How to conditionally specify markers in matplotlib graph

[Mark17]

Here's some code:

import datetime as dt
import pandas as pd
from matplotlib.ticker import LinearLocator

df = pd.DataFrame({'date':[dt.datetime(2018,8,25), dt.datetime(2018,8,26), dt.datetime(2018,8,27), dt.datetime(2018,8,28), 
                           dt.datetime(2018,8,29), dt.datetime(2018,8,30), dt.datetime(2018,8,31), dt.datetime(2018,9,1), 
                           dt.datetime(2018,9,2), dt.datetime(2018,9,3), dt.datetime(2018,9,4)], 'n':[10, 6, 7, 2, -3, -5,
                            1, 3, 7, 10, 16]})
markerlist = []
trade_date = [dt.datetime(2018,8,25), dt.datetime(2018,8,30), dt.datetime(2018,9,3)]
for i in df['date']:
    if i in trade_date:
        markerlist.append('d')
    else:
        markerlist.append('.')
fig, ax = plt.subplots()
ax.plot(df['date'], df['n'], marker=markerlist, linestyle='dashed', markersize=12)
plt.xticks(rotation = 45)
ax.get_xaxis().set_major_locator(LinearLocator(numticks=7)) #this works nicely to evenly space x-tick labels
plt.show()

This raises ValueError: Unrecognized marker style ['d', '.', '.', '.', '.', 'd', '.', '.', '.', 'd', '.']. What I really want is to iterate (?) through each point and plot accordingly.

Actually, I've been told before I want to print all points at once rather than iterate through... but the program needs to grab the corresponding marker from markerlist, which is why I say "iterate."

Also, I want this to be a line graph (solid is fine... I used dotted only because I stumbled upon that keyword string. In fact, maybe I don't even need to include that argument?). I tried this another way where I did iterate through the points (using a for i and .loc[] commands) and I lost the line connecting the dots (which I couldn't explain).

Any thoughts?

[Mark17]

I just wanted to bump this... does anyone have any thoughts? Thanks!

[Mark17]

I just wanted to bump this... does anyone have any thoughts? Thanks!

This is closer:

import datetime as dt
import pandas as pd
import matplotlib.pyplot as plt
from matplotlib.ticker import LinearLocator

df = pd.DataFrame({'date':[dt.datetime(2018,8,25), dt.datetime(2018,8,26), dt.datetime(2018,8,27), dt.datetime(2018,8,28), 
                           dt.datetime(2018,8,29), dt.datetime(2018,8,30), dt.datetime(2018,8,31), dt.datetime(2018,9,1), 
                           dt.datetime(2018,9,2), dt.datetime(2018,9,3), dt.datetime(2018,9,4)], 'n':[10, 6, 7, 2, -3, -5,
                            1, 3, 7, 10, 16]})
markerlist = []
trade_date = [dt.datetime(2018,8,25), dt.datetime(2018,8,30), dt.datetime(2018,9,3)]
fig, ax = plt.subplots()
for i in df['date']:
    if i in trade_date:
        markerlist.append('d')
    else:
        markerlist.append('.')

for xp, yp, m in zip(df['date'], df['n'], markerlist):
   ax.plot(xp, yp, m, color='green', linestyle='solid', markersize=12)

#ax.plot(df['date'], df['n'], marker=markerlist, linestyle='solid', markersize=12)
plt.xticks(rotation = 45)
ax.get_xaxis().set_major_locator(LinearLocator(numticks=7)) #this works nicely to evenly space x-tick labels
plt.show()

Now I get the differential markers... and the color argument fixes everything as green (still not sure why it was multi-colored without this). I just need a line connecting the points, now, and everything would be good.

[Mark17]

Solved! Bottom line: need to do one ax.plot() for the line and a separate ax.plot() for the markers (points).