Let’s say I have a simple model with shared variables to make out-of-sample predictions: N=5000 p=2 batch_size=100 X=np.random.normal(0,1,(N,p)) b=np.random.uniform(0,0.1,p) y=np.random.poisson(np.exp(X.dot(b)),N) model = pm.Model() with model: x0_s=shared(X[:,0]) x1_s=shared(X[:,1]) b0=p…

Did you try pickling them? Save them with the approx variable and set the values before you call approx.sample(). For me this works just fine. x0_s.set_value(newX0) x1_s.set_value(newX1) trace = approx.sample(1000)

Save and reuse model with shared variables

Questions

mgregori June 16, 2018, 6:07am 3

You are right! The easiest solution is the one that works

Thank you.

Topic		Replies	Views
Predicting from a pickled model Questions	2	487	April 7, 2019
Theano shared and prediction not working as expected Questions	2	667	July 13, 2019
Out of sample predictions from a pickled model Questions theano	14	2728	January 31, 2019
Testing hiearchical model on unseen data Questions	3	821	January 4, 2018
PyMC3: How to compute MAP Estimates Repeatedly and Efficiently? Questions	2	2617	July 25, 2017

Save and reuse model with shared variables

Related topics