Using Nvidia pre-trained models for prediction

[user14171]

Hey there!
Is there an Nvidia pre-trained NN model I can use for prediction, such as weather forecasts or time trends?
If yes, could you please provide an example?

Thank you!

[Morganh]

You can refer to Joint Intent and Slot Classification — TAO Toolkit 3.22.05 documentation and Question Answering — TAO Toolkit 3.22.05 documentation .

Jupyter notebooks: Intent Slot Classification Notebook | NVIDIA NGC

From training notebook,
Note: If you wish to proceed with a trained dataset for better inference results, you can find a .nemo model here.

Models:
From NeMo | NVIDIA NGC
[Natural Language Processing]
(NeMo Natural Language Processing models | NVIDIA NGC) (NLP) models. Popular models here are BERT base and BERT large fine tuned for several tasks such as Question Answering, Named Entity Recognition and many more. Bio-Megatron is a state of the art model for medical data
NeMo Natural Language Processing models | NVIDIA NGC

[user14171]

Thanks for the answer @Morganh !
The models that you suggested are more NLP-oriented, and Im looking more for models that are based on time series prediction.
For example, given a date and value → predict the value for the tomorrow date ( ie the temperature)

Did I miss understand you, or maybe you know more relevant Nvidia pre-train models that can be a good starting point?

Best wishes!

[Morganh]

There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one.
Thanks

@user14171

For example, given a date and value → predict the value for the tomorrow date ( ie the temperature)

OK, you can have a look at Time Series Forecasting with the NVIDIA Time Series Prediction Platform and Triton Inference Server | NVIDIA Technical Blog