11# @faim-group/sdk-forecasting
22
3- Modern zero -shot time-series forecasting with advanced neural networks . Get accurate predictions without historical tuning or complex setup.
3+ Zero -shot time-series forecasting in JavaScript using state-of-the-art foundation models . Get accurate predictions without training or complex setup.
44
55** 📦 [ View on npm] ( https://www.npmjs.com/package/@faim-group/sdk-forecasting ) ** | ** 📚 [ GitHub] ( https://github.com/S-FM/faim-js-client ) ** | ** 🌐 [ Website] ( https://faim.it.com/ ) **
66
77## Features
88
99- ** Zero-Shot Forecasting** - Works out-of-the-box without model training
10- - ** Advanced Models** - Chronos2 and TiRex for different scenarios
10+ - ** Advanced Models** - Chronos2 from AWS and TiRex from NXAI
1111- ** Point & Quantile Forecasts** - Get predictions and uncertainty estimates
1212- ** Type-Safe** - Full TypeScript support with strict type checking
1313- ** Automatic Retries** - Built-in exponential backoff for reliability
@@ -27,15 +27,6 @@ pnpm add @faim-group/sdk-forecasting
2727yarn add @faim-group/sdk-forecasting
2828```
2929
30- ### Version
31-
32- Latest version: ** 0.1.0**
33-
34- ``` bash
35- # Install specific version
36- npm install @faim-group/sdk-forecasting@0.1.0
37- ```
38-
3930## Quick Start
4031
4132### 1. Get an API Key
@@ -222,9 +213,7 @@ pnpm tsx examples/air_passengers.ts
222213```
223214
224215The Air Passengers example demonstrates:
225- - Real-world dataset download from GitHub
226- - Point forecasting with both Chronos2 and TiRex models
227- - Probabilistic forecasting with confidence intervals (90% bounds)
216+ - Point and probabilistic forecasting with both Chronos2 and TiRex models
228217- Metric calculation (MAE, MSE)
229218- Formatted table output with predictions and confidence intervals
230219
0 commit comments