Fixes the difference b/w logpt and logp_nojac #3398
Conversation
|
@junpenglao please review. |
| def logp_nojact(self): | ||
| """Theano scalar of log-probability of the model""" | ||
| """Theano scalar of log-probability of the model but without the jacobian | ||
| if transformed Random Variable is presented. |
There was a problem hiding this comment.
Does this mean that the Jacobian will be present if there are no transformed random variables? That is how I understand the sentence.
There was a problem hiding this comment.
If there are no transformed variables there is no need for Jacobian correction, hence logp_nojact will be the same as logpt
This should be rephrased to make it clearer indeed.
There was a problem hiding this comment.
| if transformed Random Variable is presented. | |
| if transformed Random Variable is presented. | |
| Note that If there is no transformed variable in the model, logp_nojact | |
| will be the same as logpt as there is no need for Jacobian correction. |
| def logp_nojact(self): | ||
| """Theano scalar of log-probability, excluding jacobian terms.""" | ||
| """Theano scalar of log-probability of the model but without the jacobian | ||
| if transformed Random Variable is presented. |
There was a problem hiding this comment.
| if transformed Random Variable is presented. | |
| if transformed Random Variable is presented. | |
| Note that If there is no transformed variable in the model, logp_nojact | |
| will be the same as logpt as there is no need for Jacobian correction. |
| def logp_nojact(self): | ||
| """Theano scalar of log-probability of the model""" | ||
| """Theano scalar of log-probability of the model but without the jacobian | ||
| if transformed Random Variable is presented. |
There was a problem hiding this comment.
| if transformed Random Variable is presented. | |
| if transformed Random Variable is presented. | |
| Note that If there is no transformed variable in the model, logp_nojact | |
| will be the same as logpt as there is no need for Jacobian correction. |
|
@divyanshu132 Do you still follow up on this question? If you don't have time, I can submit it again on your pr. I think this problem is very small, but changing it as soon as possible is very friendly to the user. |
|
@chang111 Actually I'm busy with some other organization, feel free to take it! |
|
@divyanshu132 Thanks You very much. I will push pr through your branch for you have made so much attempt for it. |
|
@junpenglao Can you tell me how to pull request on other's branch. I am sorry for that I do not know how to do it. |
|
fork the branch |
|
Thank you, I will try it today. |
4 participants
Fixes #3372
Previously:
Now: