How to use the quantstats.stats.compsum function in QuantStats

fig.suptitle(title, fontsize=14, y=.995,
                 fontname=fontname, fontweight='bold', color='black')

    fig.set_facecolor('white')

    if subtitle:
        axes[0].set_title("\n%s - %s ;  Sharpe: %.2f                      " % (
            returns.index.date[:1][0].strftime('%e %b \'%y'),
            returns.index.date[-1:][0].strftime('%e %b \'%y'),
            _stats.sharpe(returns)
        ), fontsize=12, color='gray')

    axes[0].set_ylabel('Cumulative Return', fontname=fontname,
                       fontweight='bold', fontsize=12)
    axes[0].plot(_stats.compsum(returns) * 100, color=colors[1],
                 lw=1 if grayscale else lw, zorder=1)
    axes[0].axhline(0, color='silver', lw=1, zorder=0)

    dd = _stats.to_drawdown_series(returns) * 100
    ddmin = _utils._round_to_closest(abs(dd.min()), 5)
    ddmin_ticks = 5
    if ddmin > 50:
        ddmin_ticks = ddmin / 4
    elif ddmin > 20:
        ddmin_ticks = ddmin / 3
    ddmin_ticks = int(_utils._round_to_closest(ddmin_ticks, 5))

    # ddmin_ticks = int(_utils._round_to_closest(ddmin, 5))
    axes[1].set_ylabel('Drawdown', fontname=fontname,
                       fontweight='bold', fontsize=12)
    axes[1].set_yticks(_np.arange(-ddmin, 0, step=ddmin_ticks))

returns.fillna(0, inplace=True)
    if isinstance(benchmark, _pd.Series):
        benchmark.fillna(0, inplace=True)

    if match_volatility and benchmark is None:
        raise ValueError('match_volatility requires passing of '
                         'benchmark.')
    if match_volatility and benchmark is not None:
        bmark_vol = benchmark.std()
        returns = (returns / returns.std()) * bmark_vol

    # ---------------
    if compound is True:
        if cumulative:
            returns = _stats.compsum(returns)
            if isinstance(benchmark, _pd.Series):
                benchmark = _stats.compsum(benchmark)
        else:
            returns = returns.cumsum()
            if isinstance(benchmark, _pd.Series):
                benchmark = benchmark.cumsum()

    if resample:
        returns = returns.resample(resample)
        returns = returns.last() if compound is True else returns.sum()
        if isinstance(benchmark, _pd.Series):
            benchmark = benchmark.resample(resample)
            benchmark = benchmark.last(
            ) if compound is True else benchmark.sum()
    # ---------------

if isinstance(benchmark, _pd.Series):
        benchmark.fillna(0, inplace=True)

    if match_volatility and benchmark is None:
        raise ValueError('match_volatility requires passing of '
                         'benchmark.')
    if match_volatility and benchmark is not None:
        bmark_vol = benchmark.std()
        returns = (returns / returns.std()) * bmark_vol

    # ---------------
    if compound is True:
        if cumulative:
            returns = _stats.compsum(returns)
            if isinstance(benchmark, _pd.Series):
                benchmark = _stats.compsum(benchmark)
        else:
            returns = returns.cumsum()
            if isinstance(benchmark, _pd.Series):
                benchmark = benchmark.cumsum()

    if resample:
        returns = returns.resample(resample)
        returns = returns.last() if compound is True else returns.sum()
        if isinstance(benchmark, _pd.Series):
            benchmark = benchmark.resample(resample)
            benchmark = benchmark.last(
            ) if compound is True else benchmark.sum()
    # ---------------

    fig, ax = _plt.subplots(figsize=figsize)
    ax.spines['top'].set_visible(False)

def to_prices(returns, base=1e5):
    """ Converts returns series to price data """
    returns = returns.copy().fillna(0).replace(
        [_np.inf, -_np.inf], float('NaN'))

    return base + base * _stats.compsum(returns)

hlw=1.5, hlcolor="red", hllabel="",
                   match_volatility=False,
                   log_scale=False, figsize=(10, 5), ylabel=True,
                   subtitle=True, compounded=True,
                   savefig=None, show=True):

    title = 'EOY Returns'
    if benchmark is not None:
        title += '  vs Benchmark'
        benchmark = _utils._prepare_benchmark(
            benchmark, returns.index).resample('A').apply(
                _stats.compsum).resample('A').last()

    returns = _utils._prepare_returns(returns).resample('A')
    if compounded:
        returns = returns.apply(_stats.compsum)
    else:
        returns = returns.apply(_df.cumsum)
    returns = returns.resample('A').last()

    _core.plot_returns_bars(returns, benchmark,
                            fontname=fontname,
                            hline=returns.mean(),
                            hlw=hlw,
                            hllabel=hllabel,
                            hlcolor=hlcolor,
                            match_volatility=match_volatility,
                            log_scale=log_scale,
                            resample=None,
                            title=title,
                            figsize=figsize,
                            grayscale=grayscale,