Informazioni su questo codelab
1. Introduzione
Questo codelab ti insegnerà a utilizzare la regressione lineare per creare un modello che prevede il costo per clic.
Prerequisiti
Per completare questo codelab, ti serviranno:
Per completare questo codelab, ti servirà un volume sufficiente di dati della campagna di alta qualità per creare un modello.
2. Crea una tabella temporanea
Esegui la query seguente:
CREATE TABLE
linear_regression_example_data
AS(
WITH all_data AS (
SELECT
imp.user_id as user_id,
ROW_NUMBER() OVER(PARTITION BY imp.user_id) AS rowIdx,
imp.browser AS browser_name,
gender_name AS gender_name,
age_group_name AS age_group_name,
DATETIME(TIMESTAMP_MICROS(
imp.query_id.time_usec), "America/Los_Angeles") as impression_time,
clk.advertiser_click_cost_usd AS label
FROM adh.google_ads_impressions imp
INNER JOIN adh.google_ads_clicks clk USING (impression_id)
LEFT JOIN adh.gender ON demographics.gender = gender_id
LEFT JOIN adh.age_group ON demographics.age_group = age_group_id
)
# Need just one user ID or regression won't work
SELECT
label,
browser_name,
gender_name,
age_group_name,
# Although BQML could divide impression_time into several useful variables on
# its own, it may attempt to divide it into too many features. As a best
# practice extract the variables that you think will be most helpful.
# The output of impression_time is a number, but we care about it as a
# category, so we cast it to a string.
CAST(EXTRACT(DAYOFWEEK FROM impression_time) AS STRING) AS day_of_week,
CAST(EXTRACT(HOUR FROM impression_time) AS STRING) AS hour,
FROM
all_data
WHERE
rowIdx = 1 # This ensures that there's only 1 row per user.
AND
label IS NOT NULL
AND
gender_name IS NOT NULL
AND
age_group_name IS NOT NULL
);
3. Creare e addestrare un modello
Una best practice consiste nel separare i passaggi di creazione delle tabelle da quelli di creazione dei modelli.
Esegui la seguente query sulla tabella temporanea che hai creato nel passaggio precedente. Non preoccuparti di indicare le date di inizio e di fine, poiché verranno dedotte in base ai dati della tabella temporanea.
CREATE OR REPLACE
MODEL `example_linear`
OPTIONS(
model_type = 'adh_linear_regression'
)
AS (
SELECT *
FROM
tmp.linear_regression_example_data
);
SELECT * FROM ML.EVALUATE(MODEL `example_linear`)
Riga | mean_absolute_error | mean_squared_error | mean_squared_log_error | median_absolute_error | r2_score | explained_variance |
1 | 0.11102380666874107 | 0.019938972461569476 | 0.019503393448234131 | 0.091792024503562136 | -9.8205955364568478 | -9.7975398794423025 |