Move content around, sketch out where we want certain topics

slides/01-introduction.qmd

@@ -383,14 +383,6 @@ tr = tree.DecisionTreeRegressor().fit(le.transform(X_train), y_train)
 abalone_fit = pipeline.Pipeline([('label_encoder', le), ('decision_tree', tr)])
 ```
 
-## Bind preprocessors and models together
-
-![](images/good_workflow.png){fig-align="center"}
-
-
-## What is wrong with this? {.annotation}
-
-![](images/bad_workflow.png){fig-align="center"}
 
 
 ## Your turn 🏺 {transition="slide-in"}
@@ -496,28 +488,15 @@ board |> vetiver_pin_write(v)
 
 :::
 
-# Make it easy to do the right thing
-
--   Robust and human-friendly checking of new data
--   Track and document software dependencies of models
--   [Model cards](https://vetiver.rstudio.com/learn-more/model-card.html) for transparent, responsible reporting
-
-# Make it easy to do the right thing
-
-![](images/vetiver-templates.png){fig-align="center"}
-
-
 ## Your turn 🏺 {transition="slide-in"}
 
 *Pin your vetiver model object to a temporary board.*
 
-*Open the Model Card template, and spend a few minutes exploring how you might create a Model Card for this abalone rings model.*
-
-*Discuss something you notice about the Model Card with your neighbor.*
+*Retrieve the model metadata with `pin_meta()`.*
 
 ```{r}
 #| echo: false
-countdown(minutes = 7)
+countdown(minutes = 5)
 ```
 
 # Posit Connect

slides/02-deploy.qmd

@@ -113,6 +113,29 @@ board = board_connect(server_url=rsc_url, api_key=api_key, allow_pickle_read = T
 vetiver_pin_write(board, v)
 ```
 
+
+# Make it easy to do the right thing
+
+-   Robust and human-friendly checking of new data
+-   Track and document software dependencies of models
+-   [Model cards](https://vetiver.rstudio.com/learn-more/model-card.html) for transparent, responsible reporting
+
+# Make it easy to do the right thing
+
+![](images/vetiver-templates.png){fig-align="center"}
+
+## Your turn 🏺 {transition="slide-in"}
+
+*Open the Model Card template, and spend a few minutes exploring how you might create a Model Card for this abalone rings model.*
+
+*Discuss something you notice about the Model Card with your neighbor.*
+
+```{r}
+#| echo: false
+countdown(minutes = 7)
+```
+
+
 ## You can deploy your model as a...
 
 . . .
@@ -278,10 +301,12 @@ predict(endpoint = endpoint, data = ring_test)
 
 *Predict with your endpoint for new data.*
 
+*Optional: call another endpoint from Python or R.*
+
 
 ```{r}
 #| echo: false
-countdown(minutes = 5)
+countdown(minutes = 10)
 ```
 
 ## Your turn 🏺 {transition="slide-in"}
@@ -290,7 +315,7 @@ countdown(minutes = 5)
 
 *Predict with your endpoint for new data.*
 
-*You get extra credit if your neighbor's model is in a different language than yours.*
+*You get extra credit if your neighbor's model is in a different language than yours!*
 
 
 ```{r}
@@ -342,3 +367,6 @@ h = { 'Authorization': f'Key {api_key}' }
 predict(endpoint = endpoint, data = ring_test, headers = h)
 ```
 
+# Model input prototype
+
+tktk

slides/03-deploy.qmd

@@ -0,0 +1,65 @@
+---
+title: "3 - Deeper into deployment"
+subtitle: "Deploy and maintain models with vetiver"
+format:
+  revealjs: 
+    slide-number: true
+    footer: <https://posit-conf-2023.github.io/vetiver>
+    preview-links: auto
+    incremental: true
+    theme: [default]
+    width: 1280
+    height: 720
+knitr:
+  opts_chunk: 
+    echo: true
+    collapse: true
+    comment: "#>"
+---
+
+```{r}
+#| include: false
+#| file: setup.R
+```
+
+## Plan for this workshop
+
+::: nonincremental
+-   *Versioning*
+    -   Managing change in models ✅
+-   *Deploying*
+    -   Putting models in REST APIs 🎯
+-   *Monitoring*
+    -   Tracking model performance 👀
+:::
+
+# Where does vetiver work?
+
+-   Posit's pro products, like [Connect](https://posit.co/products/enterprise/connect/)
+
+-   AWS SageMaker (R only, for now)
+
+-   A public or private cloud, using Docker
+
+# Docker
+
+tktk
+
+# Deploy preprocessors and models together
+
+tktk more substantive feature engineering (like with learned transformation from training data)
+
+## Deploy preprocessors and models together
+
+![](images/good_workflow.png){fig-align="center"}
+
+
+## What is wrong with this?
+
+![](images/bad_workflow.png){fig-align="center"}
+
+# Model metrics as metadata
+
+tktk like https://vetiver.rstudio.com/learn-more/metrics-metadata.html
+
+Transition to model monitoring at end

slides/04-monitor.qmd

@@ -101,6 +101,10 @@ tktk
 
 tktk
 
+## Connect model metrics to business outcomes
+
+tktk
+
 ## Possible model monitoring artifacts
 
 - Adhoc analysis that you post in Slack