Jupyter notebook for a Bayesian inversion of critical wedge parameters

bayes_wedge_inversion.ipynb

Hi, if it doesn't work, try calling it with calc_row_likelihood instead of calc_row_likelihood_parallel and see if that works.

Thank you for your response, Im still blocked. [image: image.png] Im write from Colombia, and try to run the same for La Sierra Nevada de Santa Marta. [image: image.png] [image: image.png] Thanks a lot. El mié., 18 dic. 2019 a las 22:48, Richard Styron (<[email protected]>) escribió:

…

Hi, if it doesn't work, try calling it with calc_row_likelihood instead of calc_row_likelihood_parallel and see if that works. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <https://gist.github.com/bb042e307e21134adba497179ca7f35e?email_source=notifications&email_token=AHWRTL723W3HAVLE5R5G2YLQZLVJDA5CNFSM4J4AVRAKYY3PNVWWK3TUL52HS4DFVNDWS43UINXW23LFNZ2KUY3PNVWWK3TUL5UWJTQAF6FLC#gistcomment-3115697>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AHWRTL2O24J7J5FD67OVXHDQZLVJDANCNFSM4J4AVRAA> .

Hi Virgilio,

The images didn't upload. Can you copy and paste the error message instead of posting a screenshot?

Dr. Richard, Scuse me,.. I ran the following. ___ t_start = time.time() #priors = calc_row_likelihood_parallel(priors) priors = calc_row_likelihood(priors) t_end = time.time() print('done with {} calcs in {:.1f} seconds'.format(n_iters, (t_end - t_start)))

…

____ done with 100000 calcs in 0.1 seconds C:\ProgramData\Anaconda3\lib\site-packages\scipy\stats\stats.py:308: RuntimeWarning: divide by zero encountered in log log_a = np.log(np.array(a, dtype=dtype))

---------------------------------------------------------------------------IndexError Traceback (most recent call last)<ipython-input-44-e539dce3a875> in <module>----> 1 priors['rel_likelihood'] = priors['likelihood'] / priors['likelihood'].max() IndexError: invalid index to scalar variable.

__________________________________________________________________________ priors.head()

---------------------------------------------------------------------------AttributeError Traceback (most recent call last)<ipython-input-43-4bc6fc59de5c> in <module>----> 1 priors.head() AttributeError: 'numpy.float64' object has no attribute 'head'

_______________________ El vie., 20 dic. 2019 a las 10:54, Richard Styron (<[email protected]>) escribió:

Hi Virgilio, The images didn't upload. Can you copy and paste the error message instead of posting a screenshot? — You are receiving this because you commented. Reply to this email directly, view it on GitHub <https://gist.github.com/bb042e307e21134adba497179ca7f35e?email_source=notifications&email_token=AHWRTL6BZEOJVOC6IZZ25FTQZTTCDA5CNFSM4J4AVRAKYY3PNVWWK3TUL52HS4DFVNDWS43UINXW23LFNZ2KUY3PNVWWK3TUL5UWJTQAF6INM#gistcomment-3117270>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AHWRTL2BWFG3YBI57P65RM3QZTTCDANCNFSM4J4AVRAA> .

Sorry, again...

=========
t_start = time.time()

priors = calc_row_likelihood(priors)
t_end = time.time()

print('done with {} calcs in {:.1f} seconds'.format(n_iters, (t_end - t_start)))

done with 100000 calcs in 0.1 seconds

---

C:\Users\vamaris\AppData\Local\Continuum\anaconda3\lib\site-packages\scipy\stats\stats.py:308: RuntimeWarning: divide by zero encountered in log
log_a = np.log(np.array(a, dtype=dtype))

priors['rel_likelihood'] = priors['likelihood'] / priors['likelihood'].max()

---

IndexError Traceback (most recent call last)
in
----> 1 priors['rel_likelihood'] = priors['likelihood'] / priors['likelihood'].max()

IndexError: invalid index to scalar variable.

priors.head()

---

AttributeError Traceback (most recent call last)
in
----> 1 priors.head()

AttributeError: 'numpy.float64' object has no attribute 'head'

---

As you see Im here.

I see the problem now.

You need to run something like

priors['likelihood'] = priors.apply(calc_row_likelihood, axis=1)

instead of

priors = calc_row_likelihood(priors)

If it doesn't work exactly, read the error messages and try to understand them to find the problem.

Ok. Perfect.

Thank you. It work correctly.

Great, I am glad I could help. Good luck with your research!

Thank you. El vie., 17 ene. 2020 a las 9:50, Richard Styron (<[email protected]>) escribió:

…

Great, I am glad I could help. Good luck with your research! — You are receiving this because you commented. Reply to this email directly, view it on GitHub <https://gist.github.com/bb042e307e21134adba497179ca7f35e?email_source=notifications&email_token=AHWRTLYOYUZYJI54K4CJRDDQ6HATRA5CNFSM4J4AVRAKYY3PNVWWK3TUL52HS4DFVNDWS43UINXW23LFNZ2KUY3PNVWWK3TUL5UWJTQAF7XZM#gistcomment-3141526>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AHWRTLY5JXWKEUHH2JHMVLDQ6HATRANCNFSM4J4AVRAA> .