Skip to content

improve prior comment on GP Birthdays data set. #704

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Sep 21, 2024
Merged

improve prior comment on GP Birthdays data set. #704

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion examples/gaussian_processes/GP-Births.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -997,7 +997,7 @@
"id": "0e8f5c41",
"metadata": {},
"source": [
"The motivation is that we have around $7.3$K data points and we want to consider the in between data points distance in the normalized (log) scale. That is why we consider the ratio `7_000 / time_str`. Note that we want to capture the long term trend, so we want to consider a length scale that is larger than the data points distance. We increase the order of magnitude by dividing by $10$. Finally, as we are setting the prior on the normalized log-scale (because that's what the GP is seeing) we take a log-transform."
"The motivation is that we have around $7.3$K data points and we want to consider the in between data points distance in the normalized scale. That is why we consider the ratio `7_000 / time_str`. Note that we want to capture the long term trend, so we want to consider a length scale that is larger than the data points distance. We increase the order of magnitude by dividing by $10$. Finally, since a {class}`~pymc.distributions.continuous.LogNormal` distribution has positive support and a common choice for length scales, we take a log-transform on the resulting quantity `700 / time_str` so ensure the mean of the prior is close to this value."
]
},
{
Expand Down
2 changes: 1 addition & 1 deletion examples/gaussian_processes/GP-Births.myst.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -347,7 +347,7 @@ ax.set_title(
);
```

The motivation is that we have around $7.3$K data points and we want to consider the in between data points distance in the normalized (log) scale. That is why we consider the ratio `7_000 / time_str`. Note that we want to capture the long term trend, so we want to consider a length scale that is larger than the data points distance. We increase the order of magnitude by dividing by $10$. Finally, as we are setting the prior on the normalized log-scale (because that's what the GP is seeing) we take a log-transform.
The motivation is that we have around $7.3$K data points and we want to consider the in between data points distance in the normalized scale. That is why we consider the ratio `7_000 / time_str`. Note that we want to capture the long term trend, so we want to consider a length scale that is larger than the data points distance. We increase the order of magnitude by dividing by $10$. Finally, since a {class}`~pymc.distributions.continuous.LogNormal` distribution has positive support and a common choice for length scales, we take a log-transform on the resulting quantity `700 / time_str` so ensure the mean of the prior is close to this value.

+++

Expand Down
Loading