Short Context Lengths Lead to Low Performance

zalandoresearch / pytorch-ts

PyTorch based Probabilistic Time Series forecasting framework based on GluonTS backend

MIT License

1.21k stars 191 forks source link

Short Context Lengths Lead to Low Performance #119

Open hdeep03 opened 1 year ago

hdeep03 commented 1 year ago

Using context lengths of 7 or less lead to poor performance on a medical timeseries dataset. Is there a minimum context_length for DeepVAR below which one wouldn't recommend using the model?