Create a gist now

Instantly share code, notes, and snippets.

What would you like to do?
Bayesian Survival analysis with PyMC3
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.

guyyos commented Dec 6, 2015

Hi, I try to run your notebook. But I keep getting :

np.exp(trace['beta'].mean())
1.0
instead of 1.645592148084472 , as your result in cell 17. In the previous cells I'm getting same results.
Do you know what could be wrong?
Thanks,
Guy

@guyyos FWIW I am getting similar results, trying to debug what is going on.

I as well am receiving these results

ppope commented Apr 19, 2016

Same here! @AustinRochford any advice on how to get this to work?

I get the same. Updated all installs, etc. Looks like a bug now is not being addressed.
Values in trace['beta'] all come out to be 0.0
Interestingly, trace['mu_beta'] and trace['tau'] all look to have values. So, seemingly, the model description is not being processed correctly.
Anyway, something looks to be quite broken.

ibarrien commented Jan 14, 2017 edited

I managed to reproduce all posted results, including:

np.exp(trace['beta'].mean())
1.6455921480844671

Some info:
Spyder Python 2.7 on a Mac OS X
package, version:
numpy, 1.11.2
pymc3, 3.0
statsmodels, 0.6.1
theano, 0.8.2
(pandas, 0.19.1)

LeoHdez commented Jul 28, 2017

Same problem:
np.exp(trace['beta'].mean()) = 1.0404104237122036
Beta plot, autocorrelation plot, Cumulative hazard and Survival function are different from your notebook (although consistent with each other.
All results from section "Time varying effects" are identical to yours

python version: 2.7.13 |Anaconda custom (x86_64)| (default, Dec 20 2016, 23:05:08)
[GCC 4.2.1 Compatible Apple LLVM 6.0 (clang-600.0.57)]
numpy version: 1.12.1
pymc3 version: 3.0
statsmodels version: 0.8.0
theano version: 0.9.0

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment