Lightwood Embedding Model

This tutorial uses the Lightwood handler to create an embedding model.

Tutorial

The following example shows how to create an embedding model using the Lightwood engine.

Start by creating an engine from the Lightwood handler.

CREATE ML_ENGINE lightwood
FROM lightwood;

Verify that the engine has been created successfully.

SHOW ML_ENGINES
WHERE name='lightwood';

Connect our sample database to use it for training the model:

CREATE DATABASE example_db
WITH ENGINE = "postgres",
PARAMETERS = {
"user": "demo_user",
"password": "demo_password",
"host": "samples.mindsdb.com",
"port": "5432",
"database": "demo",
"schema": "demo_data"
};

Create a model using this engine:

CREATE MODEL home_rentals_model_embeddings
FROM example_db
  (SELECT * FROM demo_data.home_rentals)
PREDICT rental_price
USING
    problem_definition.embedding_only = True;

Check the status of the model.

DESCRIBE home_rentals_model_embeddings;

Now you can use the model to predict home rental prices for specific criteria and get predictions in the form of embeddings.

SELECT rental_price, rental_price_explain
FROM home_rentals_model_embeddings
WHERE sqft = 823
AND location='good'
AND neighborhood='downtown'
AND days_on_market=10
USING
    return_embedding = True;

Here is the output:

+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| rental_price                                                                                                                                                                                                                 | rental_price_explain                                                                                                                                                                                                                                                                                                                                                             |
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| [1,6.712956428527832,1.247057318687439,0,0,1,0,0,0,0,0,1,0,0,0,1,2.3025851249694824,0.5629426836967468,0,1,0,0,0,0,0,0,0.7540000081062317,0.3333333432674408,0.35483869910240173,0.9583333134651184,0.7833333611488342,0.25] | {"predicted_value": [1.0, 6.712956428527832, 1.247057318687439, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 1.0, 2.3025851249694824, 0.5629426836967468, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.7540000081062317, 0.3333333432674408, 0.35483869910240173, 0.9583333134651184, 0.7833333611488342, 0.25], "confidence": null, "anomaly": null, "truth": null} |
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+

Use Cases

​Tutorial

Tutorial