Add confidence levels

Tweak the confidence level used for anomaly detection. By default, if a value falls outside the 99% confidence interval, it is labeled as an anomaly.

Modify this with the level parameter, which accepts any value between 0 and 100, including decimals.

Increasing the level results in fewer anomalies detected, while decreasing the level increases the number of anomalies detected.

import pandas as pd
from nixtla import NixtlaClient

nixtla_client = NixtlaClient(
    # defaults to os.environ.get("NIXTLA_API_KEY")
    api_key = 'my_api_key_provided_by_nixtla'
)

👍

Use an Azure AI endpoint

To use an Azure AI endpoint, set the base_url argument:

nixtla_client = NixtlaClient(base_url="you azure ai endpoint", api_key="your api_key")

# Read the data
df = pd.read_csv('https://raw.githubusercontent.com/Nixtla/transfer-learning-time-series/main/datasets/peyton_manning.csv')

# Anomaly detection using a 70% confidence interval
anomalies_df = nixtla_client.detect_anomalies(
    df, 
    time_col='timestamp', 
    target_col='value', 
    freq='D',
    level=70
)

# Plot anomalies
nixtla_client.plot(
    df, 
    anomalies_df,
    time_col='timestamp', 
    target_col='value'
)

INFO:nixtla.nixtla_client:Validating inputs...
INFO:nixtla.nixtla_client:Preprocessing dataframes...
INFO:nixtla.nixtla_client:Calling Anomaly Detector Endpoint...

📘

Available models in Azure AI

If you use an Azure AI endpoint, set model="azureai"

nixtla_client.detect_anomalies(..., model="azureai")

For the public API, two models are supported: timegpt-1 and timegpt-1-long-horizon.

By default, timegpt-1 is used. See this tutorial for details on using timegpt-1-long-horizon.

For more information, read our detailed tutorial on anomaly detection.