visionforge/plotly/docs/tutorials/HowToDrawASinus.md

220 lines
11 KiB
Markdown
Raw Permalink Blame History

This file contains invisible Unicode characters

This file contains invisible Unicode characters that are indistinguishable to humans but may be processed differently by a computer. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

### How To Draw A Sinus?
In data visualization tasks, quite often there is a need to depict mathematical
functions - for example, to compare the convergence rate of a model
with ideal one or to find an approximation for a given time series. Such charts
have a number of features that are not inherent in most other charts: for example,
visualizing the perpendicular axes (OX, OY) from (0, 0) and a large number of
additional information. This can be the expected value of the function, its extremum
points, deviation at various points, etc.
This article will tell you how to depict an ordinary sinus using the library `Plotly.kt`
so that in the end you will get a visual and informative plot.
1. Let's start with drawing perpendicular OX and OY axes. This requires to do a few things:
hide default axes lines, hide zero lines (it's impossible to draw an arrow on the end of it),
add perpendicular arrows using annotations (`Text`) with visible arrows on the end, but without
text. After that, let's sign the axes themselves. It is also important to specify the required
chart sizes (`width`,` height`) so that after saving the graph will look the same as in the browser.
```kotlin
Plotly.page { // making new html page with plot
layout {
width = 900 // width of the plot (in px.)
height = 500 // height of the plot (in px.)
text { // vertical axis
x = Value.of(0) // position of the top end of the arrow = (0, 1+eps)
y = Value.of(1 + eps)
ax = Value.of(0) // ax, ay means the offset of the bottom edge relative to the top
ay = Value.of(430) // positive (negative) value is the arrow length
// upwards (top down) and from right to left (from left to right)
}
text { // horizontal axis
y = Value.of(0) // arrow left position = (2PI + eps, 0)
x = Value.of(2 * PI + eps)
ax = Value.of(-800) // position of the left end of the arrow
ay = Value.of(0)
}
xaxis { // OX parameters
showline = false // hide OX line
zeroline = false // hide zero line
}
yaxis { // OY parameters
showline = false // hide OY line
zeroline = false // hide zero line
}
}
}.makeFile() // making temporary file and visualing plot in the browser
```
2. Now we will draw the sinus function itself - for its visualization let's choose a bright blue color,
contrasting with both the white background and the black ticks on the axes. After that using
the variation of the `mode` parameter (` ScatterMode: lines, markers`) the intersection points with the OY axis
will be marked with a slightly darker shade of the same color.
```kotlin
x1 = (-410..410).map{ it / 200 * PI } // function domain (-2PI - eps, 2PI + eps)
y1 = sin(x1) // function values
Plotly.page {
scatter { // making Scatter plot (sinus)
x.set(x1) // assigning OX values
y.set(y1) // assigning OY values
line { color(XKCD.CERULEAN) } // color of the line is cerulean
}
scatter { // making Scatter plot (OX axis dots)
mode = ScatterMode.markers // visualizing only dots
x(-2* PI, -PI, PI, 2* PI) // points of intersection of sinus with OX (values along the X axis)
y(0, 0, 0, 0) // points of intersection of sinus with OX (values along the Y axis)
line { color(XKCD.CERULEAN_BLUE) } // color of the dots is darker cerulean
marker { size = 8 } // dot's diameter is 8 px.
}
layout { ... }
}.makeFile()
```
3. Let's draw horizontal dashed lines on the chart, corresponding to values equal to -1, 1 (extremum lines) and 1/2.
This requires changing `line.dash` parameter to `Dash.dash` value. The specified values will be plotted after that
on the OY axis using the `tickvals` and` ticklabels` parameters.
```kotlin
...
Plotly.page {
...
layout {
shape { // adding new figure (line y = 1)
x0 = Value.of(-2*PI) // (-2PI, 1) and (2PI, 1) coordinates
x1 = Value.of(2*PI) // will be connected by a line
y0 = Value.of(1)
y1 = Value.of(1)
line { dash = Dash.dash } // dashed type of the line
}
shape { // adding new figure (line y = 1/2)
x0 = Value.of(-2*PI) // coordinates (-2PI, 1/2) and (2PI, 1/2)
x1 = Value.of(2*PI) // will be connected by a line
y0 = Value.of(0.5)
y1 = Value.of(0.5)
line {
color("red") // red color of the line
dash = Dash.dash // dashed type of the line
}
}
shape { // adding new figure (line y = -1)
x0 = Value.of(-2*PI) // coordinates (-2PI, -1) and (2PI, -1)
x1 = Value.of(2*PI) // will be connected by a line
y0 = Value.of(-1)
y1 = Value.of(-1)
line { dash = Dash.dash } // dashed type of the line
}
...
}
}.makeFile()
```
4. After that, it is worth adding vertical lines connecting points on OX, the values in
which equals 1/2, and the value itself in LaTeX format. Shapes are used again for this - an array
values of type `Shape.line`. To use LaTeX format you need to include the `MathJax` header.
```kotlin
...
val sub = PI / 6
val xElems = listOf(-2PI + sub, -PI - sub, 0 + sub, PI - sub)
// points, where sinus equals 1/2
val shapesList = mutableListOf<Shape>() // making list of lines
for (x: xElems) {
val shape = Shape {
x0 = Value.of(x) // (x, 0) and (x, 0.5) will be connected
y0 = Value.of(0)
x1 = Value.of(x)
y0 = Value.of(1/2)
line { color("red") } // red color of the line
}
shapesList.add(shape) // adding new figure to the list
}
Plotly.page(mathJaxHeader, cdnPlotlyHeader) {
scatter { // add string "1/2"
mode = ScatterMode.text // visualing only text
x(-0.35) // text position on the OX axis
y(0.56) // text position on the OX axis
text = listOf("$\Large{1/2}$") // increasing font size with ТеХ
textfont { color("red") } // red color of the font
showlegend = false // do not show this trace in the legend
hoverinfo = "none" // do not show anything on hover
}
...
layout { // adding figure list to the plot
shapes = shapesList
...
}
}.makeFile()
```
5. It remains to add labels on OX corresponding to the intersections of the sinus with the axis. It
is done in a similar way, using axis labels (`tickvals`,` ticktext`) and writing text in LaTeX format.
```kotlin
...
Plolty.page(mathJaxHeader, cdnPlotlyHeader) {
...
layout {
xaxis { // OX axis parameters
...
anchor = "free" // axis position is set manually
position = 0.43 // assigning axis position
tickvals(listOf(-2 * PI-0.05, -PI - 0.15, PI - 0.05, 2 * PI + 0.1))
ticktext(listOf("\$\\huge{-2\\pi}\$", "\$\\huge{-\\pi}\$", "\$\\huge{\\pi}\$", "\$\\huge{2\\pi}\$"))
// ticks positions and text
}
yaxis { // OY axis parameters
...
anchor = "free" // axis position is set manually
position = 0.485 // assigning axis position
tickvals(listOf(-0.91, 0.09, 1.09))
ticktext(listOf("\$\\Large{-1}\$", "\$\\Large{0}\$", "\$\\Large{1}\$"))
// ticks positions and text
}
...
}
}.makeFile()
```
6. As a final improvement, let's add a legend to the plot, which will be placed in upper right corner of the graph.
First you need to set the name of the corresponding line, after which with using the parameters `xanchor = right`
and` yanchor = top` set the position of the legend so that the coordinates specified by the parameters
`x = 1`,` y = 1` correspond to the position in the upper right corner of the image.
```kotlin
...
Plotly.page(mathJaxHeader, cdnPlotlyHeader) {
scatter { // sinus Scatter trace
...
name = "\$\\Large{y = \\mathrm{sin}\\,x}\$"
// the name of the plot displayed in the legend
}
layout {
legend { // legend parameters
x = 0.97 // horizontal position of the legend
y = 1 // vertical position of the legend
borderwidth = 1 // width of the legend border
font { size = 32 } // size of the legend font
xanchor = XAnchor.right // "anchoring" the position of the legend to the right corner
yanchor = YAnchor.top // "anchoring" the position of the legend to the bottom of the graph
}
...
}
}.makeFile()
```
7. Final picture:
![Картинка](https://i.ibb.co/wKTCp4H/newplot-17.png)
Source code is available at: https://github.com/mipt-npm/plotly.kt/tree/dev/examples/src/main/kotlin/tutorials/SinusPicture.kt