From b4d413e3d0aa548095ed450ac0a506af12d25a85 Mon Sep 17 00:00:00 2001
From: mmcky <mamckay@gmail.com>
Date: Fri, 3 May 2024 11:34:27 +1000
Subject: [PATCH 1/3] FIX: fixes for PDF builder

---
 lectures/bayes_nonconj.md | 16 +++++-----------
 1 file changed, 5 insertions(+), 11 deletions(-)

diff --git a/lectures/bayes_nonconj.md b/lectures/bayes_nonconj.md
index 0443499..e32b816 100644
--- a/lectures/bayes_nonconj.md
+++ b/lectures/bayes_nonconj.md
@@ -315,16 +315,16 @@ Formula {eq}`eq:ELBO` is called  the evidence lower bound (ELBO).
 
 A standard optimization routine can used to search for the optimal $\phi$ in our parametrized distribution $q_{\phi}(\theta)$.
 
-The parameterized  distribution $q_{\phi}(\theta)$ is called the **variational distribution**.
+The parameterized distribution $q_{\phi}(\theta)$ is called the **variational distribution**.
 
 We can implement Stochastic Variational Inference (SVI) Numpyro using the `Adam` gradient descent algorithm to approximate posterior.
 
 We use  two sets of variational distributions: Beta and TruncatedNormal with support $[0,1]$
 
-  - Learnable parameters for the Beta distribution are (alpha, beta), both of which are positive.
-  - Learnable parameters for the Truncated Normal distribution are (loc, scale).
+- Learnable parameters for the Beta distribution are (alpha, beta), both of which are positive.
+- Learnable parameters for the Truncated Normal distribution are (loc, scale).
 
-<u> We restrict the truncated Normal paramter 'loc' to be in the interval $[0,1]$</u>.
+We restrict the truncated Normal parameter 'loc' to be in the interval $[0,1]$
 
 
 ## Implementation
@@ -335,30 +335,24 @@ We have constructed a Python class `BaysianInference` that requires the followin
 - `name_dist`: a string that specifies distribution names
 
 The (`param`, `name_dist`) pair includes:
-- ('beta', alpha, beta)
 
+- ('beta', alpha, beta)
 - ('uniform', upper_bound, lower_bound)
-
 - ('lognormal', loc, scale)
    - Note: This is the truncated log normal.
-
 - ('vonMises', kappa), where kappa denotes concentration parameter, and center location is set to $0.5$.
    - Note: When using `Numpyro`, this is the **shifted** distribution.
-
 - ('laplace', loc, scale)
    - Note: This is the truncated Laplace
 
 The class `BaysianInference` has several key methods :
 - `sample_prior`:
    - This can be used to draw a single sample from the given prior distribution.
-
 - `show_prior`:
    - Plots the approximate prior distribution by repeatedly drawing samples and fitting a kernal density curve.
-
 - `MCMC_sampling`:
    - INPUT: (data, num_samples, num_warmup=1000)
    - Take a `np.array` data and generate MCMC sampling of posterior of size `num_samples`.
-
 - `SVI_run`:
   - INPUT: (data, guide_dist, n_steps=10000)
   - guide_dist = 'normal' - use a **truncated** normal distribution as the parametrized guide

From e513849701569bbf2ec0870ae7d7c12836d199c6 Mon Sep 17 00:00:00 2001
From: mmcky <mamckay@gmail.com>
Date: Fri, 3 May 2024 11:49:33 +1000
Subject: [PATCH 2/3] enable -n for pdflatex builds

---
 .github/workflows/ci.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
index 2e3b487..bd84137 100644
--- a/.github/workflows/ci.yml
+++ b/.github/workflows/ci.yml
@@ -63,7 +63,7 @@ jobs:
       - name: Build PDF from LaTeX
         shell: bash -l {0}
         run: |
-          jb build lectures --builder pdflatex --path-output ./ -W --keep-going
+          jb build lectures --builder pdflatex --path-output ./ -n -W --keep-going
           mkdir -p _build/html/_pdf
           cp -u _build/latex/*.pdf _build/html/_pdf
       - name: Upload Execution Reports (LaTeX)

From cb8dbfb117af92b0802582501b3f7ef21ed73206 Mon Sep 17 00:00:00 2001
From: mmcky <mamckay@gmail.com>
Date: Fri, 3 May 2024 12:11:44 +1000
Subject: [PATCH 3/3] upload build folder for pdflatex builder

---
 .github/workflows/ci.yml | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
index bd84137..719dbba 100644
--- a/.github/workflows/ci.yml
+++ b/.github/workflows/ci.yml
@@ -66,6 +66,11 @@ jobs:
           jb build lectures --builder pdflatex --path-output ./ -n -W --keep-going
           mkdir -p _build/html/_pdf
           cp -u _build/latex/*.pdf _build/html/_pdf
+      - name: Upload build folder
+        uses: actions/upload-artifact@v4
+        with:
+          name: build-folder
+          path: _build
       - name: Upload Execution Reports (LaTeX)
         uses: actions/upload-artifact@v4
         if: failure()