fixing docs

docs/index.html

@@ -4764,6 +4764,12 @@
   <a href="./week2.html" class="sidebar-item-text sidebar-link">
  <span class="menu-text">MCMC and Stan</span></a>
   </div>
+</li>
+          <li class="sidebar-item">
+  <div class="sidebar-item-container"> 
+  <a href="./week3.html" class="sidebar-item-text sidebar-link">
+ <span class="menu-text">What to do after MCMC</span></a>
+  </div>
 </li>
           <li class="sidebar-item">
   <div class="sidebar-item-container"> 

docs/site_libs/quarto-search/quarto-search.js

@@ -675,6 +675,18 @@ function showCopyLink(query, options) {
 // create the index
 var fuseIndex = undefined;
 var shownWarning = false;
+
+// fuse index options
+const kFuseIndexOptions = {
+  keys: [
+    { name: "title", weight: 20 },
+    { name: "section", weight: 20 },
+    { name: "text", weight: 10 },
+  ],
+  ignoreLocation: true,
+  threshold: 0.1,
+};
+
 async function readSearchData() {
   // Initialize the search index on demand
   if (fuseIndex === undefined) {
@@ -685,17 +697,7 @@ async function readSearchData() {
       shownWarning = true;
       return;
     }
-    // create fuse index
-    const options = {
-      keys: [
-        { name: "title", weight: 20 },
-        { name: "section", weight: 20 },
-        { name: "text", weight: 10 },
-      ],
-      ignoreLocation: true,
-      threshold: 0.1,
-    };
-    const fuse = new window.Fuse([], options);
+    const fuse = new window.Fuse([], kFuseIndexOptions);
 
     // fetch the main search.json
     const response = await fetch(offsetURL("search.json"));
@@ -1226,8 +1228,34 @@ function algoliaSearch(query, limit, algoliaOptions) {
   });
 }
 
-function fuseSearch(query, fuse, fuseOptions) {
-  return fuse.search(query, fuseOptions).map((result) => {
+let subSearchTerm = undefined;
+let subSearchFuse = undefined;
+const kFuseMaxWait = 125;
+
+async function fuseSearch(query, fuse, fuseOptions) {
+  let index = fuse;
+  // Fuse.js using the Bitap algorithm for text matching which runs in
+  // O(nm) time (no matter the structure of the text). In our case this
+  // means that long search terms mixed with large index gets very slow
+  //
+  // This injects a subIndex that will be used once the terms get long enough
+  // Usually making this subindex is cheap since there will typically be
+  // a subset of results matching the existing query
+  if (subSearchFuse !== undefined && query.startsWith(subSearchTerm)) {
+    // Use the existing subSearchFuse
+    index = subSearchFuse;
+  } else if (subSearchFuse !== undefined) {
+    // The term changed, discard the existing fuse
+    subSearchFuse = undefined;
+    subSearchTerm = undefined;
+  }
+
+  // Search using the active fuse
+  const then = performance.now();
+  const resultsRaw = await index.search(query, fuseOptions);
+  const now = performance.now();
+
+  const results = resultsRaw.map((result) => {
     const addParam = (url, name, value) => {
       const anchorParts = url.split("#");
       const baseUrl = anchorParts[0];
@@ -1244,4 +1272,15 @@ function fuseSearch(query, fuse, fuseOptions) {
       crumbs: result.item.crumbs,
     };
   });
+
+  // If we don't have a subfuse and the query is long enough, go ahead
+  // and create a subfuse to use for subsequent queries
+  if (now - then > kFuseMaxWait && subSearchFuse === undefined) {
+    subSearchTerm = query;
+    subSearchFuse = new window.Fuse([], kFuseIndexOptions);
+    resultsRaw.forEach((rr) => {
+      subSearchFuse.add(rr.item);
+    });
+  }
+  return results;
 }

materials/data.csv

@@ -0,0 +1,100 @@
+5.133879959574598
+4.937817051434988
+5.277966400350326
+4.978073616723295
+4.768264765712152
+5.00325432375216
+5.140401818212519
+5.220085220743337
+4.9103261367115305
+4.916817929108131
+4.676230504106508
+4.681155942296321
+4.939456345611404
+5.404674908950762
+5.0000565051183505
+4.946723670230919
+5.071484229192757
+4.702765071407966
+4.990440219285413
+4.757565482906698
+4.90342085169882
+4.909731907856895
+4.7502350067852035
+4.735636087526655
+4.831835910148419
+5.083641260592571
+4.941212006801077
+4.836347213445427
+4.962098544360349
+4.686040420643092
+5.124470580419098
+5.066542799622644
+5.116490362362727
+4.978068545964261
+4.733571541168396
+4.799377562666428
+4.731238173387429
+4.927630515268987
+5.276180964383851
+4.605853804427195
+4.895306070517735
+5.125441726077501
+5.048540015619509
+5.134475090531417
+5.020340763177987
+4.744641578259603
+4.9620314387166395
+4.664576116370324
+5.244742999934165
+5.279274831080141
+5.089479340563577
+5.071743145634021
+4.942265997797657
+5.089398760768866
+5.458231752210997
+4.786148013337675
+4.7972136274342025
+4.93152718719522
+4.978778063916119
+4.801232311921279
+4.686526370326429
+5.105113175288019
+4.8016552715132645
+4.978252685586892
+4.903595625784314
+5.150389785891813
+4.957095292110073
+4.851463539442262
+5.18128245975816
+5.139918945662961
+5.020216551825798
+4.84115302966151
+4.975904991006625
+5.401969287491859
+4.811230682660985
+4.7993454128378925
+5.08505429688905
+4.6388175324133565
+5.363488556427758
+4.825053384694479
+5.304806718052644
+5.321881179050992
+5.044592651869508
+4.944522400406804
+4.9930450107605076
+5.400020830329514
+5.162848379281823
+4.851419760797657
+5.021789943039787
+4.831523320170004
+4.870328937789049
+4.927266223569667
+5.023563865935118
+4.969113692374906
+4.997160262493862
+4.747684042472069
+5.070321872431043
+4.899256187007241
+5.111542353099519
+5.201047790285543

materials/week3.qmd

@@ -120,7 +120,7 @@ idata
 ```
 
 The function `az.summary` lets us look at some useful summary statistics,
-including $\hat{R}$ and divergent transitions.
+including $\hat{R}$, divergent transitions and MCSE.
 
 The variable `lp`, which you can find in the group `sample_stats` is the
 model's total log probability density. It's not very meaningful on its own,
@@ -131,23 +131,30 @@ divergent transitions.
 az.summary(idata.sample_stats, var_names=["lp", "diverging"])
 ```
 
-Sometimes it's useful to check the convergence of individual parameters. This
-can be done by pointing `az.summary` at the group where the parameters of
-interest live. In this case the group is called `posterior`.
+Sometimes it's useful to summarise individual parameters. This can be done by
+pointing `az.summary` at the group where the parameters of interest live. In
+this case the group is called `posterior`.
 
 ```{python}
 az.summary(idata.posterior, var_names=["sigma", "g"])
 ```
 
-The function `az.loo` performs approximate leave-one-out cross validation, which can be useful for evaluating how well the model might make predictions.
+The function `az.loo` performs approximate leave-one-out cross validation, which
+can be useful for evaluating how well the model might make predictions. Watch
+out for the `warning` column, which can tell you if the approximation is likely
+to be incorrect.
 
 ```{python}
 az.loo(idata, var_name="y", pointwise=True)
-
 ```
 
+The function `az.compare` is useful for comparing different out of sample log
+likelihood estimates.
+
 ```{python}
+
 idata.log_likelihood["fake"] = xr.DataArray(
+    # generate some fake log likelihoods
     np.random.normal(0, 2, [4, 500, 919]),
     coords=idata.log_likelihood.coords,
     dims=idata.log_likelihood.dims
@@ -160,3 +167,5 @@ az.compare(
 )
 
 ```
+
+az.plot_ppc(j)