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

Error in the quick start guide. #385

Closed BalancingComic closed 3 weeks ago

BalancingComic commented 1 month ago

Hello nixtla

I am trying to test your software. And when i started and used the quick start example using the electricity short.csv doing a forecast i couldnt get it to work. the problem is that the column topics in the csv is not correct. So if you change the following row in your example: 

 nixtla_client.plot(df, timegpt_fcst_df, time_col='timestamp', target_col='value', level=[80, 90])

to this:

nixtla_client.plot(df, timegpt_fcst_df, time_col='ds', target_col='y', level=[80, 90])

it will work.

jmoralez commented 3 weeks ago

Hey. The time_col argument is used to identify the column that holds the times, in the case of the quickstart the name is timestamp as you can see here. If your data has a different column name you have to set the argument accordingly.