Python matplotlib changing the major unit scale of the x-axis

Question

I have a figure that isn't showing the correct dates on the x-axis. Possibly because there are too many observations to show and it results in not showing any. I am not sure if I should use the set_xticks function, but it gives me the attribute error

AttributeError: module 'matplotlib.pyplot' has no attribute 'set_xticks'

My current code is this:

def plot_w(dataframe,ticker,benchmark):

    # a. Printing highest observed values and corresponding date
    max1 = data_df.loc[:, ticker].max()
    max2 = data_df.loc[:, benchmark].max()
    max1_date = data_df[data_df[ticker] == max1]['Date'].values[0] 
    max2_date = data_df[data_df[benchmark] == max2]['Date'].values[0]
    print("The highest adjusted close price observed at: \n", ticker, ":", max1.round(2), "USD on the date ", max1_date, 
          "\n", benchmark, ":", max2.round(2), "USD on the date", max2_date)

    # b. Setting up plot based on dropdown input
    I = data_df.columns == ticker
    mpl_figure = dataframe.loc[:, ['Date',ticker,benchmark]]
    mpl_figure.plot(x='Date', y=[ticker,benchmark], style=['-b','-k'], figsize=(10, 5), fontsize=11, legend='true', linestyle = '-')
    plt.ylabel("USD",labelpad=5)
    plt.locator_params(axis='x', nbins=20)
    title = "Adjusted close prices for " + ticker + " and " + benchmark
    plt.title(title)
    plt.set_xticks(data_df['Date'].values) # Code fails here

# c. Creating the widget for the plot
widgets.interact(plot_w,
    dataframe = widgets.fixed(data_df),
    ticker = widgets.Dropdown(
            options=data_df.columns,
            value='ATVI',
            description='Company 1:',
            disabled=False,
        ),
    benchmark = widgets.Dropdown(
            options=data_df.columns,
            value='AAPL',
            description='Company 2:',
            disabled=False,
        )
)

The figure looks like this:

Answer 1

set_xticks is for an axis, if you are setting ticks for a figure it's

plt.xticks(data_df['Date'].values)

Answer 2

Alternatively, you can try the following inside your function working on an axis object. The idea is to first create an axis instance ax, and then pass it to the plot command. Later use it to set the x-ticks.

I = data_df.columns == ticker
fig, ax = plt.subplots(figsize=(10, 5))
mpl_figure = dataframe.loc[:, ['Date',ticker,benchmark]]
mpl_figure.plot(x='Date', y=[ticker,benchmark], style=['-b','-k'], ax=ax, fontsize=11, legend='true', linestyle = '-')
plt.ylabel("USD",labelpad=5)
plt.locator_params(axis='x', nbins=20)
title = "Adjusted close prices for " + ticker + " and " + benchmark
plt.title(title)
ax.set_xticks(data_df['Date'].values) 

Answer 3

The module matplotlib.pyplot has no set_xticks function, but an xticks function instead. (link to doc). However, the class object matplotlib.axes.Axes does have set_xticks (link to doc).

The easiest fix in your code is thus

plt.xticks(data_df['Date'].values)

Side note: I am not entirely sure why matplotlib.pyplot keeps around two functions with (almost) the same effect but different names. I guess allowing it to be called from either the module or the object is an imitation of MATLAB, but in MATLAB the function name is the same.