5 ms·
What matters more to me usually in a time series context is how forecasting tools deal with non-stationarity of the underlying data process. I'm not an expert o
by BrokrnAlgorithm 3y ago
What matters more to me usually in a time series context is how forecasting tools deal with non-stationarity of the underlying data process. I'm not an expert on LLMs, but I assume they wont' be the ideal tool to use in these contexts because even finetuning will be rather expensive...?
- danielbln 3y agoFine-tuning with LoRA is pretty cheap though.