search

time-series-foundation-models / lag-llama

Lag-Llama: Towards Foundation Models for Probabilistic Time Series Forecasting
Apache License 2.0
1.08k stars 121 forks source link

Fails to Reproduce Experiment as Described in Your Paper #37

Open oublalkhalid opened 3 months ago

oublalkhalid commented 3 months ago

Hello, thank you for sharing the initial code. I aim to replicate the experiment outlined in your paper before proceeding to fine-tune the model with additional datasets. Could you kindly assist me in accessing a tutorial on reproducing the experiment results? Currently, the model seems to lack convergence, so any guidance you could provide would be greatly appreciated.

Once again, thank you for your efforts.

best, koublal

ashok-arjun commented 3 months ago

Hi @oublalkhalid, thanks for the issue. We are planning to release the code for finetuning and reproducing the results in the paper by the next week (by end of march).

ashok-arjun commented 2 months ago

Hi @oublalkhalid Can you please share additional information about the results? You can reach me by email too if you'd like (arjun [dot] ashok @ servicenow [dot] com)

ashok-arjun commented 2 months ago

Hi @oublalkhalid Please check the repo now. All scripts have been released.

julian-vp commented 2 months ago

Hi @ashok-arjun

When running the pretrain.sh scripts I am always getting this error, and I tried with different datasets 

  File "/lag-llama/data/data_utils.py", line 40, in __next__
    (dataset,) = self._rng.choices(self._datasets, weights=self._weights, k=1)
  File "/.conda/lib/python3.10/random.py", line 533, in choices
    total = cum_weights[-1] + 0.0   # convert to float
IndexError: list index out of range