search

Nixtla / nixtla

TimeGPT-1: production ready pre-trained Time Series Foundation Model for forecasting and anomaly detection. Generative pretrained transformer for time series trained on over 100B data points. It's capable of accurately predicting various domains such as retail, electricity, finance, and IoT with just a few lines of code 🚀.
https://docs.nixtla.io
Other
1.91k stars 151 forks source link

How to evaluate the performance of the training set in the TimeGPT Model? #370

Closed kestlermai closed 1 month ago

kestlermai commented 1 month ago

Hey there, as a newbie diving into the TimeGPT tutorials (https://docs.nixtla.io/docs/getting-started-timegpt_quickstart), I've only come across performance evaluation for predictions, but I'm missing out on how to assess the trained models themselves. This has got me curious! If you could guide me through it, I would really appreciate it!

elephaint commented 1 month ago

Thanks for using TimeGPT! We currently don't provide direct access to the underlying models.

Hope this helps.

kestlermai commented 1 month ago

Thank you for your reply