*[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](introduction/apis/client-sdks/collections.md)

*[Collections](../../pgml-docs/docs/guides/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)

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 deployedbefore to the current oneisdeployed.|

| 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 waslastdeployedfor this projectisimmediately redeployed, overriding the currently deployed model.|

The default deployment behavior allows any algorithm to qualify. It's automatically used during training, but can be manually executed as well:

SELECT*FROMpgml.deploy(

'Handwritten Digit Image Classifier',

<preclass="language-sql"><codeclass="lang-sql"><strong>SELECT * FROM pgml.deploy(

</strong> 'Handwritten Digit Image Classifier',

strategy => 'best_score'

);

Handwritten Digit Image Classifier |rollback | xgboost

(1 row)

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.

SELECT*FROMpgml.deploy(12);

project | strategy | algorithm

Handwritten Digit Image Classifier | specific | xgboost

(1 row)

There are 3 steps to preprocessing data:

*[Encoding](../../../../../../pgml-dashboard/content/docs/training/preprocessing.md#categorical-encodings) categorical values into quantitative values

*[Imputing](../../../../../../pgml-dashboard/content/docs/training/preprocessing.md#imputing-missing-values) NULL values to some quantitative value

*[Scaling](../../../../../../pgml-dashboard/content/docs/training/preprocessing.md#scaling-values) quantitative values across all variables to similar ranges

*[Encoding](data-pre-processing.md#ordinal-encoding) categorical values into quantitative values

*[Imputing](data-pre-processing.md#imputing-missing-values) NULL values to some quantitative value

*[Scaling](data-pre-processing.md#scaling-values) quantitative values across all variables to similar ranges

These preprocessing steps may be specified on a per-column basis to the[train()](../../../../../../docs/training/overview/) function. By default, PostgresML does minimal preprocessing on training data, and will raise an error during analysis if NULL values are encountered without a preprocessor. All types other than`TEXT` are treated as quantitative variables and cast to floating point representations before passing them to the underlying algorithm implementations.

These preprocessing steps may be specified on a per-column basis to the[train()](./) function. By default, PostgresML does minimal preprocessing on training data, and will raise an error during analysis if NULL values are encountered without a preprocessor. All types other than`TEXT` are treated as quantitative variables and cast to floating point representations before passing them to the underlying algorithm implementations.

SELECTpgml.train(

PostgresML is written in Rust, so you'll need to install the latest compiler from[rust-lang.org](https://rust-lang.org). Additionally, we use the Rust PostgreSQL extension framework`pgrx`, which requires some initialization steps:

cargo install cargo-pgrx --version 0.11.2&& \

cargo install cargo-pgrx --version 0.9.8&& \

cargo pgrx init

virtualenv pgml-venv&& \

source pgml-venv/bin/activate&& \

pip install -r requirements.txt

pip install -r requirements.txt&& \

pip install -r requirements-xformers.txt --no-dependencies

{% endtab %}

We like and use pgvector a lot, as documented in our blog posts and examples, to store and search embeddings. You can install pgvector from source pretty easily:

git clone --branch v0.5.0 https://github.com/pgvector/pgvector&& \

git clone --branch v0.4.4 https://github.com/pgvector/pgvector&& \

cd pgvector&& \

echo"trusted = true">> vector.control&& \

make&& \

cd pgml-extension&& \

cargo install cargo-pgrx --version 0.11.2&& \

cargo install cargo-pgrx --version 0.9.8&& \

cargo pgrx init