+W*tKQLvEfMUrIMf;y*xW;7rloR*FIoa2@7__9jL7QX)At8 z39&V>f#STEyzpQK>vM46&l&%kZHN^AC)Vo5Ihra}KCu3I>{F23#z2)P7P!2{3G)i4 zaeM+{2PXvT<~qxOc(7p+pabbPsSXk30yxv9InOM+U4P4v?xMF=unF6XD}O^&5dCWn zON|3+eE|yt@?w#q?J)E)YjTVuWIpQ#DOSBI1RjirsU9hT<-2j;DnO+~*uApDYDCr@ zFfxGZJ0jkPclaQPBI@rdW) zrx5dL0z^VfaW(r?_34efJ5?WNYM@=@PfV)TSN@Vp!+Q`Z&?)FixnrgYkuq)`N7kKz z$~lDZM}}@Le0tx8WKKnKUNk*3IrArKgS>W~5WH^7wMqzl&}3VcoaPCj37E4LeuTq@ zs>sJS}Ij;p7XxzQ*qDC^1KXeb2bBZR@_HzXrjul>x5}?j1rq`u@Shk!1d9w+Me0nYs$&QXL1-+s z)I(PHH{7z41+uPo$IC7N_NX%zX$~?oE?+|0=^FsT3rD7#Qu^O7TZ206nKGGCU@hdt z%;w-?(}Z3~R@xMH1wNe9!9p>)_{a@0(2h{UYy<{>!-ofCovL9zH~}G3of*WzHu<{( zHao@b8NV9RvG&+-Sf@`YZHa$O8LgBGuu<)l?=I zU^-Nz&y})dZbS8bc)$+vw0zSuoq&2*BNBk-#k)=xF2;nfeP^|a4&2sw>O(~^iK68Y z)$FtF0oK7Vbg#tc6Cu;39O0WO*X$fgzy?K>c&Ql8>~-sXkEUu)ELqB@5y+39|FsXK zY@i<{a#y1HiTdb@X^|(S6Lj=WNE>1 z?NB!Hcjv#X|3Homv4XFUg1J-pv41h3#9Moeg`OYsF4z5a5H1ru=`3sB_AZu@@VhdfR`ewj*YFQ0H3iauu>P+EERRR0rea-sO+1Vo#usa_W|^uJwE1 z;Rt`fL-uP2wtkdcr8?+vsqAWC{lo{7kC9Xe~h7k8C{R2&*eP|eZdW?kN z`#uf1mN$bRjGm^#%`i~dWQa6v3dxis_F;VmI@rqkZ&Wb4Ia9Aaa<2u(Q2_FZv`YdF zxR?PLES_UGazDNpndmW2tQRY64p;vj3mu30o^g26D_kSFCE^c&-pzhNFb@AKcrLK$ z;~9a!A!x{ZMo_?k2SYpm3DR*aWTrd>hc4N7a762GRis3sMvCqT{LlF=wEp>d;m(O{+ey?&+~ujF%~i!(ylY{65A{^1)PFX;Gi79TU9ow z0V`mIC7baUF9`(K_`w?yyaPl~J7vE;7)Sf_wm-c=1QiHwl0v{PyaOe0fN0{d3iFmq z;K*VoBDhpuGk6ISB~-KHNh`UL4s<#SpXJ3D(FV(JwbupW-gk&pIdglX!9^H3%Y zwUa{(KCQUxTEyrJ-0b{sMf1kU6S*TvgEf>tV}L58j}0LgXk$W#waX*?MCkVOZa^r# znLsy9sTYKmU(bi{2Q&h=FPud-II~>H6|a61AQfT2 zKmY{s)%f?}2q>{te9nWBR_WnLUs*;(9ejaInSAbhs5Nx?+?quITnHJ~+18;)>2fcS z<{O`Dn7>Ren5nde2FwAeb==f{OJV~QBMi_O8f|Aw>hEU|Zvh<{Pmtu_=F*zTNL*wC znby?ZIbfZ{Tndth)Q&Nu{wy?l$s%z^L?2+4OZ<^JkvAH%v9U2SDF%G$tva2u(AB&h1)-Z>#Xe zxpqD9Hj4XN6T$JaeIJ&$=5(Y`JV((xGswioFTErESNQ9^k?%4}A>Sn65|9lcrdZ0y z$nSEm5t;pB4PA$~-iIR)gAr1@vvnrC-`83E{G_JPEqB(vYd+S36K}jvO7BykHT9B2 zLzkg=+tFExBdI_=8|mP=9%va@>d8W6dS3Q}?{srtvZ~mHjYq6umyn*{gEaMP`h>=e ztLg>z(~tcF1)t^El>1?vu~I(gd@A}FVX{|R<84l}ml%<}jQ2a~ZY`JDD2y}faB^Ib z5apuq=Na5_99m}->8Tn(OP3D`(fM@psG~xqCK?;Ppbl(N-;*RPuSpy*aTi zS+(yfE(VRqiPjW`nT*RuEtv|B%rn5pE=Kzr6KXJ!lt1mp)rdWHv%|fe%PTf~|WqveE78JaB^yVQYSGyejC;oW8rh#}) z8O4Scm~^OhYC!ZCDewR#E+xu9Z1&}OTM04i5u`VVPgcwgf@V7FcQ5g-af*2EGK$UP zxa)UjMk$~aYH#sUh}^d)Yks}Mr zdb*4z7HO1mN*v7yUpEbgLdrvzbZGmGvUjTmN}KAs#9QZRq`e0C#oP#fWdxa5t-&h3 zH4nwe7Z3{dC^Gsxm1RN~tTprn?QY+Pl;BnXQwigo937QGx;z%14jqse;nB3J<~M?R zV+~qr`b<&xzd*Sd*aL=8a;NbSbO zy9W==f#-1ZP4Uy#y;zKAq@<759UzE6czT+Ul*&GvOs_g=1YfOaXV#w1nwq{d^#8Q? zmQhu%UE44sA_gjgfTY4kKw4V5kyel{LAnI#6a%Fdo-@ulkK>r|@b9cgoySiU##^LH{wwDR(TSyobY$8K ziBMGnhh1qn#%b%RxWV>Ae3gnzkghz6jW{S{|GKY9D##bKr)4dEne0JAa~etDuzb`G zt-dzEAg2Gj$iLR+|I_Qj#e89efJTzhfkh(^6;pNaSJrmg8Lq03&vb?$nGdi421;8~ zO?f!L2dx12jx!BLTKox#VCH}b*Xkp=L4htCVYV?|n2wW%$OLiE`+LuN#?488=-P)Dl_LA`c#KHW$!mrH4 zWOs2UAf$eY@s!MbWbYUe&8r{zHk?Y|oN&ZQHRgMB7|wzZ*>EnQlKrpS_$QHvuRxld zCeeNf&eE(0yBq&e?cwPfZ1BOG)n{A3Ea69*4B2qIo;yMir{Kl^l2PXBqXhTAvJgd>Wtkb%U%Hqkknm+(Q0v#8slBMJi?QA}A9 zXAgZK2FLWP>Vb)eF-BEj^9-1whevD}!v|tBIID+jJ_k9XRK;dd{`*jaCkr3^U%x07 zkMSp{i_yPN+KzTz`aA{IMFxd5b;VFHL}k`3XivNZ9az=fQUZQpCvuYlK@*8Z^H0W& z|I}5tg*Iej9cgpp2!iml{Y({}ll@His~br161*WxORoejCly+|snX`}`%KsGB<%gC zJnjG$yOR%H41%jG|t5)lzpJ<|)=ZRTW6{-c(F||M5`? znJYgn`yZn+7KK~ENxPLWC!LhB)TCtY$^U4{XF$mG_s!>NbF!e9-l#n3Fh1H6ETt&v zuft~zz(*%W?Z*GniARhGToA^C>GA&*U3@d=8u7UTr%h7q zcVWNbQam%E@yUkotVVnyM{+nvqP*`-)ZIUgH5v~AuH=Ru4xA6cQVE(b97;ka-@ym} z({F?eI^5Dw!;oTJ2%BP(j+F1t`IL+jxe6ol#gJRW(7lrdUg$J59F3c) z!Zn}sgH~@mbTGrS+^+uIJk5~+LRd8{Pwz;!KK1Peu2q>{+NL3%jP7dk!`qT}0P(v} z0H~+7E&yovsy%Wq0|-MaLU6XN#F02%)GzbP5TjwoPUkp=*G8^Pq$uvNB4#YOjE+## z5kwx{BhWm$R0@!IIovB;?VZQ2;QWZUf*Qx7Li0jZ#xb0whuwVn!>_FMSm2eags*4N zIA4&~C`nMurjgL+NAVC_AR9XWwlD~4nSS|GQuf0@sI_P{9%qFexOr^=@GD;Tm z%08U!uydVEuK)-}v4$rZog07{uyYezRdtTkP1~iN1$6qA-t@3 zsF)N?yxa#mv4O=jB+0;NpD$c%c*Vy^UjXntUo6TxZ_FKf~zyP z*I?y-U-gOO;xaLS`|q@={`Pi*(MPYVMPz;7Yl1z;Mp6a8JIT(n8u1xtn|&`G1-XJvY2EurMs%f523`0cs77>L4N%DILo9YZb?lmTFXF z?k}4-@Pu(7e0o(@C5h(ezdiU%EH}8{@&439u8hVX-k3>|l-MC>+5q;+K4Ewm+G_%P zNUH7jzk6Sh*~o!k=H@i}J*<6ST`5XPpI)Zly7|k7{&{%8`>^Bhr5ieBkR}BAtoGhQmncRw$%~#-}L#=Hvf)_~s~h#QT<=e{@Zdk3z)Ze%Z9*-uzsN zgTHu=!y~@DpM4n9%?9@yH+A*GPXNM?2*if;4E`6FFv*q~4kj+W z|JO*EA0b#nD;Ge)x7f9GT&H!kb_br0oN8awSu{)elWGO6xg)U@>y0!sc|Z$`LpJ{a zG#ua7Cpo4tsprrz08c2TTb$S;1j&j@qypZEGy?$a3)n!z?KQ4yYlxtZj*eaFjo>RR>82aasSnEA49BY1oN=%0j4~6&SV-u* zxESZH=JvHffL5GiN$=rJ<6Ofq%_rV|li9VOW^ApQ|25bwLi1yj_%TefHb4atHwbTW zoLsqmT{JKUX9Fp2B2-uX6Z)((NIzf-#4GBE$`P*y1gwp|A%bKVDa99doodv;hT}__2Zg*=z_Us#I$U4~r~trb z#7Lt{f1lgJhvUAAEbO%Y6+ks;1Au7uRyB4OVAsOqzg<-)@g&^&dnn2Sv=mxfj~iH_hZu7j3P+rwk0 zZg~Ju66z$-xU{~GAS?E^=LK4)!+uW{Vr`+^6q;UFZTUoC^Rs;XS+_|MfRPI816i3C zT?A1WYz6EDkG*G338SY2NF`2%D9YVHN-Tif3#GtXsW^J}9l`zVf^^dx=gID#d>pYjr`k$8a}S@TD6HADu!ewUO( zIfeL0taJPYI_0}>xu$Q>HFTfffUahgKVC{gN4TcV4QEK;W^P>lUZDFC^QGV@r{rTv z)EBygKtzCmtHt>dh)lN~)K8!S1WSb~HNQd)=(E7q@5NEiRwyinmYs4bKY~D2Mf^3( zT%AOwsK|)X5k)}}?A^AQy>;KBYlP(#W1pA%?6kpVhrkx&Zhwxcg*Er`cEMm7H` z8bKh_hN8!G=&RW;_6XP%JGW8g17{_>;d!f^7IEy{Ra$u`DY(?{@dVMCIe?uaYVX)L z%z=A{9T)Csjzt2vOKkQP1f2f={w0~0Y_xO0oU$`>1#)K+PugRBiD6hT8Bp6yPx$~+ zDr&u;`Zz5`_p|bcxb50Y@a}*MhC&n)FN3g+nW#6gzOEys9?JM&UGtN!&`+0BjF-8X zw>XoUEw=?KY;5L4h-d>u?1Sjc0q63p)o#M{`6mS05dfJ96gq~Wq50Pfo8mw!`EK)= z;dv}i!Xu#0Q<4K1h4f>sU0vaT!kKtC@3FsaQtfxqjDhnH^q2vDgu-}j%62y{*VMb< zEOC^|ry+penvB=Kn0j#Zj3mG~F1iq^_lF4jEpP*|`j%^PKVWhG2L1fE1)LwznaghCR~rCV(`*a!U13?I zR|m8_k?|GLZfiYiRE@x_&jCJ^L%y&SQ(T3#_R@2N(UAZ$qj3P{dL3e+F!k*uB2PSZ zvB^XDp z>Y*?wJSaD-yu2Kxk>WIW{+R{MepmXnuYG;!DJmK1-+cAT?@~1T3C^)FR9TmKB*oiU5~GGUW;ziX2@6|gcfh#Es`HzHh? zmfd(gI{o32v*54?NK)T&-EM48QcTeq8#vm~`uNOAdZ-m^L3*X;E$`CX)(VKjp%1MW zI_S^=$YKVK!Q!lO-tC?yIS8HKLgP&&jvqrnt}Bp?y(=p#TX{oJn)rx{E?ETwK}Fhj z+*UDY=z-AG2F$`%I@z{>Gyg0T0)|u`VSTmehj2{Jgrz42VZei)&x-oBxefxT4H8dE zj@~{TPo6_Okre(TkNX}^q>y;BbVkYT_bd!?V~`j^JUOBEJ4O7Tcmne~6+ZL4H}CLh zav2OpG)j=s2JtC_KD;j(`+Yzv>9L^eQO#Lj5YI`ynUCT(X@9I$Hw;KJc|=vX6(Jd! zyF6>kv;%X)19GR01*||nL1EV zxm&}ApQADiR2dViBQ{L$DpdLGvnHQYcv4*cZDkP}$3NQ+2g)$W~+>IqUxTpU z0Zi#7clU+{1_j!8qLmi#A*Y^8GnzD5Q^jqqGP(dff8ukt9nwQ7_-&B z0#NSDZqW9ebu4{Ji>TO}Y(eVLTA$oW`%I8-UBaI&Nn?!TmW0WP9K~Va>_`KFD}7jl zC+O76^?&ZC**{AqWfe47Q&b}3wV;J0sT;0ke`7J@Lh!CS9Uf0O3cA-p{tFj5Ej8CR zx&Wtslff0bxUz8hgjVoOjPUBJNA{GjVn(#Up814SXz zw26q?Ie70<*66!A349E`QK;rk5c8u%lEo>tCg;vrkqKbsmd69b)5t-UT1M4n+1CL2 ziy9oM+)IcKYip`n_Oy-vov1|_R~aOo>;;{h>?TB2cS_!vP zecKvWWVNX4TFqWYXYtzwxH%%TB8ePwe2TXq15`_)@HaojLb8oio|h6!1ydW?hTLH+ zOw2B6JL7WUDtfL*`0nm5t$Uz*;NGBNq{Xcj5q`dn;YU4Kfd~V2E`Nx>k8#Ohk##Wl zT%YV*8AS}ic0dGqpt9&*kP7cr{C%XA6e+t_XkU)M{l3MVGZS0FJxGE#46$jv?eq_| zq?0(?

L}$N5Xp{L!xD1;QBXQGd>OF73xNx*mNjR-h)1km70sOqv|4S(EiL2cB5( zl|AcTqlo!JcEQ(zq80dQb83aa9kk0UPaM!rIkvMaGOM8%y+PoR7_$_TaCfGC&BVQ4 z1KO^6%H~a(fss&6S4TB~I5r!dz)~Nm@+!7pZpDtmXVU)8tT*q25gJlkARiy|rh(;_ zB~XWt1_w?J9pBr{b@pI8$GOF^CIL9qj%R`CoyF1P+agfCacG+qqJY5j=CpYw9+b_9o(0wh=H!x3)&CaF+mRPDfT|m3ME(^a?i>^ zLh3dFdt2UgEx)bOFwFRpC(yhnWPeLhV*#NdtjzE4&6{araAq-73QfcKGy@;slnmD~ zIumGrBsTDG?|dTPV|M#?KSsI<{>)x~k{Vj!7k098Xbe<%6nV>*JH>CaTmUVA7ic=BxKoPCTYes1@>idn-W~$tq@RZ&1|D^C@7_lB zirkI*3nl$<^l_nm@&L71MbV47D{?gez;1d6EayCqnc4fZF?t#SxEZ~)yc7%P0-3A$ zxdfaV+A6no5!z}R4q;um|5!{*0?yD-%zWzo_+y>&_xd z#ft4Koj7?3Ouebgi*wc9W8L~Hus15u=`2deIIG7)TEI}{%(6ri@Ra3NlD*HVB4pA| z)w66YbMe@5WFD97bGHC}LqAO)S~=S$m=m-D1qQn9sD{)X0!}NvkTygzkezXdVh*I( zUs%Yq@bW*@VyO}no;8vvI@fSMFlfKGRG{|aXz5=q)Hu%?6R>!z3+qr|zvkP1b8In0 zJ#xqU&H{8NY3jJZhR~~lzXjgTZt!>~wal5<$eb#Q)tb2IbboYLE3@2t*&ER%H(Rm8 zFxA(=nP5AbFgb$=U7OA6jI@+-8J_Mh-+|?|@>w3m!swE-rd&rfDDAV~*}K!_FYas* zp+;ythB@#jZnlJZZbujf__ugDr+*(~vN8m-r*dT*`9a1=fN7@&u?+O=MaTP5+10DJ z_^YgkA(Xh;zfzcawchcRf8{m3sx6E}%g7u|(OykPFy8`IYj}NzC4uxYBh_wEb?unm zdOSX8`jlP1K{~}-`~mxvq&&-OE7MK6Pjy#JrE^ml(5=x)haTX1Oj_R}5(-B|f_Y2= zPQ`JVc;ki!#Y;NuZB7_%f;8TI%{OsmOhtos^+oD;*6Z6yi6^$FvJsIjtbZtA?YS1S zzZugr&?kh;$(()L!Yo$u1~J~VM@mr@)yI~I2N%4|w|&EDwLuTP=ZS5}*W4Z}Cm##O zGwL|0e&HX_;G=z~Fm-S}#ENgks!WXP?y3}Xz~CA=GhRiY6Xlkz8&x5Wc14_`dXIup zJp)oRb3Bn*f)Nbo_owze*#l1v6x<68UcxG>l>D3PvR`l~A|c*WW}4VS++faWZDYUj z<*AXwz}u2UqyFga^8M{{`oAwY(2VNQO1sO(UKrR?Iw}048FKoSWBS zJYe>6_eODGWO(Mh_ndA<=~{C764aWM)^ETIu*##~5?EuvDe$)|TlBZO`*uk`j_@+x zgeVIHTXz!}>Q)~>oas?8ot%<2Dv2R#KVa8IFT1CclE28sD7LVL)VZ8tFi*^!B4x$7 zt}{5UJD_*}oFJMMH}pN~$64Rx`%u3KB4n>cH&eeP|Bp1}Do zDVD27z+sr;8QPS~D;2$WWv|gJ0gE2BcO^i_n%<{;e}e$^y5Q#dK0}OeBT-c7?Kthe zquTu;wfVWc<>DDhFZq~~6Xr*q@4VAD(w={{tJcE|XQ1B)lWUa{pi zx^-b4<2Si!r!z`2E=w<0bCzY6+t`#eHu`9O2i%w@8eBX`(T=#qS^< zYx5y;RcEXsPuw$;w2A|FSNjCk?Qr%Y1|>3y`X$GnI_*xKi#SRR)ci_}0a*s2!}1=5NXAR2YhL(QKB9GNU`;_QR8E zrHJTYb?S6T7d6Ze6jZQtcTJ3{F`7|&c-pLfy>0=$rD<`Utnb#Cr3w2Yq7h`SJ%;}J zI@7*FVOtbKie|yr?5)vmytN>Cs#wM77>}Cm5}QwyF&#SGLg`_os47pz!Guh{F1b+O z_|&LcF{QcLz1m~9IRpaIjt2UN^2F|jo_cTYH1@cxNc(HhJT)DsLGtoj%Xs!+wu+k_ph+$?9%Uq2Djvr zrTOqQT2F9)V;i3zyO=D_@|ZdPS&g`Dk$tjerBR?_CFj#B*2G-JejMLBUMwq%#Ypt!eObygUqNS?<&AeVlA6QoyV)-mm|+uOR2LboSM0N@UtsJfnG&rh zDk;MvW&D0UyFGrjFc*DsQp(S`CZRz8W#DX*W=TdSx%VB^qTW!%NV*!Iq92W~t7$?+ z_}K(Fu5X(Wq^%aV%w1_k>8G>wI}g{q=}s7%kM-u!dnEe>XWlF+dZf`*hO#5qK00%c zTYSCp?GxBh4s7}ua{Oc4FIDm9B@&|pxJI5S6-sL8i;DMK1QstWo_R7E=epYq3G z2JqHRY`@6bvq0_9klhTlxHi~^Ga;O-=%=2jDbvwd%g37L_N?J$V&6Qo;qH>XmY}IE6e0QLazOmm{3I@(`M$0J`*kC{*J7T)Ilw{GNtZdG`JerHHFiME5f$Mt-IERQC1f78n`nN!-P*{aaj;J?K$ zE{iJD8*Z;*4=p24?CKrjL+?~?VtJu)`UyE#m6*&d*2oAY*=O%NUqxgWTPGq+9j-d{ z5(W^?=MkE3eIzJTaBGrym&A93nBI+bS-daLur^0-7u{M2uSp2?evHtENTh32M{@yn zu1i^F(NTq$A20Apr9zjh&yfDoh6-bAVT?68_dTC*D%76gd-o$@j~r83JTf{I`h;XVdr&237L7$P-eJmF+mN?aoa&3< zY#$P&XHBDRPT##r(x>Mm!Gsy9*XN5OHBA8D(QEf~i-bua0L87R;b&3K!AvycXUIdG zTu;#eBSyss2OS^ZAI9?>>87sIS6|F0nyB4$x?f^t@P63*t74?epoDeQMaSUFwZ6nu zi~jdA_leCtn!!&@Gk2d#h}o<+GL#K8^KvyE?~dqWbCwMXS6hj3+Mb@;=GsH~Z$#*M zuylts1rF$~h7deT*AasX>!k|@RY#b< z^`h>#Y)S?iF6GKCqi-6=^pY$+XsED?^Jd%3+A*4Qe&p83fj^~#i!Pb|8k5%5O+TM*{Oc6DL;o1(f0}VK0~8~ zxJ=6a4>U0zjOv&xvIx_mb8bF+smdfU74YCriJ5Kz?Hbi49`$gj#BLUW5Q=lPvdV$q zKZo?|)QC|I6?u{^3>NKTRG+zDV$AyRt#y4}!UwsXWdry0@5@bwMl{_|^KI~K6w!d( z@pOTOb;3}5N?iSYy!m_|2ewYRw*!r)%U7o-cP%T`JNdE9zWyDkteB(Z;BmdD7{L>o5!K1?dl9;4f*vvMJ!Hb?pAwS zl_+s)hp9TDWY&a-YnmrxaMb#q$F2#pGwBwTcZs)(_7bwN^d<$(T@m-JtXmm<(ui}{ zhu~W^O`hWl<+n+R2gUZPI~7H_Z_17?lhBo`466No%ynd5z*0%{l5wJ-&aA}9lNUJe zd#dv@b2%JCJNKi4tGJcn z5F3NX$l0VgoO~cM9QJ^3Lx?kv?+Vp#zEc!Oip3y^&DyAgZc7AwA1{yiZNMP(=2v zMW4dv%OPja0tKJ-+m+=7!KCvu-WiH%C8w`%dnG!O@y8qHKfziNMVkcqQL!~fuZ@d+ zmUE>0o%vD$1_X%?#oksde#7!56O`!ZHa}Yp?h%HOE4q(_H?7~dMPHn=-kgls=q%XL z!um8uL|wyjm!F)29Q8NWNwEsWd8cVrxEPc(J{imvK2nUh#H zD5WNQM2P*agR4GXbA=UVN&;i6thN8h0lz za2Ri_cN&jsy8lhhy{TH3iy@hrK|U#UYFb7r8`ow)lNP7`^%IJix92Z2gf~)!hTRyt z$uB1p*EH_LFhCqK5u|O@UxoKUA1=xMvK~vH$wJnyMzlv%GA!9Pr$`UV@;$QLqi>*6w7KeQuIIJnY>h4aMB{Kq_0zXuq<5uK`_3e{k1DX z_7Cov(qm!7d(PZqbuLTmbn_&pc^KTj)0X_qmn~jlu0Gg+pER8Dp&h0oD36^dhwAf(wI>II6Z{suf169 zOyoufS?qW&5RQ2h$z!_2AEQgdzoBIRH&>n#Q8|Ub|IIS;cgh#XJYVOWp}mL4+O$J~ zCZ5tG)mCA|Ki-!1*WZ{KMmn--tXV0-r*f9sTXsQ$V}+9;`Gj$+-Q&+3S8q4QHE}Gv z=TT*_F321uq}CuSw3kJBu_cLJOX#@VMLFg@dgg;gQQGrcbjIq>m3zYryrhP-rn(hh zdAuq&NzzIZM%30>`tUYVSNpqLTzqVI%u}97y_!o~&myID`4;^A_=WJ$Th67qEKja7 zD9)I53@TboWibgjbQZb!=R_Te6~CJHFr7|6{VrtcpcRp^0fSnK-Ctlz9P3J}jsuLr*$I5sv_+bz2xxbf2g`}{fn%WjRRH0la};!$Z9T*7VXHOx>JEt;->R>)~DK9_EE&{`9bKsQ)1~R=S02?ts@BT|#DN zyPChq;%PLlEfeiAd{pClyo(*Vv92$L#ej~PY*XvX? z46l0{uH(+F9iiA(mnuN=cCuCy*Mo6`E1u(Hl}Yb63IUt9K}Gwmb`de1=hO5Hob(jt zLGu>%L8S?r!;ZnCh4;vA?D1%wIpCKp9#lrfjDc&L-#+HP_enz@WFog}7Dl(P&loo7N+o#7SuLW#DP2A8Ih z`K|%fqS6#=43>B&R5J(TM^2o3@a3gfnBHtQ$KKt%1x}~|i3%2D%Us~*xtP+RiQZnR z8n@eke1unoz~Es`&faCTnoUoA38Mtl`J_Cho_zanv*u4Jz77@ckBM~)cfD>qMrihg z=n0^Qs^{NSHJzZ2I(L?B2UjIH!adygglS>HGgcnuJtdCR) z`5z^jkyJV5ZroY_gyndtlF0Ki)@{{5i@NcN<*k$NSR4yH-(HC%Kfg`NsnM@#pb-nZ zgzL_=1ug=H)A)To1UFVgujQvvDKxl$dcw`Q1yg?fP|N}PB*ka0PsVmTKW0QbZ7;Vj z!njjF%hJIpcq%d^i~Y+~?yP>Yn$Zq!krMM92j>umb!$bxpS}c@BuBRlnUb)=j5D

NfXo_`UMJt zq#3+xBJR_T3N~jwAVzbS>|aqBt;0S$A)?4~j4<#C{$1NoQjh$X_vM?PEcXF*JomFt zS5}N6tFDJULp>H({sr5{Z=4M!F&0TPryGy##9WJ~5+XhCNqjYrW@ahVseu0q7s}dO zib);sIhVsic>@crOY}9@jfD-THR^!NQyU+<#WmZQpK%H$G|DM0*=WBT2^huSW}T^K z=l1k5PyX)RPgdiNbxWUn##pSAhcW3(f;Mg3gRA-${n1IAbXLh(;(mt~^t_kK?|W>L zR-GiiCFP0sDu?Y@Ka%0 zt8SFZ#fP^@>bT->im1)`W2LbbcwV-$yi$jon}P9JO{ep9@5z@-t~)L>Snr#x{i_;u z*tazqwoUl*Qq|V88Jwa&tg-xt+HE2VArM^!vzh&z<>RYV8Tge4xC54XNfsd70XfezSP5f zw-;+GSswVPTuBc)bWs`wl_SLuIzrp(f4k{}3ww^7AVj;W{h~np-?vH=E2wqsG}`}> zUuzabhdGuf2a4=`1wb{_IF{Yv_Jx0*D3%T$Xj&?6F7N7}uO6N!9FG~ToyvE>U-{?S z{df)Of9awn(uL;r{q~*@-niB+ShH|?drtL3i&pf!f>q-j5`qw79#_I5{~owH&U-Nq(mNo z%PDUzb(vYU@z`BeaU2U(`B)oD_X(N^#F*Gbc&c78FrzGUWIc{xnLcu9;pV~p5b3}T zj)uS1`0>DY=Qp79@DXS$9|3N4ac^h2N!Er+kQKOc6*4-2Bbm7}pSz9%#p`p7r7Nw1 zd!gq)#DTgrjUOS4UUq&AOd?|jPThN$B;7FBcjqbTNBB81WENJ#;OG@{%f}}ym)OiYQ@eehc5>wm zn=tJHSmg_xpM2_;dyqbl9|`w0kgH;3iCrFanSU~0zRzC*9hNVk)qfTcTW8$1*BA_V z08@JpdUH?xi8-sgBm&7u!wm{y?gKMs2mm3-yIu$?aEz&OLKfgXH7mnKJf)4w`h(zb*5_(Yb-QNJw-rHPXL8>nO`T@^fn>aJzZCR z(9Yc`+>oft0313C=7Wvx5nrZ&G&|?D*%yNWN1~(|3;*H;DDYygNn89Ay7YoKD|>*_ zATIW68Jhu!@kvCebBa$L!=H%#>cVcxUIRohb&R+Wqn?nbfd{XznHN3Hfy3&Rg{|}t zWA)@XMyNr3%){7#6_WB>3r~;1ZBKNAz)-n63we?9ZN;bdAhTHAbe53$(gH8>&KGQS zj`jZg^5zDWLpc~y-eE$11fmp@A=rBhL}}iZC5zSXVG=Xul1|p$6HKT-FRzJif5a?( z$=2-P>am!f2>vFg9p-&&IKOfeJ~s(K`#=V3FKl9bLd_S$(i&4Rlqq0z8JhFdE0Zm6 zMwp!4smWp^p~lhqZuApuRKJ8fRT$81HFx zeSN9&Qai@uTU(%8|B*H!>OA2JccigwrEV+6W02R2dJiXoyF&p@%k;%+hGjnt6q{+7 zP!L<^vE>h^{`JBJAo}$2>l<&++k)MDAyb(_*O|rsGBUA3>O;(Hk*C!r6ECb&Z;jX) zl0?GUq#Gn2rO12U-w`ptzy{O0neW#8&1HCVs$rk+bGEsY_Fme-ulqBBZ(FQuX87Uk zqR=snCF|eaZmRUtM#k)I0I_^@gAaje6aXOe(@aJ=nL8L5R~&C!z7@ck(%c#5aYXe5 zCn0U1KSLy&63a%zrjMfq5VS9_#rx<3dFw30UN9ZlmY~QfLNwFKk^5sm43G>x=gnmn zmrinzofgwD+jYII^`2~_d}8`HeO&-|JHMeItEX*(69P`^lU1 zv?&wuI^K(tbN}%0n6=gk#<&$2BK|Jn0{Jq%28)P}(m&4J)9XBVcZI64+5^WgXoSUF zVIzeMkd)_ixtsSMY49V%u=9p{W{ z$=Ku%7bAG)2LaN-^V6XeWvszk9Uxt z0$(R5e^jv$*KZ5_im;ht+wvE@l{;%$Y?P|645aCXaSt}9@zoSqrXtY)6 zz^lFkqkiL>T5FBrE&@NqH~+*C88R`?>6EB`^K7(-Bd|3JA`Fn zxx(}4QLU}U0h{62-|8@?mKWkJ|Klc1FwBp%vx!Z4{n(Ubo@9K^l|nBk z8X^f$1D7*0%C@eiY8GC@^f2}pk&ZH7NGP3;gt?y)Fw{-*2q~tQp}K?mJcL9F`$cdf z2_L&N_4So)i6de34Xp-0pnT>riW2SS2l)^OM-S>Esj8V23u>6*uV5(D&Sy-InoCVt zaTgZ657vpz{{h2;m+>?M=YAxP321FOOrYVAmvy=Mnr_8|m)Qw%XVRV#HYIrM@7>v0 z_!4K*pVAN#i<9O)`BFSQ7ipEsMTj?%s}w;rHw1)$H_7b@k5y-N=vuDTAI=(1c5>F; z=bW8=_u1b*d!IvVLxUacXV?=61c#92OV<+!)(C-MMX<91oSK6jp_Xp#s<36CClKfa z0*gQgBqFdR;DJsD1R9IZ0wkgXrU($s0`usA2H7moL@ben1^5y`xQI>=vFIYe6-h)A zK(GifVO+u@NLX}<2oTG{wlo-gBry2JB{+WX-aS}Ehea${0!TDCBC-rx8V=()5P`!Y z9G2kV*|NgIyD9KL26QJw%Rfma5GXm8@3`DHeH?+{EDTw?IBZX$REAL}GsO{7O=Q!x->Y6*`rc)-`(!}bGL`>!%HvhqGA!ZA`0_y zQ$xA&Y5(<7_L`dP9d`7ta~Dr4GqDHm+o*xPLLOy58|&GSxCGhy%|aVgKy7<*{F3jQ zYBE17#K&1GFzlAa>qLQSZFE4}-LN5MUE9N4G&fL}$y|nPoebP_=X@wfcDkx4bJLfy z(@FtC!8Dt&!Q`H7S%MADIj1yD;N#onOWpxcwy0q1&ma=%etp0mYXt=~I4=lc^p zNY`=XT>Pt|mtK?jS?heJN=^Ck4aOPQOBIE?ttTwNqtIqU#ue(cqtU0(atduSMN0kP zTCuBfEwSF(kvpFsOC&LmDi`sys4I};%F9j{i0kg_JYE#+l!GC0<>*6D?XSv$?y2*TlkCteCwV(@NzT9tM$xJPpdztN4hC!!&%G#hxzG8KAQkl1Ep`16qY3Ig|b7LR~>**C~Ge#lzD_ ziwBnX7jso?>t>$s7_KHB(kM~^j!@vA;tZu`;IaVp3<5i`3$(7c~W-Wt#hPxH0s}aXVo2^ zBz5%JQgKol;D11CFz8w|+Gt+N!uIX6eS;BtbHX*C=H6$!Uv}b@D}C2%IXYdaX`^lI znTdnT&M2eQwJ+euom#!_L*xCR2@SDcQ;AKNS)LZinV&1{tJ}6{{-Qj4{`0NprxNSmAM3~LFsdIEIH-B@ z8pE@!Bg~r-?^>5--QEO3cntoyuRX`bg(}txqX)XK=K2o1Q-Qtu8Ya^$f?RZ3H_Ws> zJo);1cZdQYv63%)@z5R0rzM>bi>b8MXvVX-e?Pfo*!c}zKcC0bq2BJ>tqgB7*G>~$ zqCQX&W*12+C~aPkpczdU8EeYmK>&}qzOyu`rKTp5W9;rIIIlTZVsySUqM;|0?^GPf zt>!)LZO?eOAvsE{u^z(9JK$`zF3TCjfrAa&k35srhok=0f>%(xcjxUP)D(kIFHN7O z){~N20xF?|M+N_(jiVu$U7jIPcxf8%Ek)D6y|H@oFuJu=H?azXozNB>D^Am8bemP^ z)$x8KXgs5&=n4Z{FUZM}hXys+X@g=dmAEiQqF6Vi-!y3!o{?hCX>#IXB~Hfb9}7GN z&Bx(;{`rutY&T_pUKRra**}yO|HNiyo<&nz-si`DV(@(xrfHYQfl!Agy*yLXka^2m z7^i9J*sx2C<&j-))hO)%d0G+R#J8bdCrgew}s zRU#tK++j&0nKy=kW9NEn2;vJMyc)ky?!?Dlc6GPB=KX6C$NYqb)PKhn4oex?@aDBh zUrdp9chwy#)~Z4u=p1Qy+RdOdzu_NdrYBW#+Rp19MeGqh-s_UlWKgwD89v~&2HZbV zVjAe&$d@ILK*97w6snBNMc?sLP)Z!vEFtx015(cX8ph9QXI>>S(3kA1)dtB_6HB{;eMP z!#w)*VH@;GEAx_BhI6}2aEtt+%rrozCJp9>lBrT&kye;k+vAr^te*^W$WuhaHMAmJ z;aR90N0WduELF3%nkY@5yNE>(JI`Pw_*p4cB07@uhSXoo^ML z5X9%uS@lDvw+g{Q+!U9jnD?H+G&%>)0Z;?`Y*%m25X_X9n6V9&JL!^uUAuVSy`dqQ302Y z)P!llXTQgG)C=t1>~`g^-jH$*t=64}*~XkgX+F^T==7lIJ_n+4^BYPU)i|Sh<=Dl~ g*frQ2_y+EDAi}k2QRv}p%Ux^8ve2bvOSbO+6^kZKX8-^I literal 26810 zcmeHwXH-*Lv~CC}O$9-UsDL7((nXp;KvA(#Y@ifTK)Mt`dI+(BR22mgDT)G8r6Vm! z@1W8{@6vk>$y*xz!>G1`U%YOxI)NW{7|F}4n8+*0o12>7ziM$u_pZWSv%B|=bgw9!g25!b zysm2*?!CSX5%@KU_h8kd+VJp~JMf~fl1;CCf80EPzngTgh0oVC`Q-D`;npCdLn7U{ zk48rjyH3)7i|p|2AZUE!*>4;@;=%tRS>*%8qr>wxiwf_5uuu^4S&c_zV-}|#GskpD8;Dq%zWX*PD4L`5<>!Ys^k?rZu!!(bK zaTc{T%%;r|42I%V6v*6KpESlulpW~03&S%NCTh7*oLnlQZqS*hE|4;P$Jxb@p8`{3 z4W=7UtPXf_%uV|Y-{B}vd;!*tvY%^Kor2P89P2GD8X0@!K>3+cvFNu-TtBlPEum7+ zw3-xPA1V7J2_bj2lc&%6AHDC_=}I4Ry{E<6EndJVsgRy<_p1^01Q$#zGwaEHnh8XK z?sdnky#v;tqr|`G@ZQ_I@? zX}gNJWuD%8x6Qgl=mBnK(MByOfU4`IBnvH@C%2$#%EYM>lv)bUM|$m4Ym`q{ONk=Z zw~Q+PXqKPPAdgE0p_wBL$Fq6=2#?#L^>5QB-ZihwCunic#ANXYWe!i|_ zxIOR40v?uj%@|)YH1J*MR5*caye2Ad%U}?PVTWk3-vKMVsjuH_?a8>Vgot}&c?{T|EW#V;z3E7Ps%fG_$17B6kQwG01qxF#N_vP-{MSwIB!zvrDXO{%=O~9EzuIL|8EnlhOFF1e!i0|H~~Q zr_>wyr4Q>pk;%)?)4QlV!1VMn^&LS68}VpHFBYxLt;Ju=^`en6E-as|hRtG<>N{nx zo%P$w;}qUKCPJ=c*8Q04yzK4+Pwr(pJRX?|w{gBG(;Sa3vYIfJ-lF5A?Q|TmOE%Bq z(e8GSoXf)#wfBD|mhb5DGzv;ic{}KyU)s~HLlb}ODD$@7+|a{A83UPnzlZNg@i**+ zeQ`V{k!OA$zs!QQ|8&5ky*$)Qy)Oy%wQ-%6a zc;flX7j9nh`*3ac(?OL`igS&=uUzD)?$AER1^p773mJKOX>=+`YO2Dd%08=Fq}n#8 zd`P4?-`=1vz-?f!}s9eWoT8E}~!I z)^N*_zs-4R1XqVa5m=?|EVi*7KU}YARat^a`#GCOD_9`e|Bi7CLvWVjFIQhfY%3RX zSQ*~4|# z(`_WJ9WmRe>(h8It@T4#SXgFXN@K_q^N3&vWBxwlmm6?&E`fR*2SPe$q0+qT#z``= z7MeUJ8PmRsfC**7w6bN>U4MQNYmM}FI>bCDyTibHoTxKyzEQ`6U#BJAGGS>nQ?MmQ zb&N%GVYVNI?@#!tBvofTdg$Uh6o=yF+d~!$z5c}o&lp=v>+|zyI~=IerCl-j#i5DU z@lM?A$m#EKeNC@TP2#g{V$#|i66)1W&KMXNOb;~(Cf1C!+O~T&B^eZZ2%MzTj z=Gq(=ESs3D97j^r-Z?M+vaG$V@L=Y8vG~n-)j84H@2&6tCK2RzqWPT4Z>0uPQ*}7> z&Am{-%+zZvFqzQirH!Ow1-uc91FWLP*Vwa=*C?Bqbm^qW<(#pDg)G|dc&boS;wZ)t z{B*>W#ueE?yLQGud5)E2U^>nvBJ|KKVJ05656sLQQI?i76PrGkvvgo7P-`Yt8e7I2 zzgK%ckEQD&t$-U?q*!8glMQa7xH!vx$}fWU-rw5}4eD@N4(88F2zDGTRhX|ZK=^7G z)<;yk`x?;Qkr+w-If+l?9Sla!HWef_$dHgMa>`p&rMD0MD%mcHAI%)AQp@NMcC^vT zuJ`?fS{~9KJAwNg5j^NHww?4%(Jd4f&xVfOYfg50{GdTAjhid>Yu{U~lWkICSwCMQ z9-R7nW5L9qD}k9-iH+HJdx=*O1m^dp2k-E2{qa`j&3ulanqqb) zX{ab87J4K`)4RL8#M??9B@HhtNsa8%xR-L>nF-4yyHGkXJnV^IUMi;KaH79?L)>w8 z*lwbbMR&0FGcZou0$m@O`YrxjEM@T%KA8wdd6JK*Xpxa=Qm04)v*dEL%j2`Gcr^=Bpkj2GPT>E;pAJlxn<8b^9DOjmVRJ@?Ff)j@`cs7Dw`p782(OTXe&6=u`7zlmT1^E#+qmt+Q6MkIaeSn;Oy0Q^}hQm-K zWUD4i3)LtDu+*H!yZ%~TXiD}H1*!RtN8KNNl`cppj{fbgp+TZJJN9qmzIMk;wBAj_ zJX`cB9Y?T_m|07DI`G&RSexiqy#o3^#p2vp2wiPc)6FWEv-?-4Tzd`(>Po`a?QFYP zZrR3U~Xb;QzIo@ik|sP!0MYbuammCr7V6L5wPheso@dwi%Twrl56WbL@TvZJo zc;^4>mtNB9Dcf;K-q*VnEG@ZkV=Hk2wTfWUY3>1nL@Rg*XFPPt?Y!@yLM_|Y>))CJ zR;^7g)E4MX_K>+XQKA~u6~LZj>ve^R8>C7_)uH}Al8z+0kilEQtqgQRAc;D<$nk=m ziB+kkBr!V}+6>%k&21zVid-lYkc#Tz!AJiXP#oMcyFI+>)7E#Dp0kDqxP9JTGbc%; zEEl>5+!7bEAvJM&!zW;GnA{hMGCQsbIZu`w<7u6H{ghXSx^X&m>_CVU@fmu~FL1cy zShnq@pcA5Jz1ULV3Q~_(=UUSS?Py$l9Z~Tv>|~X3u+|gYLli;oOSsudtse#K|Bxg< zBYkd)=M@D~^`h|GMW4%Wq(es>W?eRDKagmtpZ(1Bj{z?Nv+_Hh;JW@a@qM{+AbGEvT#BpUjCW z*$U?SAo|S$seRR60D2yMKY)+;t+l1d{jYf`^esI9>2?oqZ|_FKFs~@dlSv0wozzC} zV#tE-_LdR=*~M!}^D_BLPY}&ygLN6Fu_4S#WE+XY1P2<$^-yZCQ4UzasnzuyvJU04f#P=y!prQrto| z!;P$BrZ-yby1hO>ty%MlsX(Ed{i+Ix)2Q3|27z=y1Q`7czix0vpO}J2x`Ejqcp|i& z#K_RIz%o+|v|`=Y)Q)%&dH@XWzANuIaRb(7P0@T6tRwhA%2Cqs+<`kidWGuey1OAR z-cT9%F~|D4DAMtc9^iC|KDYl8sm~0R0oFWnQ!SdPFzZtN4=hS0N{)bEo|-rzC3SIM z!|m0DpifXu$L1fb{GAelfkT_&x<#JPD#fsTvrPse#>%zB#JOg3a=Hv{v6d&=Z?ocU zkjhwsaGr`yj3Iic8V0ho(A3A5dDcZm9Du@4po1p;Q|9zyTY!i`QMZpQdsDljcyU6^p@{KSTNAIhvO3y$T6;}d)a2_HwaA_SN z0sMl?1i6?LVW}?OS+oj3;WK$@!yd!7?E)DCA8^cd!&b8z7E}Gz)osmds}g zL1Af}S$Qe|CdCuQ9GDVAHwdgYn2(}6@+g(j-xq+Ay=@-Dzcc=D$U85R>WeJ-~^ z)kEIwt#s@yBj=*yGK}$Bvj=N?n^e_}K1sWj$35iLg-`>#EUuGoO4#v48z4ghC|x0r zNICPqG>ELm`@eMb!@VWC+;HM@Cc;L(2fsXoH<`D`0Yt`Ymz+;kE6=%P-VI=JY4x|- zFim|B&e3JSmofNvDdT&Ty8wKwK#lP`mi_vX3Ubii%nRz5} zug0(=oa}5X)iY#iabk_PANfr;#cN~yro$M)l4}(?mlXVzk%Lzs1g5TJ0=YVHTKt4l zLMvj$ub}VJ*iq#lvvWto2(nQxclNzEfk%?M4Z~Ov3R8kFj@)~FUA0s`DIU;VmQ$7D#SdKt zAb+~RWGVl2Td5D$wjJ3Wo7UzI)Le{NTINCLjwH(cn)ge&6QBeo3+qT4`)Ft(7+3|t zz$TNM*g4=Q8-eV*C(8w^QM{vdYT90v;_cEv<)Y^&{V{ukGm`ma+KIXLniq)T0nzZ5 zqRvd6nh%$^#D|9>BHW}O`v4DqsDAAd=`_gTTR^h8<;!tyV&o-yWqMAPYrrpWzO*Tb zHPzVEh*jsgDTqx${Ox5oHR6AYMueJwCr?%u90yRQ=%=#mzN-;RfDFFDCDvk*&w5Pn z+#=d*&Ec(G)Z7Q@Mfv&6vGG)_nFmyAgVNLaLP6(*)AFU$hpWp*>X^ldsjbHh?L(Uc@QvcOd!Ltkp>j@ zMbE`Y95~KYP%J4ql~I_V7BfLkC;fa5Duos%8~yZ?Xxp)IJQ&)TCLJF4IAa*V7|Tl0 z25Yb5fR>J%YYH4TPEXJ+wi#KPFP3G8@Zh#RVtlty%QJ6|-Mt`<8_yR8+(PD?Pl4iH@52)ljHi+NYnyelr7@DB); z3PCm`go>OVlaX&d&0K_B!nWV)DNAX`^@&e_QG4$tVpUMMZVrWdn+g8R0|KG)NGdN9DUaahP=vVyn5ES5=UB7MdzaoGH|o?X zcOtmT;`Y@fKGg$z>Ek?OQ^8dPwezz~)&6Q7llQ4v~_!5iU# z3N6s|u48w}G?;Ujv6y8Ih|2^bYE@cW2%s!JSu|z9gcASI{S3#1|-8rWjHthG^wA@UzGVY*;(;iSG-^MO2u|njD zo3(Ul3r;RPVY%f70oitct(*S)3`h$Vk-rrrXQ_3p$Oe1Q-)AxnwUlOAuv4J`>`}Am z;Pd_R2Qs5#lafT7r^7f3cN@Ih+4S1N!+fqCh0?pz=BtbdK+V{q=4?I0CjvH#K`!*+ z3M%^55U-6J!r)EtuYQ_FT2;%4i?!Kx&sW*zFtTVnSxQgW$R~hmbnjSFgeSXIg>V9B zp(vgMR?jMBW9kGHwhC)Nr}0M8N*{sct1Zt-Bgx>Yu4(7eS-w^Y1JF$zQ^rr*8h$B& z!8~i?tfuoq?}CX@zJc0T{`4RBCdQK5J%Kq)F}IJ1Y{XeEa%Z+^lBr#}liKdQggYPE z^oTXJ<$ifym_FS}2?B1a8vqRCHddj~@rfPlDLzrH2KJGu<-Q&WZuNEmG)jLwXzDl$ zER_}N0fG1&sJQh;#6YPcNJ<>|1=i@;RP~sosuRLG1#TsInM~9p7c}3#Ujrp;7r&`- zUx&E}+2vm)x5P*t)-1CHt+8W6!{FLGYZneT0dZKqMF@iAYd@}B$Tk=ZVa;(o}IuW2lf7;vs zXhWD~9>BJhC;k4C%62qVv+gPNjTYXW1PzU5VzwJeX;7WNK?z}bIV(V}RmNtoR21Z% z!B^nvdKm?%W+fw#(v-1T;RZo~Tij4Nu|RAWsgR}sva5{#%0#SFtk3+_VNMs`omdsp z(JC3o>7Vx1wYWp#3uU2rx@|WDuqFpVPt$Xd*bG1AUKGsU1sW=h@A@x7JS%!DAU@bs z^e!HR{-cd$-UBo50osIQ98d@JSM*DU0ctJj62~kk`Yc)qK@&Z$oXa2iAD)n0sdA7T{LeY8O&bC>6ly-@r|H9l0u>$0{vg{vUoDW@>*jA4k0~ywd`Xhr+};V| zJ%8TiAUaUXekO=oC|?)?J)<#I1r@~d{+e@@3w;i}E4_aaX|$klr%vCJ@KDu@Cw9H8zldJ>pw1TIeWWpbmH~n#4)_jASvRQJuMIiZ>h!TrRf8X7hiRbv zaigyUH1c_w#C<(vo&@!v86c(t?xrd)+!dPx&t6?fZjt3CnXnt{l!1iS0Ilq9&@E*5A(vTMyKjDWE{ZVg@`t9ywwn3@Wi^*l>Io zL~8TKXC;2hhg#V_E&uXdU(@f%q;ehNuV0tO-OeQ7m%1U!9e`6AsEorZtaBP7<0rh# zox&dPo;G!Z$~~aL?gJA>a&Ercrx+xKj8NsS>Em9KCX>i#VD~75bBksgv^hXygBeJ`(?166;y9EE7jR6`!RiOz>w53X?mn!lxz^A@oOCEP z)MR%5brSY62LN%6fc~C(5ue`Wjx$aNU?jFDfb}0`hHYSlyURSh5t}@WUKDO(rKvGI(2NbVROf&j4}`aDRKe92Ak)hnO^l~_6yNRw>S5>MQ2UQ|SaaIx zkr8TE)Ma#4Rh2~lJF6Ac4bZm$VG^LM;SXL;|EOU@Z3G`aH3OW$<=C{I;}D_eAV0vi z4h_}Aq^a*w0Z=8dLtpYC?%-O|6AE?lY`z4N-AzIKR%jAIEWVYkgTW56Kz|Eh-8BDq zs#Hx{*OZ9NiXdv3;TXUzbbaj)Ht|TgK>_^qTfXrUE}!G}3-wBn0r3JDrEyZibp_I+ zdv}6%?ux)xz!9vrP1dl3iY*~qMpc&ylG}52&@KT&ODgM%UX&>|xzHtTRk^aTPj_qxj;q z(Ivsq$-umma~w!pZox~ixK19I-t}>asP`lAwx4F|q)0JSJ^-rQWy$zLNc;#VD3uDt z#ntQMQgKYYr^^#mXt+lmvj}>U{Y~ax&E)-2eJsn%4dFcsp_d>(} zZ0@}Nlj)7vs)RX%z;*`31{1f_YqPg=MSK6ddOH{E@8yjgYX`@G1Q%z9)a)do1RHeA zOD$Lq#Uv?pcJKS!vLMw~iF4uYxG-Ji{K~FD%hpTTdD8cLQ`2sUrX}AY!8w|SCS08iof&*7tYN@YOgM>E8vh9a_p!2 z3Ihy+cSEHK>80T$t&yyMB=um!*MtL2UWvA@W!Hd7lJ837ck`|#^!GcDyb(1h8ml?B?bWJb!MQ1F+h zM60hf7yT;b38_$EmAv0`Oo8pVp-Kw?Rtt6J(AhAk(=P`AgaRm?=Z#JNam+_?`L4~L z5db1uf80srg|D<_0g$~!1yz0sNUWDwb0m`aF8Auk(9G25PO4KhX^8`nREc2|x-Dm` zm@#W>W7CJmLR~5l#M@(@CG0IdSt0>g=#1v+qQDRWu74t*122^5dS=>s58%8tohAws z00@{#<}F?gEldn0&Z^xI1`=)1OgfcxjEU%+MFBR66hQ)glQI`Qe2=gDB-{qCaJHFB`Wu4uDX5{~MX|;h;?i0mh^VdgR zOIKyb!~m$wg2EDbx@l&YJ8u`*SB<#uP{hSuNFR1jHR7tP2Ov-Da?BPk!yc?~MA zLhP_nc$k7uRc21FpN7u-c&Ggvn<~kErydLKrRIjocGM8)&U=)G!WE+`%$1i*$VLC! z{}q#r^aOG1@fA&rZ(AVR-RWb4HcG~Mp z$c*0xmjH6*9blA~QMKUcX^ELs5P-iepmk~C5tIA2>_TRZ0??Yd(_i!cc|8oBxM5A` z0aud^2RDI!*Ia}n9gSLwb_BLwc)P`rH)EegF4_8d6ZBDPJz3w0;iT%bv&U`M5NPfP zc$|wyZ9dOhOk<*lDlL#sn*sA-l4#yfROnei>ZjNuV~J2y#XT3C*O{^mR}54wv)f8^ zFSX}En^S~w0Po+94{U(k2vE-y-T-RSjs4&AQWQ@O*hZ%-K|w)vf$w$`%To?FL(K(! z%-3ogXfb+L5&cW1pdJk!XiENO%!gjpiyCzhB1~jvh<~ul@MA70y>->d)7|T7<|qCY zt?|pC5penN{(l^Fr5^xQ%{WTC9sg)T1Fyl;eHbQ5Pou)n?7%BJxb&+Kb>WW<y%r%J8dja=1f*g`TCdEhR9*KCzz@lk! zQY*r|ZOxokE<$%gfAqkx;KWtID=Qe`}l(ElGQhbF}DlL`la2UU)>uU%ANR%OfM$<^}&r{O!&snm%_xQUKI&Lua(g?!Y? zHkQ=l50Fu6Qt*@7RL9j^vbnFB_3OTYXx}ef$ORrUlO8x7Khoi{)S_iVYPjce$yiQ$ zklOnEE{j~`Xq@ODY0vM^$sO!)@dzzk_XRQa(4CNa|vKh)ZCS#kK)y8CV zGtpg#<*QD1Gtu2Bl1<#P7G;PR|7&32wuU-evQui=FQNZ$$oVr$r{hlDe)xX?;nB~1 diff --git a/_site/test_style-guide_files/figure-html5/unnamed-chunk-8-2.png b/_site/test_style-guide_files/figure-html5/unnamed-chunk-8-2.png index e25ec3ca293047094ae7267bfcc8c44cfcf4690a..9526d69841a677d73ffaf87088c895d0af72c540 100644 GIT binary patch literal 7993 zcmeHM2{e>#-`AkAd!n*OmXtzCh|oevC|R;swzA&IzBVb|@>rtsNK#tJWb8A;SO$+Y z2@hgoW*Esjmh8-6zU!{%eV_9_=Xt;Lo%4O)Iq!GQcc06B&HcZY|F!)7|Lga^+&yP$ zBFHDn$HBoNXl80;&B4J%=iuPv;Nb+4>V$DikbOLN*2Wm*92~kF99|r{x*WRToyOti z#o>hoB3J-FUb=upm*=KoNzF_2R&K>0;Bous;gmLhw-Fate5}v1~RaO_!49MZtoUMge6&3_om8 zNMW-<7oY~4f@K5I_v%1)F!%5mFdJUT)ZqpPhfE^;wyfubjq|~Cj-&?jczqTFsH)>j8TRJ^VJNFZD`?hl1>_1Eu(=$>*F4fcJtA(o%=`b@|1ZIxP_U#2aXyoyDZgA$3cX6H_m?Yb zm+D$V5Gqj5IBS)l;};Rg+kei2uUxQ~Y0%|LQS+OrK65di*oiX!ygkKrn@%tJ{o}!d zG4jyc;x*r;{jW{9QkHn7^_8HvR`l_GDEAnIi50IjLsTd74E7mxp)}M6tnx&v9=+SeUF?~C9=zFrbF=`aE;a_jLL&UauKIf5^xLxbMrM&irUk|82@ly!x z!!-4r^-_ke4&6(F8>7DvLfvQy?sxy{Jw1m&Wn0jH6*s2e%*9*O^84?Zsn)BDvjlbp+?Jmj z36@U&ukrJjp!#nJaf>pC_dOLXaX^GENo^Nu{0YB5&JaJ(ou-Z=tdrxD)U`J~ZVHOx zvo{@wnOse?_MK2lwM!ai)Bc2;n`->H&3+E z_Ry@q5zphFAXcW$c2wY!kp{L6tf3XHq`gxo-B2%}quevlKJSqE*nslCH75A*O5bVS z3uV~HH}uGKWi-%Tv&-My5yXn=@MWH-7q(Tt+UR8LB3N9`4FYWuVh$A zPqZ)Qn}`)V^5~MX#j{sGBf_van{s9ebpxoHao@#w$y@Q@0=;g^?hfm>aC*fgrRy}U z=mYBsLrST2AG8!AZ`*k~@LlDgmpv0(edZ;Rc!C6Wy|0C~|5RmydcW#%u(1)`Gk#@* zXQ|&JC=$l7IXKW&GG5GWUe$~=Fu*rOoBc2vUx2&2f1EL%yYQ5jJR!{eZ=@|wdjJfc zbK+q8$wJOdG3Ijp>s(;WATde+{{QN9vOA4pEq|?kZhu;XLwlQ!+*>fgP^Jz2=FT8K z>Y_o%bbx61LjbmP?tmc8+U$u2$5ssgN>C0r>=i8NmTqp;>A=;Ro^{#G%A8#jDnQjs z9%??ir9E_DAYww9G5wNH`hmv+Dtp9#wR_y6GLcCc?>lfkjAmODYEACD4V(t)HK%rj z{n%8T_H5^R2>Zm+=V3i6wrbf^ler>k1%NLnB}C2iS(u)iKZNTh_h`|IUD&Prfde9W zajs00=$(na_2@HHM{oWaQ!r&vJB}8uJG5BOCG@%At=ghXE|>wA5RW-9N?(QWmH4IO3p8 zWo2))90Rl>C2-nzH0t@e@~>ROjck8N19l5!VfY8-x)BTu$0{MP6n$b_JpcTTVw*4OHT)Kr- zc$L!*%uxmE1X|T1Edv|I@XW*Kj!j}N z>?BpNK^_b7)5@&2Uj=gmLr!l4>J_upt_5ih$R?leXhBk*s>2R!PYu`Kz25K6@As^K zHQ0=N$VDXW7J!_8xLC-}SrOVksS52+FEi z0%Jz6Tt8-lB=5(El&V-OFL0_wHreWv%oKd}BToNheXG^TGO{A0KjVpJ?rE#1B1RA7 zpdL$~K&_!jYf)6ZqP6ol3+#Ltnf*a#>E#{-((J2l`sMkE_05iD!`_U?JkpByfxYeO z$XLD>8bSMZog#57>U&EKKI)g?sjswV)m@IRsIDJX-RaAdWgGsQk#@=$7g$B+-QyzJ zg4C}Ys)MUlBby_~AZ6QKUaZ$kE-xuL&!){SoyYn2`9+p&v?of{O)rK}d`(7CXf66o zuHmOhxr>28R|3)y&>7tzx?2T!2V{nizm{2E`@*B&)sFU+yCo=V}blb5`}0g*vJ zFqyyh1ZqkojJZwM)HdBxPiYX0%Q0N`w5Q+9R;)Vvb-eG@np#|Sb!+uwT_~jQas0}H z{#K-)YaxSfu%SVVmNHmx@wZQ*`e?D&QrX8=$yY=%v3onp3yvcqT*jj!x)#i0wkmWm z=r^}(@HyL7yx8z2h@sF)Vb^s6lwen*0$u3-(8ft#_IL{)O#IY4q&P=6UwME&BUE-Y z(=&KT*5CW+0AecHQZE{>x@Jyrup$>2>db1Ex9ejbF5*4 zp9pX~ie_}oqJIu-;0L@X0<;F>f)GelxmDIKlv^41a-}>ft5%;JTI?F=%)UO2>w8!?J$iEG&HKm1 zPHQ_9A(GWl7z)lBV!pgD!o=JRu@W#gJs%&Y(HFQUV6|qtS=nHRgQ$+<8O`+PP5WAv zO@=o*uP#PMLy6{$XqB>2jctw}t{w0}Mjx0TJ?Yj@`te)CyI^P0OUb0k*N1~o4qjDg zbNG}>h|ke(abcT>(c`x$nUy6mTgPWzEF9|gfDQW!dglc2GCT5lQv@f-W&Pe)v8n$L z3iD76a%1c}J}|iUsDK93%R*vIG}N+YSgrIOcu{1$>DM{tN^aF*(A!I1XZ;H!-#WwB zDU3+$3soTs%$(ARJvWkb1927ul{ua@7hr87_}WoB%7B*k5R9SN##Wyjn0UhR!TT_H zfq(6pPzQVvrct5iQ6*12I2zB0;4>O79h4=0D`hf0$?VY&-_dt+&<9=gw0wPa6_>BJ z$Zc&ViYKApohzD_L4i}uJg>AMqxsUOA+5wbKI!dBv|zK`IEm!b?C`-QLvjjGb~c>% z?==6R3e>Ft-CzTHXewtxpX40^W@CDw?{+2v`SOt>wDj|cVGpN7yZ>yjBRmJq?DtxU z-bcV`q}Uk^In@X@!FFn9I5g6~rzMx+WGXv2VgqAXWI8VtW>m=;J5q)EH}Np6b3Deq z`0(BNnN}@Ap^h`mA9hjQZ=#>&8ZiHDJ!lBoO)~3$cmWt3No`Taz|b{}WUGv9f?3rx zw)Qp5J?~m-1P48z9qO{`HQ%i2$pAAp1^h@opHO`&vMcD;B~(1e=st%bDLr<6=B~Bn zsUc58+CGRZL-LIZwxVBb`;aS#rC)KMU=8-I)zg*`V_Yd7c5JTu17AGXHW?Y`>1u<9 zk?+)0vVZ#`Y7~?#_DU8Ng4 z)R&c|UanJ|wX>6hYI|`+i&%*bn!d#QM7SF@(Fq@5C_ovi)reW-EyfJhv_bO^VBR*~ zsoU5YRA47o zUr*JW4|ZOFJ7-yLU4b+Y=lm8HzGV3Z2S;j_D73edgjS&2!ZqCVM?gj3&@a*aC^~~Y z`C0(}+zTz~BZ_Czs?}AX@}Ci7?aRmY;MnXu9Yct#aIt|Str@?-H(pDZtL(HDw}0=H zRpzu~=F~8%yD-pPDg+1HB+yhW`sSWiqY1I+A3XWK?%lwJ+bI_+Vqz(*Z-*SUxmfzqzpFE zBl56EI8U$BmstN6wbA^Qb}3-BjHUrg|Cw>XlFM|=LWVep1YoA~{LV6!&9T=ikAxMs-Q5Z=i5R~5IsTpdez09I?H>`tm)ACYBd@u; zg>nnw;T38P?b0cwi%Uc0H3<8)xm+aM(XX*b_-GOzRgOSxJG)}895+=hB!qVWwt6qL zg6+2=4~5LB3AGf#hq42AfRAHU1U=NCf_3?aM*a+T#m8*o35Osdstn+ezoQPO&Kkl` zN(+c>zZ2;{hr*XKMqf1JHja&Q35|bfjuxlH?imA7GnDLasQfqIxh;$~Wz|2RJD59e zO_zKAZVauSTk##h+tA-5K=t}io`SWxNs@_bF)_!WwoD(}9iWS^TkqX1k8!qFO@M;8-`MI#{e3()1<*bOm$sgV@G01W%f zEhZ3)hO4jVzsoxrh(GHggFtSKo*X=4LntuU=B-HL)9)hepF9jgv)?&5Pz;<;2diW1dQc*+!_X-YxVmSh&z=u%P))`89qG9D@6OYqi zi~6==M#PJ5fOa4{gRdsOv-5bwqW@3PDhTAjULYNT z!fg6m`MqmvB+i0boxIMO=2MX<7nP}=JCBJWRlj1yKEG;obt(9-6|@>vJ6|F5Pg!46 zSC7f`x{2bC(ffx$KhuW?L>)BMQU-VE8$3DF)g0$(dv%aY2>Y7u?hXd!XWrngePNr< z=bb2u4l3S%8$Ma|uRpG5y?@6m?PpPSs=yl5e|jbH+$(vg#F&Y70G{Vx+IKUz&|b4K z!3i+K{$+J3Q}jYYa4oSUZ(7|cHRXg<7s)O4} z7R4Z33e;hA-)QJHp)=J!OSLl>olBOs;p_PS{i}11{{kGt|JsFX(UmebbWnVKb-W3i zYnop-CFB~wIvVUTUmASnJkDbf!{>o5-_OD*7RXLoB z+py7HoF>Vgq1SzMdzLd;&u-h$j?j*)->8|+FT;?~ClsIs;jVA>8TwmLth$VbJ1k19 z2CYxZh?kld7~fFKLxG~iW^o_+wcpr{$V17()@9mHi81!IugY(ni@Cm7!3^vuro`&H z95rJaP+l^*CF=@=6{o!1ZadN#k?J#xwllLhA6C&e`~-5MY3jz}sX_~X6bqBl9O|?Y zY20@kJ?MV|5r$u_d2;*nZ9V6j*5!Gj<@||9OL?A&3-8;I}P> zJuRmRf%|aWy(zWHS-e;t|EG@EPWE^HR+QSl3(Crt?Nwd@@3^?4Dyu z{cl66cZlM($~M(~PAy^4CuJ0&?w(&WAhkJ}9%D0ZzuF<+~ zRjBV5kGQ@O>cCo*+Btf=p{nw3*SzXL5N!&Fjuxj8N4{u&JebfdrFxn^c{nKP)SN?Z z=rOXm!ular%vL~+^v~+B6zVRBWhOWCQol5b45iDWO}Ug z;S29>g4Ia9K_=%I#*qVDK>gv>|8f2N&lfTOu3rkJfmT~rHYe#(bQV|2uUsj?|H*rF z?s1Le`Ztyb8$?#oi;JrVPhDi#p-L=@IMxzp#UlN z3Q*ZJnU6J3HfTdrRx6u>#~`=S!IFicx1E*76LdR-$(3ZHi}@Z{IUps2=-tqb=Bc+C8U|8TDGYrfB6tScf)k`E}eO zmfNZtN}<$?-oA2J*n3S^S~K)4qu6riZglL`>igQXP`#|cxrQ(v>AOcbR~Dw8t6c3l z)x)_Ekdq%UEAOE^Ht4*;P9-GC-lp(^DhTZk@&TyF?PG`Zfn67;?%JiET{-V~G46+>56dW2N=i-1ucg38(c(N@FKlg4z(#2I=NSWLUYTfW<88yBrb z66qI{iT5v}-63;|&(g#v7goVrQ@ZKYJ7SMsu9$+Pnqf(ay?>pIWPbBK&@e!N+8hEc z_1#%O#_Gtvj?|&3=eSCE%Wz>98=!ej+>Jt^ECOxz4B81w)@=Bx}QPw8m#+hr7r%q*$0;MQ~M?uwEYVW z)8vJQjn(xIrf^hJIe%39;S#>TQ!P$F7;*vZL4t_5V z&JG;Ho$x0eSJ>Iv@9hh$_-}u+(Z|nMbW}U20O!LqCzaQ(0!kj$YznXwy5+*%Q z@+zhGN$-Znt5e5~)5pEo;__wV@b57$eq0uRilx{+`uuF(%O}MqrL~<)gOQW2L)|tL z-NgRItQx`H40x=RS!@GMeR(j?sC#%fWx`Tgy<=yU9-NoH+1xzDp`wqS-z)?9Iy?-jV zAuQm`Lme?z#x(Esw;d064bd;ChKlhPRV4Vs}+0ClO8TZ-&A^U#jQ}R%j9jx zy=#7Xr(H;Ub`&p!Fk1(Ic|a1Q^tIFWVKzG^uaaa#|2by&G(C>i%MuT3qG{o`dP+_u zN59#ork!i7U`cKcGbHJ`=V13ru%>F_(rN*xP9%|?t9<01!Tr=Yy?xAV&m6zgQ$Lf& zFh8vIQ8D7Fo_V-fJV zLie>J_t?HWZMQnVY_Ad(M%ee5kAj%en}I{QVEV#4H19T{%-VJOLlC%H@}vg31z1~DOW zoiP{G+t>sdZ+Zy5e)IaK)U|J9_y5{`gTvAOT>2p&3aytrtDjliOn1q+Ly>ATW}cDv z>5GWs<=~yCoRj(|&+d>h@BfSbipUW<53`q!e~r&4*}F;!ea?2TvY9p&B+zx%apn>x z6?tuW??{hVncUcPv-NOp?Y=&5OjLeB_=xA5_h0*UiL%cylkXX*h-Dmo@$Kc&#-x1( zAw~l@&z#QizP7lszDCh;J5k>zW+(6JDS_|mg?#&aGj@DjF{JWVBgvnTgjVzp4lW-4FF(8!i#2CYWDeZ&$LJT`uZrT-`)vCJtthQHE^hHr zD>CV=52F=@PYC_@iC?z)zrFg;N`5hi|NP^(=<{DJ`7et8l8pY#AKTLKfBECT{PAD@ z0O0yx=KL>n{$0BK7Zd+C#zgmJzTnJn`Xr&8dY8SA@}0TtefYqkQ!?JtJnbhw7n)H; zn7_a@r>j^Tk+dcJy~c(3DQZVVSTT?xV*Bl*!))ETyu71MV>yipSDuJPTP=*WH)iXU zk4=vk^{@vT=GF0dt*@w+?J!K*W>ctr>NV}Sy}K&&w7y- zZZa|VYc2odsX+%w?xcIm``*raJ{az>Je-_&Hlf?9RqK}4QOgQk;;8A==BGQ7tx z`%xcL(Q&b}9c$zCY+mCUekwVLwheKicZYR@g?zZ$EZ8ir>!j*sV@ZWi|aEhn4Z7{<)@Wev>=f z$}|eNuFlu6WP8J&33o&d-j%ApvSo|l^%B`=aYAP2uh^8hoJ*`cB!8`Vu6Q^pqHINZ z&r#mf{I=gn9?-ZtRIWcdbn3D?pY1#Uh{N zeSLDR{;naW1^c>xwn0SaYZ!6*U*z4Zi~WB6=gOC)zvvelyIS`fz^LQE#JG#(jLYua!Rf?1TIm+B-k)#Q!Q(VH(e<(q zzrWfqo3JV|R&%6l0Rn>f=%-AUnFOE7EZ@&@VqV4bpEQtf6Mvn)yXAOOZsWx`&s_O5 zJXkL<>78A+Z0x=><;9~gO-cIkU0BU& zOitzP*}=+p0rYxrA0B9*Q`ndl**NsQ$IWdlx0andJHS};I4zFKdAT*Jx^Qldl!KVY z`w;Rt@_Y7ev9(wOqSbzyjfdbl9E0^yRLDnaIm!h;=VS6j_2hA~uEhvX!^s##Sx37p zw40ukjb9^q(* zzDK;v-{SQKKZIF*Jin{*T#vKD8x0+&bPgO9r|i<;tznZz6~X{6r&ZtEHk$^a{_>5L z@(_mwuI$yl=Q>F=EK2RX!Si?Sr!INB+Rdtkdr?;E`)5|gy!;Czf9*cRypRpynXKcs zc>)3b{e8#6eJ<*B&nB?vGu1^4IoYV1C#fxKk4RpsiT3CVI4Yw;zODq&wZuFxQ@(d{KvhQHU32}9h z_}!~wd3JM|ck-H;t1CWDOSQcEm~Nbb}| zwe8-r9$dVQ0}`_t$+oWuDBhj|&y9(aTKo=xz!!JN&YOHh)$GS#7D*~X*fcIX4MCw{ z@Y2fiXg0kNw$rFC;+!8TnMGrq*usnO^lw{JvSH-8& z0fP^r2i8`;`}rOdFz`B)AZLYguBJ^Lx0+@z1aCjs?K4CyDT8t}`hy&?yENCUIB(B< z{f14H@;=NN*Jp|_M?tH$0G5c6O{4I`Iz6{(+v$}CZj(qGZl z6&hUVuqZF?^JM93R=AeQSi3c$$mZc=J(D)0q092zBMtNC48*xEtJBqTMdn?s%KrCL zsa@t8a|VJnK4sib9jAk1Ab&4k62T-_sC66j7sE*b_Y)+4ca=MkfF!K$&iU79fDuJzPXf|7hAGBKmO9RRQs_8y zI?0rFGe4o^GV=GCFZ*(zs_)e<8t*y&#JmP!fx)^6l`y)m$A6C%@C7NC!B-=vn?ff7 zv8!g7X^-Y0TQ!A|Y5S#t5W@$_UovRG&~nKKc(=F;KnSuE&&C>K$rN9OyM_{^q8A?IZ4hV^t)xDgAPo&gQHnbaZ)-q6bl}p6L_}3s4 zhq%C0lI`EL3Xy6iuTIjuDM@-kGjx~v?Q0?T^OHOd%YQbAFbgi?t=|4EUs(Z4oVBd> zJ(hR26D)E1!`TMb*7l^UVF#0=?A4X4M$)sXGc}5|vu2K7-3D6y%h4LEh9Yfbcy8K$ z%23PMSYHuWud@ZHuOvK*fS=QNfl9(qps&dKB6G+Wob9V7A+sWlpKSXx1K`*_@zq*%c|IG(OYDEN1t8Ii z#+DH5InOpZnr^qnRp2*U;QUqUC@O}xy(xZ;hA2)}Np|+916!*oLd85dcWxb=A3wsXUIwJu>AL8 zKaKRi6Z_3X{%2F1PhVP*(Li0}Wr_Yu_1y;x1 zA=sR%=NpQ2*jW2c;m9%m4e+X_>(W3@CAkLcjElgg4_C|(Izv{Ph74T|yo;_6_osKR zJzN&u&bK{nS;4*1#CoAPsl_vy9cu8a4QGN*bRppaRL7a^ydb96T_&qGx@f=o^H4!S9_m0j{UWh!C74e#Ee4m_QE?f`uJ%!sKYU?Fq3LyE{xWaB z_u0Or2y@TwQYXWnwzXEbHI?^;S1J)?jh!RFh6qBvXPUraFE1%{5tZ$1>iG|MA3FIK zK!3Q)(vvbANt){levcDg2Exy0zdJu~Kp1#};WJS58s;6gUgQc4p0UJ9&siOm;s5P?}FuLGYJaMu!2bnC%;79Pt9 zUs6*;G$lQTUOPpjv!;SQG=8J>sp@KPoELjdG#pQ-DR)w0PJ+lDRH__4XRi5)${e#) z`ooXN8T|2#_lvAr^0aEGuO>sunln?w!r^b%qNL)}vpRTg+Rts3KAa14&<%=R`n}_i zVXI}+@b(KYURN&2U%a5-)F@69wYKU62Hf9tO(b*d#tdYiSXNj_P2U;gw^)4aGwjog z4-ABeWrXIN6ZIRqua)*=Wt#&h0Pc^qi){K>EHj=$vW$-uM=6)U34-RrPU~6*>ax+S zmsNmoBL=SM9IgvHYxroX^a^a?$ic(h#)|eMb2jO@Onb`#XO7dewVigH`&L7Jk_N1` z<^tqFR`&SItkTy}mH!}gys(2kEw@H9oR=rDRolr7lRPDi%$h`1f_C0n>U^=#LK{6* zQi1G!u`vk+8bou9x(k`MJrn!-%nK!57XKv75+Ex&Hrgg|duFAR(%&D;Q(#I`VjO=lzkk08#;(vx^G9Q$GWHiz8I0k# z8@bdFEtHqx(a&NNDTvl1hsQJ(T-Ge(Ud)9}U}jU8$n7&9yJoMy_9qSFfOtO8nefUQr0}dLUDbt$9Yyd zk=O}hOI}HYMcM5{zkq;ENI;5D$zvL7raW-#8h)Xs%0x;<#s#% zrbR_3%50*1W8FgwSvR1uL;?hvSSMi(YQuW+Rp+`L=xcR&%OrQz!uGn8(N1~ktkPk7 z+J=cIAS_)^HFREI8tQ*6#r}Rq3p`b-j$;Qh5NR-w&a9LLR^{GEoBH?NQVp=Ya*BxT zsVeDnU8HVnqdI!X1$5SDLOiisY*khxen!v08jtlsv|G2%f7P zE81V>lgClm#Kp$+EO=d8#;e|0B$4bIgxEpTG_pF|;KxzEdS>Mg zDwPulL)>8rEeuofG;55A#pWGJiPwb)gD7!*uXhPcUnseaNhixUT=yI~bD|r#xeU+h zqc&F)j;AA}Y#a}m(yU}e$#fQ2Zsj@s*G;c?wgi}$&VMo$YeGIxWBTIOWHiB`>(o7( zo3rC3>&v6`38T)YcIRlOgtYl(D{~dL&HmI zZF#=73?F#Epg!PPH8puwP^3G9hsr}P9qXPl7y31yD94vBB4)*3?gzM*t$jTuvb3&& zdI_NSkomh7n*r#VWty7Im1vtbBNYKG&j(8hAGStcIDRp?BS&*0e&& zd3wH8ha)$I)L)9m8CA_@A1f|OezfJU-Z7bam>m7A!v`;r{F<`3dM+aLxg+Xfspm}Ya)}_nq=5XZiQY$ zkrlO02xMx8L2vd6WEk+o;Tvb}(%h~2SGcT)lCoMNW6M<`+%$nieofm(F)OHqE)n7F zqvBz$@}r@_BtHL?sUaS2ams~PvULB6O-v=b#~p}MmQQMiQM|(mPjdGsV|*I}A(ZD< zgq(-ktmdu-up}&1n4p}acuWrdWp_*)<2))}S@ZJe+wVpZ5POjhN%c!yA z&o1D90(ydXqPRb|nuzb|lTDw@B$w!+PJWt#;;P_c@4Zb9Wij}!C6qu9Q50B-YKA;0{-`SH_LrKmWe0Ye+L;H695D8+1L<}>yVc7s*l-_1c!f{t;geXSL@#5q^#rmpv zFTe&bDvJ_M?8hFib3HU0i`8wwkgzylkLF;$O#yX51w@o~LS-bDXc&`w`)3U(s|Mh) zM)LhXwV13joq)WiFnAu3)9Vb>z>D1)?d^NnKDNIvG9rMl|oDWTUgey6Koo;$v zS4{HQv2)KmR}mpp7!N1MR6!>~E9}irj>Y)-9cs&h)tw)m@e)s-2ChhAcG48WTf1zG zIE@QegqI;8X{#pbbNV#?fOaYTMAjWrm9$(ocl-WsK>(m?KKlW{O175h*9PZ687;0} z{M6DpaewfSoI;XP04HCtfHmQD#~zXp+Ks@aDSI?P>`FQDV?olkQOYq+gJFh!GqoIk zq|Pe<|D)#rdO4DoE}4D6c=M>aFJB*P%Td$nD6-H=bwg@ypA!wt&_RFUpi_ya&j{hG0uEbO-t%d^QxG>$H5+=C-j` z9#QwhE>i+U)n$(-F3+bgoU%IZ2Du>W>fe2-SY_gGxGL!!eBrk^6?3sb!gej{IP(*R@REhSNJpUEVt?T=}%Jl>X;f>{CxQOHISBMLfS1LhL$8t zyQFpMca8A27r9LHu2L9iQUwWJ65jM9+2fW_JX{8wETQpBEH} zVj~2tkgN%29DcTzR3Ae#?6GZ$_?-%(1FAsP=FKK7b*4SWK1d8yFECaLdh?ZG^P9-< zFk($8Y=G604HG}*sy0F`P`zqjwd>{(^lHfomz!eNNx>B#<{N!HVi+lBRq~So19%6o z6MteqybckxN|;zSP;1(ve60{WNM-OV8`dB z%OmLlt{jPE@)6Ke7C&lr$(ve;U&F;ot)c&K!ZUV(s5!e-y44yFUs5a<8Voro5QB}9 z-`lm{_lXQYCRIr){o@+IqK27c&Qq`chE4!c!$s8z8HJ#S32ieGAZIh0bcs{HNBj)2 zs>--Ry}?TLR``UZO0;#ao?szY9@|(Q5(!aVfDS}|O1y8);AlH!J0u0r(si}?Q5dBW zCQw^IYBn2#P}G>ECGe^!t@pjukI2m`FN$Lxz-oVx_E{W(JPDh$Wt3xK;(kHsgxT;k z&VL`}1UCH})Pl25|Z-TKj^I%f2MQzWs#RSB#i$ zUudlqLu&CVG`ZG-ba6ZDoTtGlfI>+qj zWoz-VtNf`$9okOgWI!6b!jA1~H@4}~Pf^L%=2@Ks2>QtGliJWri}tN+VuJMXI7BRw zu-V)lUO*h$`h5V7JiokoP>tjR0KlzHOQx&ZzqUa=o6;q!zJ&yl1T=eDi#>=z)(Ic_ zYKOo{^cGTtGC#@x?H)g}9iup|*Xh81rItUm%^!m_f+bAqDcwI?_0JCjD82crC`D`* z7e5o{=BY?thHt*f?&0|}V*L9nfe0M(X)Bo_4Dole{rd;TBjEmzpOZ3Q*!nlLvc3@c z2ON|Nwrn(tg*U+`BnRbF{vcuDgI0LZ|KAsjb$#D_No+F+R}o$f=VC(g6NiByl|E>( z`$5~G8W?(VA;-CgP%z|IN7f6tny#EHfZ5j@d5IYm9LFk=mAg63Ecn> zbvsa>h(qb82bjO{h25w}chS9Hx2Zr5c9uum&J2vD;tXLzwE4=!ON&pVnKDuQC}kS@ zp(a$4OcW)`@10JQrPICl_E#ZN9kD@>x6F>zfBmiH5Eud{@<^-U$XSE8fA_yP8^jQs zVAh$c_JqY_568xIsc`g>@E1yQw^k*ooclj3#_cdO=ERB(OkdU*?V zJ_1}Pgg^w~^}*x^JpcRKjlJ$D$=Z}*QK2+9+;qz6Y6Op2fFTu-|47fn{^DDls^S#e zu{n?*+$_BY6$&*RiP}v*h%KVD``zbkm|SrEu|DUz4|@@pk<~}tPhJNa#GBMx>Y^M$ zFVUGF<=~IgO34I+ZE+qLYKWnMXn(T=bPr_QAW<^QbwP0QsMXs8IdB?ECEp>zib1i$ zqTTW19j^(^E25r%VYlPx;+Oqys;N!TxnsrpHwVi5@&4uj1!|k^)#X-A8)?L#sj)!^ zBeNJ1eczoO`&3huhj&b&-blQ-pD$2d{G*&9cWqFKHbX99BcGHHJFhsb$h3hP7 zJ-snfj08qW9>JZoqbwZ1`ghG0uw2UQL5TD$j`xpRzdN1-7=}HEXZX!n?D>1Sf=ICi zWDYxL`8ia`?Qth(Q@>g1JY7kl#GLgDSMg5((mw$K$cMt_LntRFv^Fz#B1n^HEPog; zm|WLkZN8Po&ig9-wkoyz!gz<_1B&~#Duo#r)?ZHt3CaAxCqLtL1h@dVcB68px)`Nn zxcrTjPuR34$~30X>knqjj7Dq@t%W=3q7D#X{Q#jd=RA`T{uf#OvR|!*oI6{;+fMlh zbM&{hN&a}<%Rr^*-P_Ca&iFL6g(X|QzAPOCC5S?*-c1OA!0=||V%;k`i{5Dju;Klk0QNbcU;mFx)Kyj1Dk0mX7?rd@_|ThEKF@F6s%FDJ`H=~f@R^+7lRd8!q}tm<1kOnc0n zi0ZX&4+L4k-W$Gb@_#n>MH#)NCkbNF5mx{ajU2Va7tXSpO#;fy0aLANPF3jlpwND& zY$MZwc{c(^MgjT_jRP$dgle&+otL!rx9|ID9?Dge(N1Pg$#s;47*R+Dsfh`woA zzh}s4QPxlz%R^sD#_$oy`@^TNi}IZ6ytrS8G`nO*G9ktgX&|^*UEu*uqh?}Sdxn1L zRmWBlA`Mq6VhkVz*HJ?&J&U(vak`)7t`!qwh+|PHv}aL%@D#-Kzf6xWG-{;V?@oCN97j6DG|KMNqd5s1MlvkKL;6gf#jZf% zuSqStUIHl}C2-64vJa)pce9Rv9PXbV7}((f7bsFda6;U2lI`=ya<>hSg${V#Up*ee zz3f@gHF+DSF>LvBw2E*hf2sY&AoZz+b^|>d?G)lj57|QU6TlKV0^+n^XtR8?S1(9O zIW^4K_zwc!*vEKc#1+z!yqt(Lc7Vg_M{pRZEbB)~5*JyTA`KuBzf-R3051s%?TI*m z@n{CxUr{z=Q^@p3o!|B6J1j8+^#HEUm_zxx2u%@|t{S9%1R6U^NCvbRsABI2e((`u zF#uR83|kv3Q~}~V3l+Fx#MOjQENH_3{(?qR_Vg$43#d(SjwrnH@CkJt*j@W(ugjP_1AZ8P@K{4JtF-^8YQzuP;EVC z%lCJDy*8IDl4(1|c7qG@7wYsNtkVxQYu)|$^)W*cFOUILqh4EtS>Yr19lO|@-(-;< zSGg30gpV;U3u={mVv$XJ|AI$AO`9ebd@WWIw-{ReXAs6{(+|B*GUL>S!9kfr!-W1fvWWaV#M~#xnR8p@a&L&qX(Py&aW#n_7x2RvMX7Q zJ2{Dogd2lRFqJ9g%GOr@NSAr|Wes!sB`*OsF6O|(K?Y@YdFBs zw&fGojn{vf7PO)|?$P<8$>jtwMyQDj#yB0OdRYmB9pmjLyVqW}oBl(Vk))8}4clB+ zC!7EX=m$Q){0TnLhDnZ!15+-DTcd=g50IoP$YYVuC*1JBC^d8HRk2!y>Z)rVpC9}T zhKb5bPtfb757sC>9ei@?nlHQ73+VpPGy6&kwPR`aP?G2%sFY zcJ19Tj944$feNtChXS-%gPfZ6P~nwRw@wTHC3Exw_lTgzNs%ZK1E zR_?myEVt%zcr%kgjS|dNDt=X#C~+W`Zj3w&DL{RH-Lhsj^| z?L8jm(mT-J#<=ilfl3R>`w&?W&SOS7TswlCaC$Jp!a$)g(s|}Rr)6&iszs&Eb6=om zh#nzgRlu3J=C-VG6fgy`B|!5NJiu=&oLgef68`tu$v&AuRM#A;~Sw=*~;hzL3^jD-<= zsTW=uL)~x1#EhC~glR8Mn(3FXEa=Dy$>9V{2^-b>2y8v1% zfyX9O(Xvw_hj!iH=s57PXwVKzD7vkxkA)N*C+GtDOs5i|oetAH>BKV3GPJpJEmGOb zS7#0jIof_YJ60{SZdEdel}h|7pZ_8OM^7?U!+E5#P!KoIyz7OSwZK52!kMDEU00xF z@KjbI_*Y27$l(5FzlfmxRHnEewH5FdV$0UcD@@~3ZZyda^hh(M35cj&B7kP?n|M#d z2=}HprtrAe+^SqIXO0FAvt7_4wf0sxjvaK1)hMQF>}O?!x)Al?0Bl{}jnLFKuD53h z7HT+!HdbfV@~|{}y_WAiO*0dos@TD@?>3NtEQMx2Ln9Ekr`);n=b+W&1#nhBULts* zXbvoi9JMhnTSD6_~7W z0X?!hcS!)uUbv}HEF0r{z``>H?!~iDnsw5<^Yl#sdWnVY4Hd866MBH4%h>XbH zv71Zz7-0fC?*|=FPArhzQDh2PrO!b&5L&?D(lLc}brfpL26TIIWtg@ArPI>gFTW6F&Tak92NC+IfANYMQ&CW^o1zq z?0yiOhMqN)d49lXzx(3@Xj48$u&fOoJ8N)Nj&d5bZiWMBKM*fWbtYW-o^*O>QcwhN z)LQy4TXo!oA{2Rql5GKtsyyGb>tBONVZuw#$dy!X0X9=9wZ>x=mSy|w1AdJ}FA{+F zhnr+Qg79ZZjFJ zL-ZC@f`>(bJ!kI94HQX4vx zgJ^f8muZ6_K@rgd5uG(cG$_sx!q^Q|od@6La(EA#-W@!c0N8gV~l2}E& zB_?tS_U(uWPyp$Qm!)f*5pkm~_yy^Jn-~67B7XJ-&}NV>+LEb}Sns7I>lzV zWf6!tJezH@G%2pI=C_`8A0bJZEysPvT$jHHjTwW|B53lIdM+p7XpS@nIXXe5uAay^ z|4p*AU)~~%)K=2wG-|U!$IPt~EHYw#rxZbgAZCx_jofC>3$>R0u%qv)I+Gu9G@b6Q zNxHw6>M~Ku8VGI}Yx*dHge<6+kY>aI#3s4Du z;6bBnUhFbJuB}|vN!_8(?BpJ^HfCrAQX50!TM5*UvAMeyQ+H|di!JXDQ|=S&y|aV) zzMn58H#0UqjmhWiHK8xY7pK_T^G(~&pr*iBz~D!zu6_mDvuip2_mh)E&VXo`1h{P< zE<-UQqoiJgW%sa!U&eEcy+oJgMWheMIF?7jhy~}ZiO$8AY}bN!gUOYBsO7-A7`$#T zGpl(8-#A@#Su^ae#qfr#{ymHL%=1U3FBU0~2)Jk?A-xI9Q~N`3VN4Bfv7#6ECsq2< z$^}qx8(x%E@GDHq2!1?hBf*l9pySQNs6^FtnaYfG`1rrMh4z^dB7O~LE27-im)dRF z-PUJg==Y`~V35m^0#e@y>CqWTU7TgmK_4R?dqC~T7@={yk>Rsc>?*;@YtfP1XY3qO zjn!S|TkD7+)@dmE@iPurw!Fwkz`QNx{k}@mj?lk^+z(5n9hU|Dqe|Y#4Toyk{e7UN;qDM z32F|+;^%r;oDOyb$-zHRzD6$KnsF_NSMxhfyo^~y~NXX}fByseI7!i@3 zdk$Z(U+@SVqm?}hlRIAhDWpGl@J9sz+6c_V+x+lhLgWbB6*@+Via1WsNi|utRqGki1@MRT5{vRa1;o8Z?ITGPsV^aaBjTg9QN*> zb6vJg7ueA_8F}m|HnH}?B+N&OPUEMM-jV8(BR+JNJez2PX&5r7_uSUImE8LXO@R!V z?zCyB1VW%64Sqlas}jJg9whH=E1vnc5cpno`Qu6Q!`Wz5}>6<+BbO2J&b*_d9Ed?7g z$&5h^-PPZy+D&_i6@<}6sk8f_pI8jul6MgRXzgNqgR62Qy}4SA&7Q5L8qg%xWU3Iy zU!(S;F4V;DHPPhNUMV(}6XXBljvr+)EbdtbUhUC4Qm0B_9wP4aEe$9`bU`{*tG>Vb zm$>rt*BPbU2Z$;%!E+md!P$el{ZuqWc?%zv)Xj zU@C+aWYuy^Tbv9Pq~uJpi8!5TaNymAnNB`v{v8%^b&gJ!{aY>a^VQLLx_2VW6aW=W z92Be5(D41L{Pe;rn91^Nk`Z7-BhoMpa^psk;qQ*L`3!DJRN7}2erAq_;Jn!}j=i_f z^8o#2($btmC$Bz4qD7#34*^H+4UL{kARd+GW*Pl|Et`M501%`A_9j4<=M8nHkg|@; zd@*nragYyDNpIku3y#2M<1z%}gn*8W&{T2`M7zqRp%@wf(T`yKt_oT=dDlD7P)vya zVX-ZYvgK(*>kw-wQAv{UCFNF^oeC;=+BYXw5hg{v5S3ZJ92%g`0qZoxEr86g9C?zU z13QHUMX+J%a5M;)k+3!Qj>hpOfX2r_67tZNeY>UK5tSxSvBqKNE(9&dCbtCsd>92C zLEvbSn$7ci=dJS)(4Gvyf84$bBp&3eqo8EoFc0<|hjF{p^@6h>7NKf@8PCs3H=q5% zNMi5rhFUf2V^u!wMaL58gteb;MoIuscBjAYXtyCMbb)%OUC;g8`lJ{TTe^U29|8x< zHI6Q1QFhXSts~5+>QdWH%7CLnk_vzr>Gj&>>+ z?3(2o^9&nQ^?bN>p(;#kD^Ku&M;KnQaQef?MSV!&;}gJ&c)!-PK>8_7;NZCrUQGOz z#{c^jNHt;^(_>xd8pX-x`9Xlq{A=y}bH_NOU~ z+A@u}nFQd4*g3VK-DLId_=C4+8w3vlq`#HmapZ@J@Ap*{XrXsH`4%amz?G>Hhg-g} zHi>#7sz5*J#ryEo+)rlTyG&8OD36`hHl2qHYMmEC54u_uvB7D3Ec%7;P9_c=hUDwY(=%@S-9 zM{uG#&G)Z{o_?A#z~sI4)1Z0>OECk66wThgg%EcHN!yM>-IMrPdtQB~JA7V5z0mZ? z>|mY7{dfw&n>B$lF(ff7l+F-Sks-P_R zi^Df9yQbwU z0;xS^4Ed-tM7hzsE z<_KynAmwu<%)BG6sOrk8C`GFe&{FP)kRv6{t!S~$qHyuNP{yBtn)AIfpS0m~0Bm}Y z=+(z|m#ov!vJBR2eKQ?d5=K(hC;KVu0qMvz)9Eyi`4Bu zwcQR#7|)^n1G=d2=g`kcz`{)q{!ooK@HCxP{u6qAj$&79p;^ZVeMkt1W>kFt4XM9_ z%VC&ZiC@Ox5AVc%9=Wsq_-Cd3f@g;>R{AoSCOl0?dY<5mIOe-$ySJNmN)E~%nQ-q= zV}VKQo-b+XS)LoM-BA@7r52wZ8@Taww8h8Y5i>l#doZr=E^$zHa52YHco1=1(!eOC zv2Ie>V+|v7qk*i%1>?BE#~7vwa?rstO;C9<#?eiE?_6(a+Vi2p`-Ao5taHQAS;6nm zzjr#h4LKc+3G0b;st46Yv4lSYf5^NSftiX=N=kBGr@N+zMnJasny5Q|I0J6TXC$$% zIw-{&!0iJ44j+VxVH%4Hy{9w4DwNtBVC+i|AiQF#ya8PDKKkA>ME}T? z|8*I~P*G`{f@DyT1zava26ui;11b0x1Z10&IvQZ|>UmZaD6DiMcumiyyzK zaNm*FXyFDT+&)-=_xmTO40W24ulc#IE$jP5XF+K(4Q4L_P)muwj^z&z@+37{z}J+Q zw0Sj-y1l!TYMaJr8fez80QRU?1$u${QOi;chOn{Bd+eH%7NB`p*K8Z5P$1Y zgPx0+O9O8LS~q_2zWmIS<8?01nC9#1LJQC?@Pxa+@f<-XBitp$1oJ_ zIFjlB++*G>Qj{Q(Qa{*cn$0nbf9?P0*UT`XO#^pfHPLcxwtXVa^WIR(hOz~Qg%53Z zjJ{znkRs?jmfs{pd}##$H5+22Lj`YJN5ZWw=+ZQFaSC~g;dUz*EyMqOPMjB2`SxL4 z;puaX#x|Nolv!5j@{78w?`t6sCv|_-NP7lKMg2!F0u!Y4VO*j;-R%zxWkUvTJ)p|E!G; zF7*gfgMJDJx}h5Ep=p7QMHC;Q8)s;1sU@IZeLS5E{TKNwerv+eQS;#!2i|nc>f8$t z85QDp9)EE3Ak)$U&=i$u-0=B_XxC*-d_Fn02+4EjzG0X#wict)>a;IFVsH~K zCX?!ci4dO`7susnpzJeg=40NTn)0JW^D^#+Q^%uG3M$h|$eUDx%KWl>XTyYmTFik7 zlVI^@iG`CzJ+9L*8Nf9Gq=AH&YT)I!L7IbZYtWZa>TUu0ZsrdV)7_bfOA0RwQdZ;3 zTa%T>H=v6}F}g|@DF|xBl+;wDx~YQsVqL_rLLk3v?Bu9<3xJZGz$@jK(oB4tTr*(O z+%61zvGM72v+I9PVf~8>WG|r|O1%9Z4YAM_mRIUUyIs;y7d-Csmy`~LmaSaWir+B- z9^?t>j0L z7@mGNC!vp26d+0T1feT{f0rnlccM2Mq#M@#=+eWP zLZ77Fta%^K5J&oxoAH8DfxOM)55CWdd_NQ+?WHrQeR8|wPb7sgg;6lRrXyD_*DW>m z%vn$dr`fBF|EWOS4X?(Y>FNlK{@CJt`ykQT4V+BQ+!#TlH370 z4#-?tR3y?Hn76t9a(PcClF9-VkLRrGL-=pLHmPqrF7qXwYZv z(`^SVi^-|Kb>BG>lSVQyv4Sgl+a4aoj^d-Fc#YJGv9WO+ zG~OTWD!3&h4tkO7bn|)EgIil3FFHUZQK4ks!=q6P4Sot3<<-d{PzM;)hY+B>0XAn7@r?{dTJ1IfXyLdl!n`+xhL- zcX$Ua#U_MuwwF6!wMFsRziXY?&RM(D5r^)`@^H$At{Ku9^uXdqm^2>WlU4+euIVT`*CTe}RLwsM8vCt*a?bG6K5B zL3QK{FOf$!F-E@Sq#)p#vC-nTN(z01I-q#!HE}!y?e<7Zz(eA|!^HDatgDPSr|y> zvzpC+h-;rOiQ;8Rw^6t@L6YStxf{QIJ3S6e5gOlkAsM^I* z<1x(azP)H6m4mN*SbB1MUF`lE#ge3Se93LcEqe(H9lC|^(CxcRL=RyuH0$8D8!_*Q z9w0c;>)B?a`>Q75$38Q9!-@R&Qjp&MJE@;p;(wp{xrfv3WE=dnd3&$m`eHvypOt0P zAHvR3tWdIV5&HwPz?h4c>~w8x>+)bOn!^81?f+L80>SI~_qQ_Jf|tJ& z1h39p0XW-(mnSO(udm&Q1h!=c(Fq7%|GgC2%>Pd6r<1MfrTG*z!byo4{ie?AfLl+eeL3aDpI)hKe%gPm&oFIsOeI> z1zbU&i>?9~Gl3gTiJOOe%H1O2+P)k#>SX!hw?V_r>d<`^0v$mARW#zeNg386%FqJO z93F>sXjO1WaN~6@B)@@sMmdrC5GjqCo4>u-nRE2#(f(6uew7dtNEML68=Mju_nYcUtU`^dmOo8C(?WUNrWLAb1q-baqL>1!ymw#4GLi=$ip7MFK z9dnwj*kM)QPPu6`T20`gcYbn&^1U$rPvA-kA>ztqQW)G#~1zw04kj@ z-Goe<->&2rmMB&qN)Pp_0y&j&xDdJ!i%D~aYirJpM1GDR#5xNTp_HUpB`l^K`nga`AxdubHtgGw7(=UFYvCS#rr;G7u0>S=H6k36VQ-< z$5r|RlxLY1&=ob5d{u9utbHNECCW3 zs}jEFrTvxNg!i5*@}qMP9TJApMWqEsB)35Wp=Z0XHAyAWL0Pg4QOBBn3-9xs$<&PA_j<4QzxTSPzq+n- zIp_2JezxbipZmTa?Ppo=45XWMBVAom!qtbk%PI!0H zo;6?7gw2g1PCMo+ROAg@Gcn&fy+l;3-hCIU`&u~(3E0?-0h&it*1Om2L)hHo${CRV zgO!viOJPwVcP_r4e8O|K!md$b?{#yH>AP$)t;Z&Mc_X?vCB#BB-uRf za0>le>dC>Vs3?RlZ9rxo?2{(m!Lmu;Kt-5+`tm8D_3#&5*LP002Jg>7@VmFIS>pqS ze3`yazvgzz0Y3JIkSzbFAi?~As9bwk*UyE3sB~Nqzj6U~@77b(NbkfZ0NPwr(NL@j zNhBG3H8y)Qww)fiw=>gNfEATr0MQPXJ?mp0-N&G?5Jq(}a?A#S^Q-0kRx#b(8+?Vz zV6{i@a}4GmQ3h)F?;mpc2L!i$xe0<49{=3_vyg^}-7EQ%&C$R9wnh`KX>L$#u;VXh|kS@d24TB9jSnmc={Zfwgea{d!Y zhw95F5pC#Zz4Gv9kKm2vlqwwVj@Btb+QnSB)_&*P0Pf1n+RozaYDD{OO)wO0F z9G@?S$(27Yu~eV>&#Hy-H=oQQc1x;?C)fqz=3gac6Tje|Mm*p{DHWz04??OKzsy(1 z%%5L*mOXlIF~77clOFj5$`7}c`F)Y^;#rCk)R1u_DlX)&MlEW<30Go7b)eka^I-9# zBn)zC+v1AehO8T--twCN4Bnrn!7dUSeF0ATJb1E(4X8s&2iu)^S`D-m-|(3gQ2ql& zHc)6SGShqz1O*H%{bF+b94I{?G%;5Mt1B40@vF`#YNgv%K#O#FCtF0=(O(E-E>!@L!TT5&4Wc}daspu zO^x;45DEd;Jh^RmS1AX3#?HB_XFgpnMedy&_n#9GusU@Z*|II*lk@?inA^bKMzi+v zjW?Va$%#(6&7LfmHIKl04>ZN=sNVg9v!R92Ke-2uR-!YTyE6*=FCG5O8o}nq5T9bl8psE;nrHq-8%7P8^2NW3Gj^M zq*X;L+I0QQC4b?yw?oqFK4%`O+eGh1XF$X+g6G5&C>6QYp>}t|&+3KnAV;?iVQbe2 zte-bJB`74p4luoU3luNOEw55XKj13=*^{)%*F%c6*_-uQa-9G|jXPXarT} zr6A{hf~R5`Pn2F9lK{%#mwKeR*3p?7_nmfzxew4|mksK)Dkc7@(>oWX&k+a}o3i$L zz6MkH(cY9;1y|0bmB}Qk;wi}p#nG1#oTf73LWP`nUFYv_Mu)GLK=AC9$66zny3jfb z2_)zbA2M6e2nvJV3E*{x%>;}B55xYCWXo>^dO3KRn|i|~*vaYR0swwJOnZVNw9AVSt+;L+L#t!%noax-q^eGpGsT}~A zUs2RhjZh>U$1sZVv1YR8%tAyPBqf}A{QYUt2NI8Zt=@9<>8^oV1gfFHZiD2~?it|t zT}IDKA+_9U-gOJv08evZ%W-*CK^Y;cIUa7hz@h`@Y@_Fej$g?1>6A5VMl$gVl@D=y zZYB}_vAJRVei*RjxuCmk?+T=vB=m88`ar&y09w(rp3V(?>U?+e3Zhw zX*ZC8M`p;UUsnNSr5?`d4ltEG4V9JL0CCULe?OP-aPI;@)lsj~J0rsQr)Id_?uo-D zwSaC$f@WL#pcI;tWlYgtEvM1sVbh=| zbn{)%YXy}I!!=|BQP5`aI@XL$5L-AQ>bNpfM9@u}dA9}ykMOgxa<{9)E&M@l9R!Rp zRjie35C84~NrQA!+E?M&=hM4dlQ8F(aX09=RZ#A}tD%YPjRJBQXVVk{{&~I=<5t5o zD)%OV0AC5X7uWZAg6psr6z@;2rvFn*1fvxchjPIasT{W$^!Y}|VBgAU<*bN-{%_PQ zDy1Q2h>prwSTP|UQip{`^9% zAk_&%*$0F40Pfg*ysJ0cXrStJD+CFNG070yV31!A*{2Qgd~boEigyviT?I`+K=zh) z(Joalu3*tMQ=5xBzCnP4BM#ije5i- z*+VPx+Q(v^eig`mj*FIx!aNb&Lsu^R=#ng;Xm3k+7Yz0+kqs8#(rTSu8^)L3+w|Rc zl6!PC{$uzVYW$MCN@c*2b7ETW8>G=iP<53hN^*X77X1>W5##b;ot5>ehAgaBt|G!= z@LkLXjfGph9bOS%?tsRZ6mvf00J=G6-v9jp|9z7&>12VTOD4PPy^=h76*L?)O^&gROxJbZq zC0go@snyVRBYTJ6H?_`sf`oY;-z&`^(P}aQO^R7Y+8~kAf(l#$n&ZPY_@v5jEDQPd z=H2%*bmLqJAXT-7!t#1!%M)dq3eG}57sBFe+exsC@i0aq!DbNK%=JdQq@lqcD4NQ*l?(<=UOAUTUVi}p<&|a{%ka2LaAva4ihqO``UKeS7PEVj(Fz3;w=!^x z!>ehquysZrd!O~5Lm)y|@zZ-J3-(ltLqJ~|OtYSm*=_2OLKpMBie46s4?XC*jLkr7 z%dbJ|rrqjD(YU`WPH+)pTto1o@ziepBRsCm=#lAagbqXK1F_f~V78}pR<<7~;uC1f z3gop*K<_O>JZ*{asY6N({=~}9713lTDum|iffEv~cIyr#iW(%rUVvCU1i?VKcLHF< zXx}{A_ss95)cS+FWx;MhiUq&{0>H=2Os9*nnS_2MwY3u9HVwCDS?aIDP#L`lJWc_1 zjad&#V+T}ED#nt^_k(f6T%jH?2n`b+z{M+a!|9$5?blR>_zAcKTrjrqkV54Kwg=4e zW&V(aqLgeGmyI+P?G{7GJ*?m%d~JZ-+u1ZGGc;Aa*j$g>$1sBV(;A<#>o88me92 ze!}K{p9q1ydnwKOJoKFnk!VG9u3$;AI5dH91p~HgVJ?(+4?aE1H9P?*#);?BN#Rxb zu%{}Z{ib2o184yiCD9uC5Xh?8#38x9o#;UipyJ*e4PnceuT%qwIUMN3y+UoI8nK9; z4XuJTvq<{95vf==!-ywK>hGJBpRB4t!pDFz%xsg7h%WHJ8_^-uL-Y?N;1M(T^fj6m_1df!G~|iL<)8lUst~^VxCMIyYLD%`$q3+59_3Zh4b~qZ(|Pu zyJ7Lf*k{m9Hb(AJ-Rclz@(!ZejZux5d5%^|o%+sw@xfhB^&~3GuHO_S4@nz@Ro&6p z1BT0-w7NAl=V;9R`6B{110o=x&G*@^W|IUErHN=IZ>Zk`-Oi`tS6_6kNU```mB1cK z6ufvK%0#jNJ1CCJ^$3en02&Auav`scOobgt$u?yr{&Gx4K z@o@;&kFaAjsx7jGax-R+?`q^RcIqZhrUTM6D<0gHH3ja$B21;qu)I6mYIDjJ{M7DM z$!a`FV;nKvwbc%UeFCO%^gWlPFB0gMW!}vjNp?s9F3N^Kas=Vn@?>;0Q2KMjeH=-r zp%=})Zpyp2xBTAgBxLE*bil+#Te88WiF$q55oBw1dFo#TE40YVg^y+%V?`v5phH z{~LCQr~Cm7xr>Og8rj7?WRDUvm=0kmlH$i4+~_?tEv$VeNGjeEcjnRLk?3cx;DHuLLzYoz=T!Nj9tsHp>f>}wq#+d;#`w}9A)r4BcX{o z?X;f`cweUs^UVHdNOV!3-eU+&jKqV552yh^d{3O2sf}X3o|Iz(Fn!uPo#;mTHQh-} z?q9J}fy(;qC?YRADa|G607H=j!7PL{846pZslg=gy`yM1Fj7$s&nE!MV9-dpXux+= ze=e4x@!JD%jaQXeqs>0BB$YvoB_?sFCczy7S4XodWJa-G_^q5K|uczN{ z9hf`o+`2>zxDVU}idD!+pneLCPk4HSdBD3Tf>_5ztNR7nQQES~iAP6ob&+p-|M(rd zx9cBD2=4_8m5Q8hQ5!94ySs$bcpc5^{FNLS~frl((ik)?BjnCJ+5rN9ENMa^Cys#;OL&^#+{;id###t8T(_2Kx|Y zL~kmGAU3$ctraQV4Y)Vi4Juu3Yu<$7ZYXu)dNOl~dJBFTgslV*Ul_nfz) z%DUs(1uobm3_K1qW|J=syhkfl_8t8ncFfxwFuAF|Ju63q?c!?S_49Vx&h z7wUl;M8M2C;65S)LSA017ane)?f;)*h_A8)5s5~7Wh{Pfbv@tAf?wMLIhP44Ci}F5 zPUB$9ar+LKVp$Q=#p3})B`(^|PvIo#{RZCFt&ZMna|{{LCtw3}37Dck{!)tTBS?FT ztDjR^%E)F=P0_t!b#kLYSk|d?yU!B&xM=0=C8OysWw6Slq>w-xkLp^ z(IBY_@G~g*NQxENcd>7T!%K$wYd)MR%1y^#T!_7W%b&wA`2oO7rpFr~K(dCxC5$O- zFYy;7-zzm|U%XG5v#V?~S1ih-UcZLTfh&jH0F-LT?ZX-^dO)ufl{lJj9Ivi2|q zlLf)kK}sZbrop&{A}&Da>+-d7F4UU5klrX6!!*L?T>xe%Trxh{JH*F|+!o_eb&pz3e%SC6hm}3JK-&x3Lt@Eh;WK>iYa+cdwxqDgs%4B0|woC_x$CPkA8KwdaRn7_9 zu?JXh^3nCGA|cR6%$xK^nar720%>LG{9 z2Wew!e9;+8WYfxR+f7NPE?xBZWE6D{K<0u=6eMV@Iqoc;Zl z12%|a<5+#=*&l!)Umhc_hzAE6&H-bp)ZpYe9LU%LV?5xc2M^T&c3axDo3CFg$b5vB zX=PXaMiTM>&h4=j3t~tI$!ITOhy>1pkFQxkchmZg!4%t$42Pz_+|LIE#Uf0p~nBlbe0 zX&bI*T5Kod(x{TsoK=T{??8B{%fSNoXPOVyfX?6W$-Xjn${lE3LPzug-VvVEdbie zWLf82T7YbdN+Pn!+9S2Ce=W+=fQBr4HZdfvX4Tfg6MS(fk|u@=lWX!YKZoEY_U=} z*=D`+ut;c!7+I{$@IZ&wi`O4LfB5sji`!1ne888KsqqcP0TI!7&Ulp9cS5M9C} z{ntS{{~sb7oZU<7DZ9R*IJk0ljay)iiP)drogiP*X#*ZK%eetflUdb4Om)*T3RwN& z&?YsjjOR6!O6wuC>%c2?6NGVG#j|1h>T3XJwMB<4G$VVA2BY9f9f%cM>QoEX%s==88!ljtC%tq?0OvU~R7t(` zfP^|s;yHPUiB>{BJr=E(07ty97sI%L>ii6FGB77IUdS2t4L z%w`~I=M05*&fI*Fl0-wdEShl-02YBX@TnDc(FR23||qDKH#JfQltoZPxsqTWt;(S-M1@`seGJYP-swM34_U z%j!i7_-P>cbIX;COHLD@BA?tHQz)mlz>+0{!Tva6mDzzV00a zw;)r*K;82`6@U>rBE&a~^%JJZLS&tUVa7CaOd6)ozpPp@g{@mz`|}cEZt`LhpG&v# z=W(+U(q-=egK19`H0=;JFQjqX17Xwtw25GVaUqR%kW@k+%8H%YB=pdD`c$JZwDAR% zr!Z)ohG#)?t03752+40T2UI7o$s}>Rqm#8$_AAYXd?t4dW5@9vg==Gx!Vxnp0i~5U z#dG5FPcNk;Tgg0wkiBR4MzQ#!xW;dALM3xevC#aWsJ+F*Wm!MZEhci|kFUI4G0%q@+VC+V(}(`hmF$*oh*Tnm z?bantp#RPqr)Zc(olK7pHQwy}L72kcj4!#j7Mim@fpg()^|5<1$X1;^wL4IgR%hZ5 zZgxjU$&4SwNPAyq?&sAYH!6LK1 z5F7gGx8zR1U5b75fB{v}wA`7dQU%^lgkPbQCP;s5WxgSP^Bb9iw0L$)6+fgW43{G5 zV^fv#HX#2vMbt4SA#wgQZaWP{yNcZk7v^bN)jvbE;02UUlqOdoZ+0>oyqh9&g&IGA zYPk4U#4Uzq6-j+ir()aN64y+{EV2(njw$y>hl?aNzX^)qMRulZJ)XZ1 zo?ax9-bRp^`)e7~Q&#V-C$kNcAd{w7YF~|%cX)W-t+Q$yITT(e&pB*U8}mEXZV&?t z>~Ai7(=fo|8N$fp>2AoYIJFoVw79xeW_ zwxRWS5pZuBnAnmLnOSFIH6I6w?3@j>dVP9+$hHD^ICf`+6^MR5TI04L0rr9%n zKxyCub1=s{hnc3kuLdT*vWKoePM-a+4R59&Z)^)t--)^5v6F@S08D_C*~q01n{cBW zlWjR8xd7oJGLL)&TJSSY%yTD#1>mkL8acOpJ~mh8H3f9p2D=R2B0bB2@n$^a0PTCD zO5oNP)+gY2Z^#HYKwahgotE!suoQ~rRJQLIYnU4zo(<)P$P`MKD>>c&jI7CF#HLBZ>P)0 z(9UFVWHC6=R}tErNnqJnTlV}`L!>fFsfl?%U|WKhW4GbHi}w-cLa=6NcC2r1#tS@h z-`h^37o2ulq7|TPW|p7^kbfiROxsUkzN(&lZ)X8)RcypAXH;R8TW=GX2NA$MEnVFu zh&%$--jsGfNlUNh78m`M3tfe&)Yl5VcPcibk!jG@#$->vrun68%m>O^qx|N1Cp?24 zR0uAfK=EQN7Ae5QN!pPJ^t4RyHM1MeJ_z^qD<~iaY?ZIuH(WT=so;MRYyu*9&*y}> zfGb8|pnQQ87Ck@`q39s5Mve+~NuiaS_UM+-k5D$BNBbFP+fl@qwckV8ly$;T@AJSQ z(%&`DH|#YaQuWm*H{2-g`^<8YfI_UL-3F@zQ6t#k;A%!<84rV$0$~`6-c?BnJyC9Q z9`hUhG(VhICkHMB%}>lki+tb0E%f;$`0JO>692$3vo%2QOKvm&VXz%+KA_BQlRCVOvu-M z_j##(87FlI=nXpay&f;1zls{1g716e6~FjPdOe?TzcT||p{iJX>%#AQ1mDM3!lg7n zsxRXdt%5k^bpN`$3vr57Ao0)mcrcmm%PW|Fi2w1eLs@-b*Yf#MeHrJ|EO@my&uBR< z{JsF--^>0xRWL2of2V4Gd;Pns{@Si^mi_m;>cQOiDFN~2PuacAzJq@{d-OCib{+cp FzW{V#17QFF diff --git a/images/plotly/taxonomy_navigation.html b/images/plotly/taxonomy_navigation.html index b51c3aa8..fb2b3660 100644 --- a/images/plotly/taxonomy_navigation.html +++ b/images/plotly/taxonomy_navigation.html @@ -3,23 +3,1934 @@ plotly - - - - - - - - + + + + + + + +

-
+
- - + + diff --git a/index.Rmd b/index.Rmd index 36a96d8b..a9a4e93f 100644 --- a/index.Rmd +++ b/index.Rmd @@ -153,19 +153,19 @@ tags$div(class = "category-row", tags$div(class = "values-row", create_values_box1( - value = "\ Openness", + value = "\ Openness", icon = "fas fa-book-open", text = "Data are most useful when they are widely available and easy to use. We try to encourage the sharing of tools that make data from the Atlas of Living Australia more open and useful for everyone" ), create_values_box2( - value = "\ Scientific transparency", - icon = "fas fa-atom", + value = "\ Scientific transparency", + icon = "fas fa-dna", text = "Transparency is necessary for reproducible science. We encourage that decisions, methods and deviations in our workflow are clear and transparent from the planning phase, to the retrieval and analysis of data, to the final output" ), create_values_box3( - value = "\ Robust methods", + value = "\ Robust methods", icon = "far fa-chart-bar", text = "No method or analytic procedure is perfect. We value consideration for strengths and limitations of each method or analysis to help choose the methods that provide robust, reliable results" )