Scale_color_manual(values=c('#00AFBB','#FC4E07'))Ĭoursera - Online Courses and Specialization Data science

Scale_linetype_manual(values=c('solid', 'dashed'))+ # Change manually line type and color manually To be able to apply these functions, you should create a geom_line, which line types, color and size should be controlled by groups.

scale_size_manual(): change the size of lines.

scale_color_manual(): change line colors.

scale_linetype_manual(): change line types.

Change the appearance of line types manually:.

Geom_line(aes(linetype = supp, color = supp))+ Ggplot(df2, aes(x = dose, y = len.mean, group = supp)) + # Compute the mean of `len` grouped by dose and supp

Create a line plot for multiple groups.

Ggplot(data = df, aes(x = dose, y = len.mean, group = 1)) + Add a legend to the plot and set legend lty Plot(x, y1, type = 'b', frame = FALSE, pch = 19, The option cex is used to set the legend text size. Finally, we add a legend on the plot using the R base function legend(), which take the same col and lty arguments as the lines function. Next, we add a second line with a dashed line style ( lty = 2). We start by plotting a first single line with a solid line type ( lty = 1).

xlab and ylab: for x and y axis labels, respectively.Ĭreate a plot with multiple lines and set the legend lty.

col: change the color of point and line.

lty, lwd: set line types and thickness.

pch and cex: set points shape and size, respectively.

Lowed values: l (display line only), p (show point only) and b (show both).

type: display the data as line and/or point.

x, y: variables to be used for the x and y axes, respectively.

# Draw these two points with big triangles to make it clearĪx. # And we'll have the line stop at x = 5000 # anywhere so we're going to have it go through 0,0 # A line with a fixed slope can intercept the axis # Now add on a line with a fixed slope of 0.03 import numpy as npĭf = pd.DataFrame( )ĭf = df * 0.88 + np.random.rand(25) * 5000 In the example below I've put in a line with slope = 0.03. If you know you want the slope to be 0.88 then you can calculate the y position of the other point: y_1 = 0.88 * (5000 - 0) + 0įor the data you've provided in the question a line with slope 0.88 will fly off the top of the y axis very quickly ( y_1 = 4400 in the example above). Now you can rearrange the formula for y_1 as: y_1 = slope * (x_1 - x_0) + y_0 For this example let's assume you want the line to go through the origin (0,0) x_0 = 0 # We already know this as it was set earlier There are an infinite number of parallel lines with this slope so we'll have to set one of the points to start off with. You define the slope of a line as increase in y / increase in x which would be: slope = (y_1 - y_0) / (x_1 - x_0)Īnd this can rearrange to: (y_1 - y_0) = slope * (x_1 - x_0) Let's assume you want the x coordinates of those two points to span the x axis so you're going to set x_0 and x_1 manually: x_0 = 0Īlternatively you can just take the minimum and maximum values from the axis: x_min, x_max = ax.get_xlim() These two points have x and y coordinates so for the two points you'll have four numbers: x_0, y_0, x_1, y_1. Y_min, y_max = c, c + slope*(x_max-x_min)īuilding on answer you just want to add in a line connecting two points. # need a slope and c to fix the position of line

If you would like to plot scatter with regression line, I would recommend to do it in seaborn with a minimum lines of codes.ĭf = pd.DataFrame(np.random.multivariate_normal(,, ], size=100), columns=) The correlation coefficient won't give the slope of the regression line, because your data are in different scales.

Thin line scatter plot matplotlib