pyplot.scatter() plots points but has no meaningful axes?-CodePudding

I am trying to make a simple scatterplot with Matplotlib. I am passing an Numpy array for x and another one for y:

df = df.to_numpy()    # this was originally a pandas DataFrame
print(df)

plt.scatter(df[:,1], df[:,2])
plt.show()

The print outputs:

[['B' '-693.3127738066283' '19.14412552031358']
 ['B' '-1633.974496310751' '40.13395450795514']
 ['B' '-2010.8973373308845' '-37.64969595561755']
 ...
 ['R' '-1034.7669874549774' '-76.93110447814361']
 ['R' '745.6579736997674' '-51.74835753276244']
 ['R' '-1473.8940519681794' '-28.58246870754514']]

However, the plot outputs this:

To give a better view of what's happening, if I plot only the first three datapoints it looks like this:

So the x- and y- coordinates are being "plotted", but the axes have no meaningful scale or value. Why is this happening, and how can I make a regular scatterplot?

CodePudding user response：

As you can see from your prints, df contains only strings. matplotlib has no idea what to do with them.

Change:

plt.scatter(df[:,1], df[:,2])

With:

plt.scatter(df[:,1].astype(float), df[:,2].astype(float))

What it does is converting the arrays to numeric types with which matplotlib knows how to deal. Of course, this is assuming the arrays contain strings representing valid numbers only.