MaxCompute:Scenario practices

Write the Python UDF script and save it assum_pandas.py. The following sample script is shown below:

from odps.udf import annotate
import pandas as pd

@annotate("string, string -> string")
class SumColumns(object):
    def evaluate(self, arg1, arg2):
        # Convert input parameters to pandas DataFrame
        df = pd.DataFrame({'col1': arg1.split(','), 'col2': arg2.split(',')})

        # Perform data processing operations by using pandas
        # Calculate the sum of two columns as an example
        df['sum'] = df['col1'].astype(int) + df['col2'].astype(int)

        # Convert the processing result to a string and return
        result = ','.join(df['sum'].astype(str).values)
        return result

Upload the sum_pandas.py script as a resource to your MaxCompute project. For details, see Add resources. Use the following command:

ADD PY sum_pandas.py -f;

Register the sum_pandas.py script as the SumColumns UDF. For details, see Create a UDF. Use the following command:

CREATE FUNCTION SumColumns AS 'sum_pandas.SumColumns' USING 'sum_pandas.py';

Set up the test table testsum with the corresponding test data.

CREATE TABLE testsum (col1 string, col2 string);
INSERT INTO testsum VALUES ('1,2,3','1,2,3'),('1,2,3','3,2,1'),('1,2,3','4,5,6');

Specify the image through the Flag parameter when calling the UDF.

set odps.sql.python.version=cp37;
set odps.session.image = ;
SELECT SumColumns(col1,col2) AS result FROM testsum;

Expected result:

+------------+
| result     |
+------------+
| 2,4,6      |
| 4,4,4      |
| 5,7,9      |
+------------+

Prepare the test tabletest_float_col and insert the test data.

CREATE TABLE test_float_col (col1 double);
INSERT INTO test_float_col VALUES (3.75),(2.51);

Write the PyODPS code to compute psi(col1) and save it aspsi_col.py. The following sample code is shown below:

import os
from odps import ODPS, options

def my_psi(v):
    from scipy.special import psi

    return float(psi(v))

# If the project enables isolation, the following option is not required
options.sql.settings = {"odps.isolation.session.enable": True}

o = ODPS(
      # Ensure that the ALIBABA_CLOUD_ACCESS_KEY_ID environment variable is set to your AccessKey ID,
      # and the ALIBABA_CLOUD_ACCESS_KEY_SECRET environment variable is set to your AccessKey secret.
      # It is not recommended to directly use the AccessKey ID and AccessKey secret strings.
      os.getenv('ALIBABA_CLOUD_ACCESS_KEY_ID'),
      os.getenv('ALIBABA_CLOUD_ACCESS_KEY_SECRET'),
      project='your-default-project',
      endpoint='your-end-point'
)

df = o.get_table("test_float_col").to_df()
# Execute directly and get the result
df.col1.map(my_psi).execute(image='scipy')
# Save to another table
df.col1.map(my_psi).persist("result_table", image='scipy')

View the results in theresult_table table.

SELECT * FROM result_table

Expected result:

+----------------------+
| col1                 |
+----------------------+
| 1.1825373886117962   |
| 0.7080484451910534   |
+----------------------+

Create the test tabletest_float_col and insert the test data.

CREATE TABLE test_float_col (col1 double);
INSERT INTO test_float_col VALUES (3.75),(2.51);

Write the MaxFrame code to compute psi(col1) and save it aspsi_col.py. The following sample code is shown below:

import os
from odps import ODPS, options
from maxframe.session import new_session
import maxframe.dataframe as md

from maxframe.config import options
from maxframe import config

# Use the built-in scipy image
config.options.sql.settings = {
    "odps.session.image": "scipy"
}
def my_psi(v):
    from scipy.special import psi
    return float(psi(v))

o = ODPS(
      # Ensure that the ALIBABA_CLOUD_ACCESS_KEY_ID environment variable is set to your AccessKey ID,
      # and the ALIBABA_CLOUD_ACCESS_KEY_SECRET environment variable is set to your AccessKey secret.
      # It is not recommended to directly use the AccessKey ID and AccessKey secret strings.
      os.getenv('ALIBABA_CLOUD_ACCESS_KEY_ID'),
      os.getenv('ALIBABA_CLOUD_ACCESS_KEY_SECRET'),
      project='your-default-project',
      endpoint='your-end-point'
)

# Create a MaxFrame session
session = new_session(o)
df = md.read_odps_table('test_float_col')

# Execute and get the result
print(df.col1.map(my_psi).execute().fetch()

Expected result:

0    1.182537
1    0.708048
Name: col1, dtype: float64