Skip to content

Notebook: sampler_stats; update links, clean up sections #426

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.

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 3 commits into from
Nov 5, 2022
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
1 change: 0 additions & 1 deletion .pre-commit-config.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -68,7 +68,6 @@ repos:
examples/case_studies/blackbox_external_likelihood.ipynb|
examples/case_studies/blackbox_external_likelihood_numpy.ipynb|
examples/case_studies/item_response_nba.ipynb|
examples/diagnostics_and_criticism/sampler-stats.ipynb|
examples/gaussian_processes/GP-MaunaLoa2.ipynb|
examples/generalized_linear_models/GLM-logistic.ipynb|
examples/generalized_linear_models/GLM-out-of-sample-predictions.ipynb|
Expand Down
39 changes: 22 additions & 17 deletions examples/diagnostics_and_criticism/sampler-stats.ipynb
Original file line number Diff line number Diff line change
Expand Up @@ -7,13 +7,8 @@
"(sampler_stats)=\n",
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is there a "keyword" way to reference/link to this bit of documentation?


Reply via ReviewNB

"# Sampler Statistics\n",
"\n",
"When checking for convergence or when debugging a badly behaving\n",
"sampler, it is often helpful to take a closer look at what the\n",
"sampler is doing. For this purpose some samplers export\n",
"statistics for each generated sample.\n",
"\n",
":::{post} May 31, 2022\n",
":tags: diagnostics \n",
":tags: diagnostics\n",
":category: beginner\n",
":author: Meenal Jhajharia, Christian Luhmann\n",
":::"
Expand Down Expand Up @@ -65,6 +60,8 @@
"cell_type": "markdown",
"metadata": {},
"source": [
"When checking for convergence or when debugging a badly behaving sampler, it is often helpful to take a closer look at what the sampler is doing. For this purpose some samplers export statistics for each generated sample.\n",
"\n",
"As a minimal example we sample from a standard normal distribution:"
]
},
Expand Down Expand Up @@ -151,7 +148,7 @@
"cell_type": "markdown",
"metadata": {},
"source": [
"- `Note`: NUTS provides the following statistics( these are internal statistics that the sampler uses, you don't need to do anything with them when using PyMC3, to learn more about them, [check this page](https://docs.pymc.io/api/inference.html#module-pymc3.step_methods.hmc.nuts)."
"- `Note`: NUTS provides the following statistics (these are internal statistics that the sampler uses, you don't need to do anything with them when using PyMC, to learn more about them, {class}`pymc.NUTS`."
]
},
{
Expand Down Expand Up @@ -695,7 +692,7 @@
"metadata": {},
"source": [
"Some points to `Note`:\n",
"- Some of the sample statistics used by NUTS are renamed when converting to `InferenceData` to follow [ArviZ's naming convention](https://arviz-devs.github.io/arviz/schema/schema.html#sample-stats), while some are specific to PyMC3 and keep their internal PyMC3 name in the resulting InferenceData object.\n",
"- Some of the sample statistics used by NUTS are renamed when converting to `InferenceData` to follow {ref}`ArviZ's naming convention <arviz:schema>`, while some are specific to PyMC3 and keep their internal PyMC3 name in the resulting InferenceData object.\n",
"- `InferenceData` also stores additional info like the date, versions used, sampling time and tuning steps as attributes."
]
},
Expand Down Expand Up @@ -1746,6 +1743,22 @@
");"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Authors\n",
"* Updated by Meenal Jhajharia in April 2021 ([pymc-examples#95](https://github.com/pymc-devs/pymc-examples/pull/95))\n",
"* Updated to v4 by Christian Luhmann in May 2022 ([pymc-examples#338](https://github.com/pymc-devs/pymc-examples/pull/338))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Watermark"
]
},
{
"cell_type": "code",
"execution_count": 16,
Expand Down Expand Up @@ -1777,14 +1790,6 @@
"%watermark -n -u -v -iv -w"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"* Updated by Meenal Jhajharia\n",
"* Updated by Christian Luhmann"
]
},
{
"cell_type": "markdown",
"metadata": {},
Expand Down Expand Up @@ -1813,7 +1818,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.4"
"version": "3.9.10"
}
},
"nbformat": 4,
Expand Down
26 changes: 13 additions & 13 deletions myst_nbs/diagnostics_and_criticism/sampler-stats.myst.md
Original file line number Diff line number Diff line change
Expand Up @@ -15,13 +15,8 @@ kernelspec:
(sampler_stats)=
# Sampler Statistics

When checking for convergence or when debugging a badly behaving
sampler, it is often helpful to take a closer look at what the
sampler is doing. For this purpose some samplers export
statistics for each generated sample.

:::{post} May 31, 2022
:tags: diagnostics
:tags: diagnostics
:category: beginner
:author: Meenal Jhajharia, Christian Luhmann
:::
Expand All @@ -43,6 +38,8 @@ az.style.use("arviz-darkgrid")
plt.rcParams["figure.constrained_layout.use"] = False
```

When checking for convergence or when debugging a badly behaving sampler, it is often helpful to take a closer look at what the sampler is doing. For this purpose some samplers export statistics for each generated sample.

As a minimal example we sample from a standard normal distribution:

```{code-cell} ipython3
Expand All @@ -57,7 +54,7 @@ with model:
idata = pm.sample(2000, tune=1000, init=None, step=step, chains=4)
```

- `Note`: NUTS provides the following statistics( these are internal statistics that the sampler uses, you don't need to do anything with them when using PyMC3, to learn more about them, [check this page](https://docs.pymc.io/api/inference.html#module-pymc3.step_methods.hmc.nuts).
- `Note`: NUTS provides the following statistics (these are internal statistics that the sampler uses, you don't need to do anything with them when using PyMC, to learn more about them, {class}`pymc.NUTS`.

```{code-cell} ipython3
idata.sample_stats
Expand Down Expand Up @@ -94,7 +91,7 @@ The sample statistics variables are defined as follows:
+++

Some points to `Note`:
- Some of the sample statistics used by NUTS are renamed when converting to `InferenceData` to follow [ArviZ's naming convention](https://arviz-devs.github.io/arviz/schema/schema.html#sample-stats), while some are specific to PyMC3 and keep their internal PyMC3 name in the resulting InferenceData object.
- Some of the sample statistics used by NUTS are renamed when converting to `InferenceData` to follow {ref}`ArviZ's naming convention <arviz:schema>`, while some are specific to PyMC3 and keep their internal PyMC3 name in the resulting InferenceData object.
- `InferenceData` also stores additional info like the date, versions used, sampling time and tuning steps as attributes.

```{code-cell} ipython3
Expand Down Expand Up @@ -189,15 +186,18 @@ az.plot_density(
);
```

## Authors
* Updated by Meenal Jhajharia in April 2021 ([pymc-examples#95](https://github.com/pymc-devs/pymc-examples/pull/95))
* Updated to v4 by Christian Luhmann in May 2022 ([pymc-examples#338](https://github.com/pymc-devs/pymc-examples/pull/338))

+++

## Watermark

```{code-cell} ipython3
%load_ext watermark
%watermark -n -u -v -iv -w
```

* Updated by Meenal Jhajharia
* Updated by Christian Luhmann

+++

:::{include} ../page_footer.md
:::