How to do model fine tuning?

[colemanhindes]

Really cool project! Enjoy the paper and have had fun testing it out. Will instructions on fine tuning be released?

Thanks for your time

[abdulfatir]

@colemanhindes Thanks for your interest. We are planning to release the training scripts soon but due to some other engagements there's no ETA yet. In the meantime, @canerturkmen and @shchur are working towards integrating Chronos into AutoGluon-TimeSeries (autogluon/autogluon#3978) and they're also planning to offer ways of fine-tuning the models.

[Saeufer]

+1 for this, if possible please mind #22 too for some custom data. Thanks!

[HALF111]

+1, looking forward to releasing the scripts of training and fine-tuning!

[TPF2017]

+1, looking forward to releasing the scripts of training and fine-tuning!

[0xrushi]

I caught a glimpse of it and noticed it's utilizing a torch.nn model. I've put together this notebook for training/finetuning. Could someone verify if it's set up correctly? The losses seem unusual, but I suspect it's due to the dataset being quite small and my use of:

sequence_length = 10
prediction_length = 5

notebook: here

[iganggang]

+1 for this, if possible please mind #22 too for some custom data. Thanks!

[lostella]

Training and fine-tuning script was added in #63, together with configurations that were used for pretraining the models on HuggingFace. We still need to add proper documentation, but roughly speaking:

• required dependencies can be installed with pip install ".[training]" (or pip install "chronos[training] @ git+https://github.com/amazon-science/chronos-forecasting.git"
• python scripts/training/train.py --help lists all available options
• the config files in .scripts/training/config can be adapted by
  • changing data files to other GluonTS-compatible files (arrow format recommended for efficiency, but parquet and json lines also supported)
  • pointing to Chronos models (instead of the original T5), setting random_init: false, and adjusting learning rate and number of steps for fine-tuning

Happy training! cc @colemanhindes @Saeufer @HALF111 @TPF2017 @0xrushi @iganggang

[lostella]

More detailed examples at: https://github.com/amazon-science/chronos-forecasting/tree/main/scripts

[Alonelymess]

I get this error when training chronos-t5-small:
ValueError: --tf32 requires Ampere or a newer GPU arch, cuda>=11 and torch>=1.7

[abdulfatir]

@Alonelymess that means your GPU does not support TF32 floating point format. Please run training/fine-tuning with the --no-tf32 flag or set tf32 to false in your yaml config.