Matplotlib helpers

matplotlib_helper.py

# Refs:
# https://matplotlib.org/3.9.3/users/explain/quick_start.html
# https://stackoverflow.com/questions/7908636/how-to-add-hovering-annotations-to-a-plot
# https://www.geeksforgeeks.org/how-to-draw-a-line-inside-a-scatter-plot/
# https://python-graph-gallery.com/scatterplot-with-regression-fit-in-matplotlib/
# https://stackoverflow.com/questions/22239691/code-for-best-fit-straight-line-of-a-scatter-plot
# https://stackoverflow.com/questions/27878217/how-do-i-extend-the-margin-at-the-bottom-of-a-figure-in-matplotlib
# https://stackoverflow.com/questions/51473993/plot-an-histogram-with-y-axis-as-percentage-using-funcformatter

def plot_cols(inp_df: pandas.DataFrame, *args):
    plt.figure(figsize=(10, 10))
    for x in args:
        plt.plot(inp_df[X_AXIS_KEY], inp_df[x], label=x)
    plt.xlabel("Graph Name")
    plt.ylabel("Values")
    plt.legend()
    plt.xticks(rotation=90)
    plt.subplots_adjust(bottom=0.7)
    plt.show()

    
def plot_scatter_y_x(inp_df: pandas.DataFrame, col_y: str, col_x: str):
    plt.figure(figsize=(10, 10))

    slope, intercept = numpy.polyfit(inp_df[col_y], inp_df[col_x], 1)
    line = slope * inp_df[col_y] + intercept

    sc = plt.scatter(inp_df[col_y], inp_df[col_x])
    fig = plt.figure(1)
    ax  = fig.axes[0]

    plt.plot(inp_df[col_y], line, label=f"Linear best-fit: Slope: {slope:,.2f} Offset: {intercept:,.2e}")
    plt.xlabel(col_x)
    plt.ylabel(col_y)
    plt.legend()
    annot = ax.annotate("",
                        xy=(0,0),
                        xytext=(-20,-20),
                        textcoords="offset points",
                        # arrowprops=dict(arrowstyle="->"),
                        bbox=dict(boxstyle="round", fc="w")
                        )
    annot.set_visible(False)
    xlim = ax.get_xlim()
    ylim = ax.get_ylim()
    xl = (xlim[1] - xlim[0]) * 0.2 + xlim[0]
    xr = (xlim[1] - xlim[0]) * 0.8 + xlim[0]
    yt = (ylim[1] - ylim[0]) * 0.2 + ylim[0]
    yb = (ylim[1] - ylim[0]) * 0.8 + ylim[0]


    def update_annot(ind):
        data_idx = ind["ind"][0]
        pos = sc.get_offsets()[data_idx]

        df_entry   = inp_df.iloc[data_idx]
        graph_name = df_entry[X_AXIS_KEY]
        xval       = df_entry[col_y]
        yval       = df_entry[col_x]
        out_text   = graph_name
        out_text  += f"\n{xval:,.2e}, {yval:,.2e}"
        out_text  += f"\n{xval:,.2f}, {yval:,.2f}"

        out_pos = pos
        if pos[0] > xr:
            out_pos = (out_pos[0] - 100, out_pos[1])
        if pos[1] > yt:
            out_pos = (out_pos[0], out_pos[1] - 100)
        if pos[0] < xl:
            out_pos = (out_pos[0] + 20, out_pos[1])
        if pos[1] < yb:
            out_pos = (out_pos[0], out_pos[1] + 20)

        annot.set_text(out_text)
        annot.xy = out_pos

    def hover(event):
        vis = annot.get_visible()
        if event.inaxes == ax:
            cont, ind = sc.contains(event)
            if cont:
                # breakpoint()
                update_annot(ind)
                annot.set_visible(True)
                fig.canvas.draw_idle()
            else:
                if vis:
                    annot.set_visible(False)
                    fig.canvas.draw_idle()

    fig.canvas.mpl_connect("motion_notify_event", hover)

    plt.show()


def plot_hist(inp_df: pandas.DataFrame, col_x: str, bins=5):
    plt.figure(figsize=(10, 5))

    plt.subplot(1, 2, 1)
    plt.hist(inp_df[col_x], bins=bins, edgecolor='black')
    plt.xlabel(col_x)
    plt.ylabel("Frequency")

    plt.subplot(1, 2, 2)
    n, bins, patches = plt.hist(
        inp_df[col_x],
        bins=bins,
        edgecolor='black',
        weights=[1/len(inp_df[col_x])]*len(inp_df[col_x])
    )
    plt.xlabel(col_x)
    plt.ylabel("Percentage")

    plt.tight_layout()

    # Display the plot
    plt.show()