SELECT statement
The SELECT
statement is used to get a predictons from the model table. The data is not persistent and is returned on the fly as a result-set. The basic syntax for selecting from the model is:
SELECT target_variable, target_variable_explain FROM model_table
WHERE when_data='{"column3": "value", "column2": "value"}';
Model table columns
The below list contains the column names of the model table. Note that target_varaiable_
will be the name of the target variable column.
- target_variable_original - The original value of the target variable.
- target_variable_min - Lower bound of the predicted value.
- target_variable_max - Upper bound of the predicted value.
- target_variable_confidence - Model confidence score.
- target_variable_explain - JSON object that contains additional information as
confidence_lower_bound
,confidence_upper_bound
,anomaly
,truth
. - when_data - The data to make the predictions from(WHERE clause params).
- select_data_query - SQL select query to create the datasource.
- external_datasource - Name of the pre-existing datasource that the model was built from.
rental_price | number_of_rooms | number_of_bathrooms | sqft | location | days_on_market | initial_price | neighborhood | rental_price_original | rental_price_min | rental_price_max | rental_price_confidence | rental_price_explain | when_data | select_data_query | external_datasource |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2450 | 4 | 2 | 800 | good | 12 | 2222 | downtown | nan | 2407 | 2493 | 0.99 | {"predicted_value": 2450, "confidence": 0.99, "confidence_lower_bound": 2407, "confidence_upper_bound": 2493, "anomaly": null, "truth": null} | {"sqft": 800, "number_of_rooms": 4, "number_of_bathrooms": 2, "location": "good", "days_on_market" : 12, "neighborhood": "downtown", "initial_price": "2222"} | nan | nan |
SELECT example
The following SQL statement selects all information from the home_rentals_model
for the property that has "sqft": 800, "number_of_rooms": 4, "number_of_bathrooms": 2,
"location": "good", "days_on_market" : 12, "neighborhood": "downtown", "initial_price": "2222".
SELECT * FROM mindsdb.home_rentals
WHERE when_data='{"sqft": 800, "number_of_rooms": 4, "number_of_bathrooms": 2,
"location": "good", "days_on_market" : 12,
"neighborhood": "downtown", "initial_price": "2222"}';
The following SQL statement selects only the target variable rental_price
as price
and the home_rentals_model
confidence as accuracy
:
SELECT rental_price as price,
rental_price_confidence as accuracy
FROM mindsdb.home_rentals WHERE when_data='{"sqft": 800, "number_of_rooms": 4, "number_of_bathrooms": 2,
"location": "good", "days_on_market" : 12,
"neighborhood": "downtown", "initial_price": "2222"}';