Snippet of drawing some MAICoS plots

[comcon1]

Reflecting the MAICoS profiles. Here is the thread for MAICoS profiles where we can share code snippets and plots.

[comcon1]

Let's start with this idiotic example.

%matplotlib widget
fig = plt.figure(1)
fig.clear()
ax = fig.subplots()
plt.show()

and the cycle looks like

co = 0
for s in ss:
    dieln = os.path.join(FMDL_SIMU_PATH, s['path'], 'DielectricWater_perp.json')
    if not os.path.isfile(dieln):
        continue
    st_props = os.stat(dieln)   
    if (st_props.st_size < 100):
        continue
    # plot some random selections
    if random.randint(1,5) > 1:
        continue
    
    with open(dieln) as ft:
        dft = pd.read_json(fp)
        dft.columns=['z', 'eps', 'ee']
        dfn = pd.read_json(ft)
        dfn.columns=['z', 'eps', 'ee']
    # stupid plot
    ax.plot(dfn['z'].tolist(), dfn['eps'].tolist(), alpha=.5, label=s['ID'])

    # pre-counter
    co += 1
    if co == 20: 
        break

ax.set_xlim(0, 5)
ax.set_ylabel('Epsilon (normal)')
ax.set_xlabel('z, nm')
ax.legend()

[comcon1]

[comcon1]

There is definitely some things to discuss already. Why is it negative? Should it be?

[PicoCentauri]

Yes the inverse perpendicular component (which is what you are plotting) can be negative. I would probably reduce the bin width as the profiles are very noisy.

[comcon1]

I wouldn't recalculate anything. Just can do smoothing instead. We all don't have much time

[PicoCentauri]

Yes this is what I mean.