Snowpark ML is another public preview feature we’re very excited to get playing with. Now, data science skills are specialist skills, and so we don’t want to go into too hard here, so we’ve made this as easy as we can.
(1) Ingest the data
(2) Create a connection.json
file that looks a little like the one below
{
"account" : "",
"user" : "",
"password" : "",
"role" : "",
"warehouse" : "",
"database" : "",
"schema" : ""
}
(3) Connect to your data as in the code below
connection_parameters = json.load(open('credentials/connection.json'))
session = Session.builder.configs(connection_parameters).create()
(4) Fill in the gaps
#Load table
week59 = session.table('week59').to_pandas()
# Initialize the DecisionTreeClassifier with column names
model =
# Fit the model to the data
# Make predictions
# Display the predictions
Your result should look something like:

Remember, if you want to participate:
Thank you for pointing out new features. This is my default place for what’s new in Snowflake
Playing catch-up as I was on leave last week. Thanks for another fun new challenge, I hadn’t had a chance to play with Snowpark ML before!