RetroSearch Browse

Home - News ( United States | United Kingdom | Italy | Germany ) - Football scores

Showing content from https://docs.nixtla.io/docs/forecasting/timegpt_quickstart below:

TimeGPT Foundational model for time series forecasting and anomaly detection

TimeGPT is a production-ready generative pretrained transformer for time series. It can accurately predict domains such as retail, electricity, finance, and IoT with just a few lines of code. Get started below!

Implementation Guide

Create a TimeGPT account and generate your API key

• Visit dashboard.nixtla.io to activate your free trial and create an account. • Sign in using Google, GitHub, or your email. • Navigate to API Keys in the menu and select Create New API Key. • Your new API key will appear on the screen. Copy this key using the button on the right.

Dashboard displaying TimeGPT API keys and controls.

Install Nixtla

Install the Nixtla library in your preferred Python environment:

Import the Nixtla TimeGPT client

Import the Nixtla client and instantiate it with your API key:

from nixtla import NixtlaClient

nixtla_client = NixtlaClient(
    api_key='my_api_key_provided_by_nixtla'
)

Verify the status and validity of your API key:

nixtla_client.validate_api_key()

API Validation Output

INFO:nixtla.nixtla_client:Happy Forecasting! :), If you have questions or need support, please email support@nixtla.io

True

Start making forecasts! 1. Load the AirPassengers Dataset

We will use the classic AirPassengers dataset to demonstrate forecasts.

2. Preview the Dataset

Quickly examine structures like timestamps and values before forecasting.

3. Plot the Time Series

Visualize historical data to understand trends or seasonality.

import pandas as pd

df = pd.read_csv('https://raw.githubusercontent.com/Nixtla/transfer-learning-time-series/main/datasets/air_passengers.csv')
df.head()

timestamp value 0 1949-01-01 112 1 1949-02-01 118 2 1949-03-01 132 3 1949-04-01 129 4 1949-05-01 121 Plot the dataset:

nixtla_client.plot(df, time_col='timestamp', target_col='value')

Historical AirPassengers data from 1949 to 1960.

Data Requirements

The target variable must not contain missing or non-numeric values.
Date stamps must form a continuous sequence without gaps for the selected frequency.
Pandas should correctly parse the timestamp column (see Pandas documentation).
The forecast method does not fill or handle missing dates.

For more details, visit Data Requirements.

Saving Figures from TimeGPT

The plot method automatically displays figures in notebook environments. To save a plot locally:

fig = nixtla_client.plot(df, time_col='timestamp', target_col='value')
fig.savefig('plot.png', bbox_inches='tight')

Short and Long-Term Forecasting Examples

Generate a longer-term forecast

Forecast the next 12 months using the SDK’s forecast method:

timegpt_fcst_df = nixtla_client.forecast(
    df=df,
    h=12,
    freq='MS',
    time_col='timestamp',
    target_col='value'
)
timegpt_fcst_df.head()

Display the forecast:

nixtla_client.plot(df, timegpt_fcst_df, time_col='timestamp', target_col='value')

12-month forecast for AirPassengers data.

You may also generate forecasts for longer horizons with the timegpt-1-long-horizon model. For example, 36 months ahead:

Forecast 36 Months Long Horizon

timegpt_fcst_df = nixtla_client.forecast(
    df=df,
    h=36,
    freq='MS',
    time_col='timestamp',
    target_col='value',
    model='timegpt-1-long-horizon'
)
timegpt_fcst_df.head()

nixtla_client.plot(df, timegpt_fcst_df, time_col='timestamp', target_col='value')

36-month forecast using the 'timegpt-1-long-horizon' model.

Generate a shorter-term forecast

Forecast the next 6 months with a single command:

timegpt_fcst_df = nixtla_client.forecast(
    df=df,
    h=6,
    freq='MS',
    time_col='timestamp',
    target_col='value'
)
nixtla_client.plot(df, timegpt_fcst_df, time_col='timestamp', target_col='value')

6-month forecast for AirPassengers data.

RetroSearch is an open source project built by @garambo | Open a GitHub Issue

Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo

HTML: 3.2 | Encoding: UTF-8 | Version: 0.7.4