Verarbeiten von Daten mit benutzerdefinierter Logik über Partitionen hinweg¶

from snowflake.ml.utils.connection_params import SnowflakeLoginOptions
from snowflake.ml.modeling.distributors.distributed_partition_function.dpf import DPF
from snowflake.ml.modeling.distributors.distributed_partition_function.entities import RunStatus

# Define function to process each region's data

def process_sales_data(data_connector, context):
    df = data_connector.to_pandas()
    print(f"Processing {len(df)} records for region: {context.partition_id}")

    # Perform region-specific analytics
    summary = {
        'region': context.partition_id,
        'total_sales': df['amount'].sum(),
        'avg_order_value': df['amount'].mean(),
        'customer_count': df['customer_id'].nunique(),
        'record_count': len(df)
    }

    # Store results in stage for subsequent access
    import json
    context.upload_to_stage(summary, "sales_summary.json",
        write_function=lambda obj, path: json.dump(obj, open(path, 'w')))

dpf = DPF(process_sales_data, "analytics_stage")
run = dpf.run(
    partition_by="region",  # Creates separate partitions for North, South, East, West
    snowpark_dataframe=sales_data,
    run_id="regional_analytics_2024"
)

from snowflake.ml.runtime_cluster import scale_cluster

# Scale cluster before running tuner
scale_cluster(2)  # Scale to 2 nodes for parallel trials

final_status = run.wait()  # Shows progress bar by default
print(f"Job completed with status: {final_status}")

if final_status == RunStatus.SUCCESS:
    # Get results from each region
    import json
    all_results = []
    for partition_id, details in run.partition_details.items():
        summary = details.stage_artifacts_manager.get("sales_summary.json",
            read_function=lambda path: json.load(open(path, 'r')))
        all_results.append(summary)

    # Combine results across all regions
    total_sales = sum(r['total_sales'] for r in all_results)
    total_customers = sum(r['customer_count'] for r in all_results)
else:
    # Handle failures - check logs for failed partitions
    for partition_id, details in run.partition_details.items():
        if details.status != "DONE":
            error_logs = details.logs

# Restore completed run from stage and access same results as above without re-running.
from snowflake.ml.modeling.distributors.distributed_partition_function.dpf_run import (
    DPFRun
)
restored_run = DPFRun.restore_from("regional_analytics_2024", "analytics_stage")