---

title: "Python Example"

Python Example

This example scores a CSV file using the Profile Scoring API with Python's requests library.

Prerequisites

pip install requests pandas

Full example

import requests
import pandas as pd
import os

API_KEY = os.environ["AUSYNTH_API_KEY"]
SCORE_URL = "https://api.ausynth.com/api/score"

# Upload the file
with open("my_customers.csv", "rb") as f:
    response = requests.post(
        SCORE_URL,
        headers={"Authorization": f"Bearer {API_KEY}"},
        files={"file": ("my_customers.csv", f, "text/csv")},
    )

response.raise_for_status()
result = response.json()

# Check the response
if result["status"] == "complete":
    # Small file - results are ready
    print(f"Scored {result['rows_scored']} rows, charged {result['credits_charged']} credits")

    # Download the scored CSV
    download = requests.get(result["download_url"])
    with open("scored_output.csv", "wb") as out:
        out.write(download.content)

    # Load and inspect
    df = pd.read_csv("scored_output.csv")
    print(df[["id", "profile_name", "profile_confidence"]].head())
    print(df["profile_name"].value_counts())

elif result["status"] == "processing":
    # Large file - poll for completion
    import time
    poll_url = f"https://api.ausynth.com{result['poll_url']}"
    while True:
        time.sleep(5)
        poll = requests.get(poll_url, headers={"Authorization": f"Bearer {API_KEY}"})
        poll_result = poll.json()
        if poll_result["status"] == "complete":
            download = requests.get(poll_result["download_url"])
            with open("scored_output.csv", "wb") as out:
                out.write(download.content)
            print(f"Scored {poll_result['rows_scored']} rows")
            break
        print(f"Still processing... ({poll_result.get('estimated_completion_seconds', '?')}s remaining)")

Checking data quality flags

The response includes a data_quality object that reports any values the API did not recognise:

result = response.json()
dq = result.get("data_quality", {})

if dq.get("rows_with_invalid_values", 0) > 0:
    print(f"Warning: {dq['rows_with_invalid_values']} rows had unrecognised values")
    for detail in dq["invalid_value_details"][:5]:
        print(f"  Row {detail['row']}: {detail['field']}='{detail['value']}' -> mapped to Not stated")

Handling errors

response = requests.post(SCORE_URL, headers={"Authorization": f"Bearer {API_KEY}"}, files={"file": f})

if response.status_code == 401:
    print("Invalid API key. Check AUSYNTH_API_KEY.")
elif response.status_code == 402:
    err = response.json()
    print(f"Need {err['credits_required']} credits, have {err['credits_available']}")
elif response.status_code == 429:
    err = response.json()
    print(f"Rate limited. Retry in {err['retry_after_seconds']} seconds.")
elif response.status_code >= 400:
    print(f"Error {response.status_code}: {response.json()}")

Using environment variables for the key

Set the key in your shell before running the script:

export AUSYNTH_API_KEY="ask_live_abc123def456"
python score_my_data.py

Never hard-code the key in your script or commit it to version control.