deploy specific models (#1265)

montanalow · web-flow · commit a8eb55b4af6b · 2024-01-03T10:42:13.000-08:00
diff --git a/.github/workflows/ubuntu-packages-and-docker-image.yml b/.github/workflows/ubuntu-packages-and-docker-image.yml
@@ -4,7 +4,7 @@ on:
   workflow_dispatch:
     inputs:
       packageVersion:
-        default: "2.8.1"
+        default: "2.8.2"
 jobs:
   #
   # PostgresML extension.
diff --git a/.github/workflows/ubuntu-postgresml-python-package.yaml b/.github/workflows/ubuntu-postgresml-python-package.yaml
@@ -4,7 +4,7 @@ on:
   workflow_dispatch:
     inputs:
       packageVersion:
-        default: "2.8.1"
+        default: "2.8.2"
 
 jobs:
   postgresml-python:
diff --git a/pgml-cms/docs/SUMMARY.md b/pgml-cms/docs/SUMMARY.md
@@ -36,7 +36,7 @@
     * [pgml.tune()](introduction/apis/sql-extensions/pgml.tune.md)
   * [Client SDKs](introduction/apis/client-sdks/README.md)
     * [Overview](introduction/apis/client-sdks/getting-started.md)
-    * [Collections](../../pgml-docs/docs/guides/sdks/collections.md)
+    * [Collections](introduction/apis/client-sdks/collections.md)
     * [Pipelines](introduction/apis/client-sdks/pipelines.md)
     * [Search](introduction/apis/client-sdks/search.md)
     * [Tutorials](introduction/apis/client-sdks/tutorials/README.md)
diff --git a/pgml-cms/docs/introduction/apis/client-sdks/collections.md b/pgml-cms/docs/introduction/apis/client-sdks/collections.md
diff --git a/pgml-cms/docs/introduction/apis/sql-extensions/pgml.deploy.md b/pgml-cms/docs/introduction/apis/sql-extensions/pgml.deploy.md
@@ -26,11 +26,11 @@ pgml.deploy(
 
 There are 3 different deployment strategies available:
 
-| Strategy      | Description                                                                                                           |
-| ------------- | --------------------------------------------------------------------------------------------------------------------- |
-| `most_recent` | The most recently trained model for this project is immediately deployed, regardless of metrics.                      |
-| `best_score`  | The model that achieved the best key metric score is immediately deployed.                                            |
-| `rollback`    | The model that was last deployed for this project is immediately redeployed, overriding the currently deployed model. |
+| Strategy      | Description                                                                                      |
+| ------------- |--------------------------------------------------------------------------------------------------|
+| `most_recent` | The most recently trained model for this project is immediately deployed, regardless of metrics. |
+| `best_score`  | The model that achieved the best key metric score is immediately deployed.                       |
+| `rollback`    | The model that was deployed before to the current one is deployed.                               |
 
 The default deployment behavior allows any algorithm to qualify. It's automatically used during training, but can be manually executed as well:
 
@@ -40,11 +40,12 @@ The default deployment behavior allows any algorithm to qualify. It's automatica
 
 #### SQL
 
-<pre class="language-sql"><code class="lang-sql"><strong>SELECT * FROM pgml.deploy(
-</strong>    'Handwritten Digit Image Classifier',
+```sql
+SELECT * FROM pgml.deploy(
+   'Handwritten Digit Image Classifier',
     strategy => 'best_score'
 );
-</code></pre>
+```
 
 #### Output
 
@@ -121,3 +122,22 @@ SELECT * FROM pgml.deploy(
  Handwritten Digit Image Classifier | rollback | xgboost
 (1 row)
 ```
+
+### Specific Model IDs
+
+In the case you need to deploy an exact model that is not the `most_recent` or `best_score`, you may deploy a model by id. Model id's can be found in the `pgml.models` table. 
+
+#### SQL
+
+```sql
+SELECT * FROM pgml.deploy(12);
+```
+
+#### Output
+
+```sql
+              project               | strategy | algorithm
+------------------------------------+----------+-----------
+ Handwritten Digit Image Classifier | specific | xgboost
+(1 row)
+```
diff --git a/pgml-extension/Cargo.lock b/pgml-extension/Cargo.lock
diff --git a/pgml-extension/Cargo.toml b/pgml-extension/Cargo.toml
@@ -1,6 +1,6 @@
 [package]
 name = "pgml"
-version = "2.8.1"
+version = "2.8.2"
 edition = "2021"
 
 [lib]
diff --git a/pgml-extension/sql/pgml--2.8.1--2.8.2.sql b/pgml-extension/sql/pgml--2.8.1--2.8.2.sql
@@ -0,0 +1,27 @@
+-- src/api.rs:317
+-- pgml::api::deploy
+DROP FUNCTION IF EXISTS pgml."deploy"(BIGINT);
+CREATE  FUNCTION pgml."deploy"(
+    "model_id" BIGINT /* i64 */
+) RETURNS TABLE (
+                    "project" TEXT,  /* alloc::string::String */
+                    "strategy" TEXT,  /* alloc::string::String */
+                    "algorithm" TEXT  /* alloc::string::String */
+                )
+    LANGUAGE c /* Rust */
+AS 'MODULE_PATHNAME', 'deploy_model_wrapper';
+
+DROP FUNCTION IF EXISTS pgml."deploy"(text, pgml.Strategy, pgml.Algorithm);
+CREATE  FUNCTION pgml."deploy"(
+    "project_name" TEXT, /* &str */
+    "strategy" pgml.Strategy, /* pgml::orm::strategy::Strategy */
+    "algorithm" pgml.Algorithm DEFAULT NULL /* core::option::Option<pgml::orm::algorithm::Algorithm> */
+) RETURNS TABLE (
+                    "project" TEXT,  /* alloc::string::String */
+                    "strategy" TEXT,  /* alloc::string::String */
+                    "algorithm" TEXT  /* alloc::string::String */
+                )
+    LANGUAGE c /* Rust */
+AS 'MODULE_PATHNAME', 'deploy_strategy_wrapper';
+
+ALTER TYPE pgml.strategy ADD VALUE 'specific';
diff --git a/pgml-extension/src/api.rs b/pgml-extension/src/api.rs
@@ -287,7 +287,7 @@ fn train_joint(
     };
 
     if deploy {
-        project.deploy(model.id);
+        project.deploy(model.id, Strategy::new_score);
     } else {
         warning!("Not deploying newly trained model.");
     }
@@ -300,8 +300,40 @@ fn train_joint(
     )])
 }
 
-#[pg_extern]
-fn deploy(
+#[pg_extern(name = "deploy")]
+fn deploy_model(
+    model_id: i64
+) -> TableIterator<
+    'static,
+    (
+        name!(project, String),
+        name!(strategy, String),
+        name!(algorithm, String),
+    ),
+> {
+    let model = unwrap_or_error!(Model::find_cached(model_id));
+
+    let project_id = Spi::get_one_with_args::<i64>(
+        "SELECT projects.id from pgml.projects JOIN pgml.models ON models.project_id = projects.id WHERE models.id = $1",
+        vec![(PgBuiltInOids::INT8OID.oid(), model_id.into_datum())],
+    )
+        .unwrap();
+
+    let project_id =
+        project_id.unwrap_or_else(|| error!("Project does not exist."));
+
+    let project = Project::find(project_id).unwrap();
+    project.deploy(model_id, Strategy::specific);
+
+    TableIterator::new(vec![(
+        project.name,
+        Strategy::specific.to_string(),
+        model.algorithm.to_string(),
+    )])
+}
+
+#[pg_extern(name = "deploy")]
+fn deploy_strategy(
     project_name: &str,
     strategy: Strategy,
     algorithm: default!(Option<Algorithm>, "NULL"),
@@ -378,7 +410,7 @@ fn deploy(
     let algorithm = algorithm.expect("No qualified models exist for this deployment.");
 
     let project = Project::find(project_id).unwrap();
-    project.deploy(model_id);
+    project.deploy(model_id, strategy);
 
     TableIterator::new(vec![(
         project_name.to_string(),
@@ -922,7 +954,7 @@ fn tune(
     };
 
     if deploy {
-        project.deploy(model.id);
+        project.deploy(model.id, Strategy::new_score);
     }
 
     TableIterator::new(vec![(
diff --git a/pgml-extension/src/orm/project.rs b/pgml-extension/src/orm/project.rs
@@ -89,14 +89,14 @@ impl Project {
             .unwrap()
     }
 
-    pub fn deploy(&self, model_id: i64) {
+    pub fn deploy(&self, model_id: i64, strategy: Strategy) {
         info!("Deploying model id: {:?}", model_id);
         Spi::get_one_with_args::<i64>(
             "INSERT INTO pgml.deployments (project_id, model_id, strategy) VALUES ($1, $2, $3::pgml.strategy) RETURNING id",
             vec![
                 (PgBuiltInOids::INT8OID.oid(), self.id.into_datum()),
                 (PgBuiltInOids::INT8OID.oid(), model_id.into_datum()),
-                (PgBuiltInOids::TEXTOID.oid(), Strategy::most_recent.to_string().into_datum()),
+                (PgBuiltInOids::TEXTOID.oid(), strategy.to_string().into_datum()),
             ],
         ).unwrap();
         let mut projects = PROJECT_ID_TO_DEPLOYED_MODEL_ID.exclusive();
diff --git a/pgml-extension/src/orm/strategy.rs b/pgml-extension/src/orm/strategy.rs
@@ -8,6 +8,7 @@ pub enum Strategy {
     best_score,
     most_recent,
     rollback,
+    specific,
 }
 
 impl std::str::FromStr for Strategy {
@@ -19,6 +20,7 @@ impl std::str::FromStr for Strategy {
             "best_score" => Ok(Strategy::best_score),
             "most_recent" => Ok(Strategy::most_recent),
             "rollback" => Ok(Strategy::rollback),
+            "specific" => Ok(Strategy::rollback),
             _ => Err(()),
         }
     }
@@ -31,6 +33,7 @@ impl std::string::ToString for Strategy {
             Strategy::best_score => "best_score".to_string(),
             Strategy::most_recent => "most_recent".to_string(),
             Strategy::rollback => "rollback".to_string(),
+            Strategy::specific => "specific".to_string(),
         }
     }
 }

Original file line number	Diff line number	Diff line change
`@@ -4,7 +4,7 @@ on:`
`4`	`4`	`workflow_dispatch:`
`5`	`5`	`inputs:`
`6`	`6`	`packageVersion:`
`7`		`- default: "2.8.1"`
	`7`	`+ default: "2.8.2"`
`8`	`8`	`jobs:`
`9`	`9`	`#`
`10`	`10`	`# PostgresML extension.`