Plotting
Plots are the language of data analysts. There are mutiple dozens of types of plots, and the majority of them are possible in Rust. Plotly, the very popular JavaScript open source graphing library, has a Rust interface! This chapter will use the plotlars crate, a wrapper around the Plotly library that takes polars dataframes as input. This is a great bridge between Polars and Plotly. Plotlars allows you to build all types of graphs, like bar plots, box plots, line plots, pie charts and sankey diagrams.
Run this code using cargo run -r --example 5_2_1_plots.
Bar Graph
Setup
Lets get some summary statistics to output as a bar graph. Here is a table of the mean income by sex and region:
use plotlars::{Axis, BarPlot, Legend, LinePlot, Orientation, Plot, Rgb, Text};
use polars::prelude::{pivot::pivot_stable, *};
// Connect to LazyFrame
let args = ScanArgsParquet::default();
let lf = LazyFrame::scan_parquet(PlPath::from_str("./data/large/partitioned"), args).unwrap();
// Modify var
let lf = lf
.filter(col("keep_type").eq(lit(1))) // Usual resident
.filter(col("income").is_not_null());
// Mean income by region and sex
let df_bar = lf
.clone()
.group_by([col("sex"), col("region")])
.agg([col("income").mean().round(2, RoundMode::HalfAwayFromZero)])
.sort(["sex", "region"], SortMultipleOptions::new())
.with_column(col("region").replace_strict(
lit(Series::from_iter(vec![
"E12000001",
"E12000002",
"E12000003",
"E12000004",
"E12000005",
"E12000006",
"E12000007",
"E12000008",
"E12000009",
"W92000004",
])),
lit(Series::from_iter(vec![
"North East",
"North West",
"Yorkshire and The Humber",
"East Midlands",
"West Midlands",
"East of England",
"London",
"South East",
"South West",
"Wales",
])),
None,
Some(DataType::String),
))
.with_column(col("sex").replace_strict(
lit(Series::from_iter(vec!["1", "2"])),
lit(Series::from_iter(vec!["Female", "Male"])),
None,
Some(DataType::String),
))
.collect()
.unwrap();
println!("{df_bar}");
// Convert from Polars 0.51 to Polars 0.50
let df_bar = Interchange::from_polars_0_51(df_bar)
.unwrap()
.to_polars_0_50()
.unwrap();shape: (20, 3)
┌────────┬──────────────────────────┬──────────┐
│ sex ┆ region ┆ income │
│ --- ┆ --- ┆ --- │
│ str ┆ str ┆ f64 │
╞════════╪══════════════════════════╪══════════╡
│ Female ┆ North East ┆ 54834.9 │
│ Female ┆ North West ┆ 55016.59 │
│ Female ┆ Yorkshire and The Humber ┆ 55089.42 │
│ Female ┆ East Midlands ┆ 54911.42 │
│ Female ┆ West Midlands ┆ 55141.59 │
│ … ┆ … ┆ … │
│ Male ┆ East of England ┆ 55189.26 │
│ Male ┆ London ┆ 54936.82 │
│ Male ┆ South East ┆ 55201.77 │
│ Male ┆ South West ┆ 55239.98 │
│ Male ┆ Wales ┆ 54865.14 │
└────────┴──────────────────────────┴──────────┘
Building the bar graph
To build a graph, you start with the data, you give it an x axis and a y axis, a group if you have groups and then various options, like x_title, y_title, plot_title, size for text, etc. Most functions are self-explanatory, but they are described in the documentation of the bar graph.
let html = BarPlot::builder()
.data(&df_bar)
.labels("region")
.values("income")
.orientation(Orientation::Vertical)
.group("sex")
.colors(vec![Rgb(255, 127, 80), Rgb(64, 224, 208)])
.plot_title(
Text::from("Income by sex and region")
.font("Arial")
.size(18),
)
.x_title(Text::from("Region").font("Arial").size(15))
.y_title(Text::from("Mean Income").font("Arial").size(15))
.legend_title(Text::from("Sex").font("Arial").size(15))
.legend(
&Legend::new()
.orientation(Orientation::Horizontal)
.y(1.1)
.x(0.37),
)
.build()
.to_html();
In this example, to_html() was used to imbed in this page to be interactive. You can save it to a .html file:
let mut file = std::fs::File::create("./data/output/bar.html").unwrap();
std::io::Write::write_all(&mut file, html.as_bytes()).unwrap();
Opening this file, will give you this interactive bar chart:
Instead of the to_html(), you can also write an image using this syntax: .write_image("./data/output/out.png", 800, 600, 1.0).unwrap()
Line Plot
Setup
Lets also get some summary statistics to output as a line plot. Here is a table of the mean income by sex and ours worked (groupped), pivoted on sex:
// Mean income by sex and hours worked (groupped)
let df_line = lf
.clone()
.filter(col("hours_worked").neq(lit(-8)))
.group_by([col("sex"), col("hours_worked")])
.agg([col("income").mean().round(2, RoundMode::HalfAwayFromZero)])
.sort(["sex", "hours_worked"], SortMultipleOptions::new())
.with_column(col("sex").replace_strict(
lit(Series::from_iter(vec!["1", "2"])),
lit(Series::from_iter(vec!["Female", "Male"])),
None,
Some(DataType::String),
))
.collect()
.unwrap();
// Pivot to make two lines out of sex values
let df_line = pivot_stable(
&df_line,
["sex"],
Some(["hours_worked"]),
Some(["income"]),
false,
None,
None,
)
.unwrap();
println!("{df_line}");
// Convert from Polars 0.51 to Polars 0.50
let df_line = Interchange::from_polars_0_51(df_line)
.unwrap()
.to_polars_0_50()
.unwrap();shape: (4, 3)
┌──────────────────┬──────────┬──────────┐
│ hours_worked ┆ Female ┆ Male │
│ --- ┆ --- ┆ --- │
│ str ┆ f64 ┆ f64 │
╞══════════════════╪══════════╪══════════╡
│ 15 hours or less ┆ 54997.22 ┆ 54990.49 │
│ 16 to 30 hours ┆ 54949.22 ┆ 55026.07 │
│ 31 to 48 hours ┆ 54904.08 ┆ 55022.64 │
│ 49 or more hours ┆ 55195.33 ┆ 55353.39 │
└──────────────────┴──────────┴──────────┘
Building the line plot
Similar to the bar graph, for the line plot, you start with the data, you give it an x axis and a y axis (and here, you add another y axis with additional_lines to add a new line to the line plot), and then various options, like x_title, y_title, plot_title, size for text, etc. Most functions are self-explanatory, but they are described in the documentation of the line plot.
let html = LinePlot::builder()
.data(&df_line)
.x("hours_worked")
.x_axis(
&Axis::new()
.tick_values(vec![1.0, 2.0, 3.0, 4.0])
.tick_labels(vec![
"15 hours or less",
"16 to 30 hours",
"31 to 48 hours",
"49 or more hours",
]),
)
.y("Female")
.additional_lines(vec!["Male"])
.size(12)
.colors(vec![Rgb(178, 34, 34), Rgb(65, 105, 225), Rgb(255, 140, 0)])
.plot_title("Mean income by hours worked (groupped) and sex")
.y_title("Mean income")
.x_title("Hours worked (groupped)")
.legend_title("Sex")
.build()
.to_html();
In this example, to_html() was used to imbed in this page to be interactive. You can save it to a .html file:
let mut file = std::fs::File::create("./data/output/line.html").unwrap();
std::io::Write::write_all(&mut file, html.as_bytes()).unwrap();
Opening this file, will give you this interactive bar chart:
Instead of the to_html(), you can also write an image using this syntax: .write_image("./data/output/out.png", 800, 600, 1.0).unwrap()