ibm-granite
/

granite-timeseries-ttm-r1

Time Series Forecasting

pretrained models

foundation models

time series foundation models

Model card Files Files and versions

vijaye12 commited on Oct 19, 2024

Commit

c3fc0b2

·

verified ·

1 Parent(s): e5198b8

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -70,7 +70,7 @@ The below model scripts can be used for any of the above TTM models. Please upda
 TTM outperforms popular benchmarks such as TimesFM, Moirai, Chronos, Lag-Llama, Moment, GPT4TS, TimeLLM, LLMTime in zero/fewshot forecasting while reducing computational requirements significantly.
 Moreover, TTMs are lightweight and can be executed even on CPU-only machines, enhancing usability and fostering wider
 adoption in resource-constrained environments. For more details, refer to our [paper](https://arxiv.org/pdf/2401.03955.pdf) TTM-Q referred in the paper maps to the `512-96` model
-uploaded in the main branch. Please refer to our paper for more details.
 ## Recommended Use
 1. Users have to externally standard scale their data independently for every channel before feeding it to the model (Refer to [TSP](https://github.com/IBM/tsfm/blob/main/tsfm_public/toolkit/time_series_preprocessor.py), our data processing utility for data scaling.)

 TTM outperforms popular benchmarks such as TimesFM, Moirai, Chronos, Lag-Llama, Moment, GPT4TS, TimeLLM, LLMTime in zero/fewshot forecasting while reducing computational requirements significantly.
 Moreover, TTMs are lightweight and can be executed even on CPU-only machines, enhancing usability and fostering wider
 adoption in resource-constrained environments. For more details, refer to our [paper](https://arxiv.org/pdf/2401.03955.pdf) TTM-Q referred in the paper maps to the `512-96` model
+uploaded in the main branch. For other variants (TTM-B, TTM-E and TTM-A) please refer [here](https://huggingface.co/ibm-granite/granite-timeseries-ttm-r2). For more details, refer to the paper.
 ## Recommended Use
 1. Users have to externally standard scale their data independently for every channel before feeding it to the model (Refer to [TSP](https://github.com/IBM/tsfm/blob/main/tsfm_public/toolkit/time_series_preprocessor.py), our data processing utility for data scaling.)