Hi, could someone explain what is exactly the average loss that appears in the advacement bar during the initialisation (with ADVI)? What behaviour should we expect from that number? Average Loss = 2.1759e+07: 100%|██████████| 10000/10000 [02:11<00:00, 89.31it/s] I’m new to variational inference, …

Hi @mcavallaro , in ADVI the loss is the negative of evidence lower bound (ELBO) within some window. You can not compare it across model, as it is non-normalized. What people usually do is plot it so you can check whether your model converge or not (converge to a local minimum at least). eg: http://d…

Great, thanks @junpenglao ! Can I print the curren value of ELBO without running any variational fit? This would help to debug the model, I think. EDIT: what concerned me is that I get inf as here. Average Loss = inf: 100%|█████████████████████████| 1000/1000 [01:48<00:00, 9.18it/s] Finished [10…

@mcavallaro Maybe this is helpful: http://docs.pymc.io/notebooks/howto_debugging.html The model could be misspecified and seeing which values lead to infs in the logp might be enlightening.

@twiecki Thanks! Good to know that there is a canonical way to debug theano functions! In my case the loglikelihoods are finite, still, it’s strange that the Average Loss in my post above is reported to be infinite first and then equal to 1,162.4.

You can first check your model to make sure the logp is not inf: [image] Get nan or inf from model.logp (model.test point) is an attestation of incorrectly configured model? Questions A few thoughts: 1, In your model, the observed part you are doing: with model: ... …

It’s not uncommon for the ELBO to be inf initially and the converge, I wouldn’t worry about it.

thanks @twiecki Still I’m confused. Running with model: f = pm.fit() I obtain that the trace for the ELBO, f.hist contains lots of inf, (just inspecting np.where(f.hist== np.inf) , see the orange points in the pic below), but no warnings or errors… [ADVI_hist] Conversely, with model: s = …

Yes, you can track the parameters following this doc: http://docs.pymc.io/notebooks/variational_api_quickstart.html#Tracking-parameters [image] mcavallaro: eventually raises the bad initialisation / model mispecification error I think this is an indication of some hidden problem in your mode…

Thanks! Tracking the parameters reveals that none of them goes to infinite during ADVI, so I’m still looking for the problem. [image] junpenglao: I think this is an indication of some hidden problem in your model (e.g., some parameter goes negative and then feed to something needs a positive in…

Hi everyone, I was experiencing a similar average loss inf problem in some of my models since updating to 3.2 and was able to recreate it in an extremely simple regression model (the models didn’t produce this in earlier versions of pymc3). It appears as though the model converges but then produces…

Average Loss in optimization output

Questions

junpenglao September 28, 2017, 12:26pm 6

You can first check your model to make sure the logp is not inf:

Topic		Replies	Views
Infinite loss with ADVI Questions	1	1133	January 24, 2018
Inf Average loss ADVI in Correlated Topic Model Questions	9	1593	July 25, 2018
Negative "Average loss" in ADVI Questions	4	1438	April 15, 2019
Negative loss (Stochastic Volatility model)	2	63	October 7, 2025
Average loss in ADVI optimization Questions	3	1417	July 23, 2018

Average Loss in optimization output

Related topics