PostgresML adds extensions to the PostgreSQL database, as well as providing separate Client SDKs in JavaScript and Python that leverage the database to implement common ML & AI use cases. 

PostgresML adds extensions to the PostgreSQL database, as well as providing separate Client SDKs in JavaScript and Python that leverage the database to implement common ML & AI use cases.

The extensions provide all of the ML & AI functionality via SQL APIs, like training and inference. They are designed to be used directly for all ML practitioners who implement dozens of different use cases on their own machine learning models. 

The extensions provide all of the ML & AI functionality via SQL APIs, like training and inference. They are designed to be used directly for all ML practitioners who implement dozens of different use cases on their own machine learning models.

We also provide Client SDKs that implement the best practices on top of the SQL APIs, to ease adoption and implement common application use cases in applications, like chatbots or search engines.

Postgres is designed to be_**extensible**_. This has created a rich open-source ecosystem of additional functionality built around the core project. Some[extensions](https://www.postgresql.org/docs/current/contrib.html) are include in the base Postgres distribution, but others are also available via the[PostgreSQL Extension Network](https://pgxn.org/).\

Postgres is designed to be_**extensible**_. This has created a rich open-source ecosystem of additional functionality built around the core project. Some[extensions](https://www.postgresql.org/docs/current/contrib.html) are include in the base Postgres distribution, but others are also available via the[PostgreSQL Extension Network](https://pgxn.org/).\

There are 2 foundational extensions included in a PostgresML deployment that provide functionality inside the database through SQL APIs.

Learn more about developing with the[client-sdks](client-sdks/"mention")

Machine learning always depends on input data, whether it's generating text with pretrained LLMs, training a retention model on customer data, or predicting session abandonment in real time. Just like any PostgreSQL database, PostgresML can be configured as the authoritative application data store, a streaming replica from some other primary, or use foreign data wrappers to query another data host on demand. Depending on how frequently your data changes and where your authoritative data resides, different methodologies imply different tradeoffs.

PostgresML can easily ingest data from your existing data stores. 

PostgresML can easily ingest data from your existing data stores.

[foreign-data-wrapper.md](foreign-data-wrapper.md)

{% endcontent-ref %}

psql -c"\copy your_table TO '~/Desktop/your_table.csv' CSV HEADER"

If you're using anotherdata store, it should almost always provide a CSV export functionality, since CSV is the most commonly used data format in machine learning.

If you're using another data store, it should almost always provide a CSV export functionality, since CSV is the most commonly used data format in machine learning.

Creating a table in Postgres with the correct schema is as easy as:

CREATETABLEyour_table (

nameTEXT,

ageINTEGER,

If your data changed, repeat this process again. To avoid duplicate entries in your table, you can truncate (or delete) all rows beforehand:

TRUNCATE your_table;

To connect to your database from PostgresML, first create a corresponding`SERVER`:

CREATE SERVER live_db

FOREIGN DATA WRAPPER postgres_fdw

OPTIONS (

Once you have a`SERVER`, let's authenticate to your database:

FORCURRENT_USER

SERVER live_db

Replace`Postgres user` and`Postgres password` with details collected in the previous step. If everything went well, we'll be able to validate that everything is working with just one query:

SELECT*FROM dblink(

'live_db',

Instead of creating temporary tables for each query, you can import your entire schema into PostgresML using foreign data wrappers:

CREATESCHEMAlive_db_tables;

IMPORT FOREIGN SCHEMA public

All your tables from your`public` schema are now available in the`live_db_tables` schema. You can read and write to those tables as if they were hosted in PostgresML. For example, if you have a table called`users`, you could access it with:

SELECT*FROMlive_db_tables.usersLIMIT1;

To speed up access to your data, you can cache it in PostgresML by copying it from a foreign table into a regular table. Taking the example of the`users` table:

CREATETABLEpublic.users (LIKElive_db_tables.users);

INSERT INTOpublic.usersSELECT*FROMlive_db_tables.users;