Generalize plotting script for sigma tau. by jwbishop · Pull Request #11 · uafgeotools/array_processing

jwbishop · 2019-11-25T03:52:20Z

Note changes to the arguments!

Flag for sigma_tau added. Additionally, optional flag for plotting the M(d)CCM on it's own subplot also added. Tested and consistent with wlsqva_proc.

Note: This update isn't 100% complete. I'm note sure yet what the (currently silent) argument of mcthresh does.

This reverts commit cdde0a8.

This reverts commit 7340af0.

jwbishop · 2019-11-25T03:56:45Z

Concerning mcthresh, it looks like the indices of array_thresh would be helpful?

liamtoney

@jwbishop I made a few mostly minor suggestions. We probably should go thru and change every instance of az to baz in the array plotting script...

array_processing/tools/plotting.py

liamtoney · 2019-11-25T05:39:46Z

array_processing/tools/plotting.py

+    vplot = 1
+    bplot = 2
+    splot = bplot
+    if ccmplot is not False:


Suggested change

if ccmplot is not False:

if ccmplot:

Good catch.

liamtoney · 2019-11-25T05:42:16Z

array_processing/tools/plotting.py

+
+    # Start Plotting
+    # Plot Trace
+    fig1, axarr1 = plt.subplots(num_subplots, 1, sharex='col')


Suggested change

fig1, axarr1 = plt.subplots(num_subplots, 1, sharex='col')

fig1, axs1 = plt.subplots(num_subplots, 1, sharex='col')

liamtoney · 2019-11-25T05:43:13Z

array_processing/tools/plotting.py

+    # Plot Trace
+    fig1, axarr1 = plt.subplots(num_subplots, 1, sharex='col')
+    fig1.set_size_inches(10, 9)
+    axs1 = axarr1.ravel()


Suggested change

axs1 = axarr1.ravel()

liamtoney · 2019-11-25T05:45:08Z

array_processing/tools/plotting.py

+    # Colormap
+    cm = 'RdYlBu_r'
+    # Colorbar/y-axis for MdCCM
+    cax = 0.2, 1


Suggested change

cax = 0.2, 1

cax = (0.2, 1)

This just makes things a little more clear

liamtoney · 2019-11-25T05:49:02Z

array_processing/tools/plotting.py

-    axs1[2].set_ylim(.25,.45)
-    axs1[2].set_xlim(t[0],t[-1])
+    # Plot MdCCM on its own plot
+    if ccmplot is not False:


Suggested change

if ccmplot is not False:

if ccmplot:

liamtoney · 2019-11-25T05:52:39Z

array_processing/tools/plotting.py

+        baz: Array of back-azimuth estimates
+        ccmplot: Flag to plot the Mean/Median cross-correlation
+            maxima values on the y-axis in addition to the color scale.
+        sigma_tau: Array of sigma_tau values


Is sigma_tau just a boolean? This docstring entry is confusing. If boolean, it should be default False. If it can be an array, it should be default None.

It should be an array

liamtoney · 2019-11-25T05:53:07Z

array_processing/tools/plotting.py

+        t: Array processing time vector
+        mdccm: Median cross-correlation maxima
+        vel: Array of trace velocity estimates
+        baz: Array of back-azimuth estimates


We have baz here, but az in the function definition and elsewhere...

liamtoney · 2019-11-25T05:55:05Z

array_processing/tools/plotting.py

+    axs1[bplot].set_ylabel('Back-azimuth\n [deg]')
+
+    # Plot sigma_tau
+    if sigma_tau is not False:


Suggested change

if sigma_tau is not False:

if sigma_tau:

liamtoney · 2019-11-25T05:55:54Z

example.py

@@ -1,4 +1,4 @@
-#%% User-defined parameters
+# %% User-defined parameters


Does this break the iPython cell?

Fixed in update.

liamtoney · 2019-11-25T06:04:30Z

Note that this addresses #8

davidfee5 · 2019-11-25T16:21:30Z

This looks good to me and I can review it a bit more. Question though: should we try to generalize to accept LTS-flagged elements as well? Perhaps that can wait until the LTS code revisions are approved for public release.

davidfee5 · 2019-11-25T16:25:11Z

array_processing/tools/plotting.py

-    example: array_plot(stf,tvec,t,mdccm,vel,az,mcthresh):
+    Generalized plotting script for velocity - back-azimuth array processing.
+
+    @ Author: David Fee


Remove all author names in functions.

I was wondering about this, ok!

davidfee5

Overall looks good to me. Thanks!

davidfee5 · 2019-11-25T16:28:23Z

array_processing/tools/plotting.py

+    fig1, axarr1 = plt.subplots(num_subplots, 1, sharex='col')
+    fig1.set_size_inches(10, 9)
+    axs1 = axarr1.ravel()
+    axs1[0].plot(tvec, st[0].data, 'k')


We could also add an option to plot the beam for a specified vel and baz, but perhaps that is best left for a future change

davidfee5 · 2019-11-25T16:29:18Z

array_processing/tools/plotting.py

+    if ccmplot is not False:
+        sc = axs1[1].scatter(t, mdccm, c=mdccm,
+                             edgecolors='k', lw=0.3, cmap=cm)
+        # axs1[1].plot([t[0], t[-1]], [mcthresh[0], mcthresh[1]], 'r--')


davidfee5 · 2019-11-25T16:30:53Z

array_processing/tools/plotting.py

+        mdccm: Median cross-correlation maxima
+        vel: Array of trace velocity estimates
+        baz: Array of back-azimuth estimates
+        ccmplot: Flag to plot the Mean/Median cross-correlation


Should it only be median xcorr? The rest of the code assumes it is mdccm. Is there a time we would want mccm? Probably not. Future option to think of: semblance instead of mdccm.

Should it only be median xcorr? The rest of the code assumes it is mdccm. Is there a time we would want mccm? Probably not. Future option to think of: semblance instead of mdccm.

I'm actually a bit fan of the fisher statistics as well. We could modify to something like detection_metric. The code is agnostic to the actual meaning of that array. It's currently just used for the colorbar.

jwbishop · 2019-11-25T16:40:50Z

This looks good to me and I can review it a bit more. Question though: should we try to generalize to accept LTS-flagged elements as well? Perhaps that can wait until the LTS code revisions are approved for public release.

I thought about adding that as well, but it I think it's best to initially make the simple generalization. Adding LTS plotting functionality adds two more import statements and a minimum 20 code lines.

… to full sentences.

Jordan W Bishop added 6 commits November 23, 2019 15:03

Fix Linter warnings

0c032b4

Add sigma_tau plotting option to plotting script

cdde0a8

Revert "Add sigma_tau plotting option to plotting script"

7340af0

This reverts commit cdde0a8.

Revert "Revert "Add sigma_tau plotting option to plotting script""

5a6bd97

This reverts commit 7340af0.

Fix colorbar boundary and logicals for plotting parameters

bb271dd

Update example.py with new plot function

21df94b

liamtoney requested changes Nov 25, 2019

View reviewed changes

davidfee5 reviewed Nov 25, 2019

View reviewed changes

davidfee5 requested changes Nov 25, 2019

View reviewed changes

jwbishop mentioned this pull request Nov 25, 2019

Add an option to plot the beam for a specified vel and baz #12

Open

Fix pr_request comments. Fix sigma_tau plot bus. Update code comments…

1937a3d

… to full sentences.

jwbishop closed this Nov 25, 2019

jwbishop mentioned this pull request Nov 25, 2019

Fix for #8, Add sigma tau functionality to array_plot function. #13

Merged

	fig1, axarr1 = plt.subplots(num_subplots, 1, sharex='col')
	fig1, axs1 = plt.subplots(num_subplots, 1, sharex='col')

		@@ -1,4 +1,4 @@
		#%% User-defined parameters
		# %% User-defined parameters

Conversation

jwbishop commented Nov 25, 2019 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

jwbishop commented Nov 25, 2019

Uh oh!

liamtoney left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

liamtoney commented Nov 25, 2019

Uh oh!

davidfee5 commented Nov 25, 2019

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

davidfee5 left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

jwbishop Nov 25, 2019 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

jwbishop commented Nov 25, 2019

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

jwbishop commented Nov 25, 2019 •

edited

Loading

jwbishop Nov 25, 2019 •

edited

Loading