# "ValueError: Mass matrix contains zeros on the diagonal" error, how to debug?

**URL:** <https://discourse.pymc.io/t/valueerror-mass-matrix-contains-zeros-on-the-diagonal-error-how-to-debug/4477>\
**Category:** Questions\
**Created:** [February 10, 2020, 7:30pm UTC](https://discourse.pymc.io/t/valueerror-mass-matrix-contains-zeros-on-the-diagonal-error-how-to-debug/4477 "2020-02-10T19:30:36Z")\
**Posts on this page:** 8\
**Page:** 1

<div class="post-metadata">

**Author:** ![hamburgled](https://yyz2.discourse-cdn.com/flex036/user_avatar/discourse.pymc.io/hamburgled/32/2480_2.png) [@hamburgled](https://discourse.pymc.io/u/hamburgled)\
**Post date:** [February 10, 2020, 7:30pm UTC](https://discourse.pymc.io/t/valueerror-mass-matrix-contains-zeros-on-the-diagonal-error-how-to-debug/4477/1 "2020-02-10T19:30:36Z")

</div>

Hi, I’m relatively new to pymc3 but hope someone can help. I have a very large model. When I run it for most of my datasets, it works fine and samples without problems.

However, with a few of my datasets, it works _intermittently._ It sometimes finishes successfully like the others, but often crashes with the “ValueError: Mass matrix contains zeros on the diagonal” error. It will also say:

```auto
The derivative of RV `myvar_log__`.ravel()[0] is zero.

```

for a bunch of the variables.

I’ve read a few threads here about this error, but I haven’t yet seen an explanation or a way of figuring out the cause of it. I know cause of the problem can probably be specific to my model, so I’m wondering about a general way of solving this: what does the error actually mean, and where can I look to debug it?

any tips are appreciated, thank you.

---

<div class="post-metadata">

**Author:** ![junpenglao](https://yyz2.discourse-cdn.com/flex036/user_avatar/discourse.pymc.io/junpenglao/32/8_2.png) [@junpenglao](https://discourse.pymc.io/u/junpenglao)\
**Post date:** [February 10, 2020, 8:04pm UTC](https://discourse.pymc.io/t/valueerror-mass-matrix-contains-zeros-on-the-diagonal-error-how-to-debug/4477/2 "2020-02-10T20:04:20Z")

</div>

You can find some pointer here: [Frequently Asked Questions](https://discourse.pymc.io/t/frequently-asked-questions/74/11)

---

<div class="post-metadata">

**Author:** ![hamburgled](https://yyz2.discourse-cdn.com/flex036/user_avatar/discourse.pymc.io/hamburgled/32/2480_2.png) [@hamburgled](https://discourse.pymc.io/u/hamburgled)\
**Post date:** [March 4, 2020, 5:47pm UTC](https://discourse.pymc.io/t/valueerror-mass-matrix-contains-zeros-on-the-diagonal-error-how-to-debug/4477/3 "2020-03-04T17:47:09Z")

</div>

Hi @junpenglao ,

thanks for the response. I read that thread. It appears to be giving a way to identify the RV causing the trouble, but I already know which it is (it’s in the error above).

Also, it seems like the error he had was a “bad initial energy: nan”, whereas I have the one with “derivative of RV is zero”. Are those actually the same ones?

I also have tried replacing ‘jitter+adapt\_diag’ with just ‘adapt\_diag’, but no change.

What’s strange is that I run the same model with several data sets, and they all work except for this one.

What steps can I try to take to diagnose the problem?

thank you!

---

<div class="post-metadata">

**Author:** ![AlexAndorra](https://yyz2.discourse-cdn.com/flex036/user_avatar/discourse.pymc.io/alexandorra/32/9142_2.png) [@AlexAndorra](https://discourse.pymc.io/u/AlexAndorra)\
**Post date:** [March 7, 2020, 4:19pm UTC](https://discourse.pymc.io/t/valueerror-mass-matrix-contains-zeros-on-the-diagonal-error-how-to-debug/4477/4 "2020-03-07T16:19:59Z")

</div>

Hi,  
Yeah I think usually these errors relate to the same problems – NaN values or zeros somewhere, making the log of these values unidentifed (if I remember correctly).

It’s hard to say without looking at the model, but have standardized your predictors / outcome variables? If they are on large scales, it can mess up sampling in exactly that way.

---

<div class="post-metadata">

**Author:** ![hamburgled](https://yyz2.discourse-cdn.com/flex036/user_avatar/discourse.pymc.io/hamburgled/32/2480_2.png) [@hamburgled](https://discourse.pymc.io/u/hamburgled)\
**Post date:** [March 7, 2020, 4:38pm UTC](https://discourse.pymc.io/t/valueerror-mass-matrix-contains-zeros-on-the-diagonal-error-how-to-debug/4477/5 "2020-03-07T16:38:03Z")

</div>

Hey,

yeah, unfortunately the model is very large at this point, fitting to a bunch of data, so it’s hard to provide a MWE ☹

However, one clue is that it works for all the other data sets, but for the one that it fails with, that dataset doesn’t have one of the observables (i.e., a “column” of the data) that the other ones all do. So, it’s underconstrained. My current theory is that because it’s underconstrained, the “energy landscape” might be really flat there, leading to 0 derivative?

What do you mean by standardized, do you mean normalizing them so that they have mean ~ 0 and SD ~ 1 or something?

So I did kinda get it to work, though it’s definitely hack-y. I used the `start` arg of `pm.sample()`, and use the last trace point of a previous successful run (with the same dataset), and it works. But that kinda depends on getting a successful run at some point…

---

<div class="post-metadata">

**Author:** ![AlexAndorra](https://yyz2.discourse-cdn.com/flex036/user_avatar/discourse.pymc.io/alexandorra/32/9142_2.png) [@AlexAndorra](https://discourse.pymc.io/u/AlexAndorra)\
**Post date:** [March 7, 2020, 4:54pm UTC](https://discourse.pymc.io/t/valueerror-mass-matrix-contains-zeros-on-the-diagonal-error-how-to-debug/4477/6 "2020-03-07T16:54:06Z")

</div>

> [@hamburgled](#):
>
> What do you mean by standardized, do you mean normalizing them so that they have mean ~ 0 and SD ~ 1 or something?

Yeah exactly – that can really do magic!  
Other than that, it’s difficult for me to say anything else without the model.  
Hope this helps anyway, and good luck!

---

<div class="post-metadata">

**Author:** ![hamburgled](https://yyz2.discourse-cdn.com/flex036/user_avatar/discourse.pymc.io/hamburgled/32/2480_2.png) [@hamburgled](https://discourse.pymc.io/u/hamburgled)\
**Post date:** [March 7, 2020, 5:56pm UTC](https://discourse.pymc.io/t/valueerror-mass-matrix-contains-zeros-on-the-diagonal-error-how-to-debug/4477/7 "2020-03-07T17:56:48Z")

</div>

ah yeah. I know that’s a standard for features is deep learning, but I hadn’t heard of using it here… it works well for MCMC too?

thanks!

---

<div class="post-metadata">

**Author:** ![AlexAndorra](https://yyz2.discourse-cdn.com/flex036/user_avatar/discourse.pymc.io/alexandorra/32/9142_2.png) [@AlexAndorra](https://discourse.pymc.io/u/AlexAndorra)\
**Post date:** [March 7, 2020, 6:37pm UTC](https://discourse.pymc.io/t/valueerror-mass-matrix-contains-zeros-on-the-diagonal-error-how-to-debug/4477/8 "2020-03-07T18:37:08Z")

</div>

Definitely works like a charm – should be the default behavior!