Random straight lines when plotting two data series on matplotlib-CodePudding

I'm new to python and I'm trying to plot a line graph of two data series using matplotlib.

Here's the sample code:

import matplotlib.pyplot as plt

# choronologically sorted dates with frequencies
date_1, freq_1 = [...], [...]
date_2, freq_2 = [...], [...]

fig, ax = plt.subplots()
ax.plot(date_1, freq_1, c='blue')
ax.plot(date_2, freq_2, c='red')
plt.legend()

plt.title("Date and frequency")
plt.ylabel("Frequency")
ax.get_xaxis().set_visible(False)
fig.autofmt_xdate()

plt.show()

Commenting out the first or second plot produces no issues, as shown: first plot , .

But when trying to plot both at the same time, I get weird straight lines in my plot: overlaying first and second plot .

Plotting the second plot first gives this .

What is happening here and what's the most beginner friendly way to correct it?

CodePudding user response：

Most likely you have an outlier or a datapoint that is out of place. I would suggest plotting the data like this first just to make sure if it is indeed an out-of-context point:

ax.plot(date_1, freq_1, '.', c='blue')
ax.plot(date_2, freq_2, '.',c='red')

Then maybe try to sort the point according to the x. Or remove the outlier.

From the graph, most likely you have a point that is (0, 0) somewhere in the middle of your data.

CodePudding user response：

Without data this is hard to answer - it looks as if you got either non-sorted data with a regressing to 0 in your plots data:

import matplotlib.pyplot as plt

# choronologically sorted dates with frequencies
date_1, freq_1 = [1,2,3], [4,5,6]
date_2, freq_2 = [1,2,1,3,], [6,7,0,8]

fig, ax = plt.subplots()
ax.plot(date_1, freq_1, c='blue')
ax.plot(date_2, freq_2, c='red')
plt.legend()

plt.title("Date and frequency")
plt.ylabel("Frequency")
ax.get_xaxis().set_visible(False)
fig.autofmt_xdate()

plt.show()

Having a NaN in your data would give you a break in your graph: