CODEWITHSUNDEEP
pythonpandasmatplotlib
Sigmarabi1
Sigmarabi1

Reputation: 25

Python matplotlib how to display more ticks

The following code generates a graph with few ticks but I need to display a tick every 4 or 6 hours. How can that we accomplished?

import matplotlib.dates as mdates
# Plot train and test so you can see where we have split
gdf = x_test[['AllAPIs']] \
    .rename(columns={'AllAPIs': 'Test Set'}) \
    .join(x_train.rename(columns={'AllAPIs': 'Training Set'}),
          how='outer')
ax = gdf.plot(figsize=(12,5),title='API Calls')
plt.xticks(
    rotation=45,
    horizontalalignment='right',
    fontweight='light',
    fontsize='medium',
)
plt.show()

The output graph is this: enter image description here

If I add the HourLocator as follows, I lose my ticks:

# set monthly locator
ax.xaxis.set_major_locator(mdates.HourLocator(interval=4))
# set formatter
ax.xaxis.set_major_formatter(mdates.DateFormatter('%Y-%m-%d %H'))

enter image description here

What am I missing?????

Upvotes: 0

Views: 303

Answers (1)

jcaliz
jcaliz

Reputation: 4021

Why don't you set the ticks from the plt object?

plt.xticks(
    ticks = gdf.index.values
    rotation=45,
    horizontalalignment='right',
    fontweight='light',
    fontsize='medium',
)

This might set a lot of values, however, you can easily construct the xticks parameter based on the index.

For example, asumming index values are sorted:

start = gdf.index.values[0]
end = gdf.index.values[-1]

ticks = pd.date_range(start, end, freq='4H')
labels = [x.strftime('%Y-%m-%d %H') for x in ticks]

plt.xtics(ticks=ticks, labels=labels)

Upvotes: 1

Related Questions