Matplotlib savefig() saving plots as blank images-CodePudding

I'm trying to save graphs in folders I'm creating and they all show within Jupyter notebook but save as blank images within my folders. This is my code:

    munilist = ["Adjuntas", "Anasco", "Ciales", "Jayuya", "Lares", "LasMarias", "Maricao", "Mayaguez", "Orocovis", "Penuelas", "Ponce", "SabanaGrande", "SanGerman", "SanSebastian", "Utuado", "Yauco"]
    for municipality in munilist:
        
        x = np.array([1987, 1992, 1998, 2002, 2007, 2012]).reshape(6,1)
        y = np.array(df[df["Municipio"]==municipality].iloc[0, 1:7]).reshape(6,1)
        print(y)
        mask = ~pd.isnull(y) #creating a mask to get rid of NaN columns values in certain csvs
        print(mask)
        xlin = np.arange(1987, 2013,1) #range of years to plot
        reg = LinearRegression(fit_intercept=True).fit(x[mask].reshape(-1,1), y[mask])
        a0 = reg.intercept_
        a1 = reg.coef_[0]

        fit = a0   a1*xlin #fitted value

        b0 = float(a0) #f strings do not accept array. Need to convert to float
        b1 = float(a1)
        eq = f"$farms = ${b0:.3f} $ $ {b1:.3f}$*year$"

        plt.scatter(x,y, label = name)
        plt.title(municipality)
        plt.plot(xlin, fit, 'r', label = "fitted line")
        plt.xlabel("Year")
        plt.ylabel(name)
        plt.text(2000,1000, eq)
        plt.legend()
        plt.figure(figsize=(3, 3)) #had to change figure size because they were coming out way too large
        if not os.path.isdir("RegressionsOutput/"   municipality):
            os.makedirs("RegressionsOutput/"   municipality)
        plt.savefig("RegressionsOutput/"   municipality   "/"   name   '.png', format="png", dpi=300, bbox_inches = 'tight') #save before showing
        plt.show()

This is what saves in my folders

CodePudding user response：

You're calling plt.figure(figsize=...) after having done all of the work. This will reset the figure. Just move this line to the top so that it looks like

eq = f"..."
plt.figure(figsize=(3,3))
...
plt.savefig(...)

I'm 95% sure this is your problem.

plt.figure() is designed to be used to create a new figure, so you will be overwriting any work you've done.

CodePudding user response：

You should explicitly define the figure (e.g. "fig") initially (i.e. prior to plt.scatter()):

fig = plt.figure(figsize=(3, 3), num=1, clear=True)

and then show:

plt.show()

and then save:

fig.savefig("RegressionsOutput/"   municipality   "/"   name   '.png', ...)