gifteval
Datasets
All datasets matching “gifteval”GiftEvalPretrain
GIFT-Eval Pre-training Datasets
Pretraining dataset aligned with GIFT-Eval that has 71 univariate and 17 multivariate datasets, spanning seven domains and 13 frequencies, totaling 4.5 million time series and 230 billion data points. Notably this collection of data has no leakage issue with the train/test split and can be used to pretrain foundation models that can be fairly evaluated on GIFT-Eval.
📄 Paper
🖥️ Code
📔 Blog Post
🏎️ Leader Board
Ethical Considerations… See the full description on the dataset page: https://huggingface.co/datasets/Salesforce/GiftEvalPretrain.GiftEvalPretrain
GiftEval Pretrain (Derived)
This dataset is a derived version of Salesforce/GiftEvalPretrain, the pre-training dataset aligned with the GIFT-Eval benchmark for time series foundation models.
Differences from the original
Converted from Arrow to Parquet format
Same 144 dataset subsets, no changes to underlying time series values
Usage
from datasets import load_dataset
dataset = load_dataset("theforecastingcompany/GiftEvalPretrain", "m1_yearly")… See the full description on the dataset page: https://huggingface.co/datasets/theforecastingcompany/GiftEvalPretrain.GiftEvalPretrain
GIFT-Eval Pre-training Datasets
Pretraining dataset aligned with GIFT-Eval that has 71 univariate and 17 multivariate datasets, spanning seven domains and 13 frequencies, totaling 4.5 million time series and 230 billion data points. Notably this collection of data has no leakage issue with the train/test split and can be used to pretrain foundation models that can be fairly evaluated on GIFT-Eval.
📄 Paper
🖥️ Code
📔 Blog Post
🏎️ Leader Board
Ethical Considerations… See the full description on the dataset page: https://huggingface.co/datasets/CollectionStudio/GiftEvalPretrain.GiftEval
GIFT-Eval
We present GIFT-Eval, a benchmark designed to advance zero-shot time series forecasting by facilitating evaluation across diverse datasets. GIFT-Eval includes 23 datasets covering 144,000 time series and 177 million data points, with data spanning seven domains, 10 frequencies, and a range of forecast lengths. This benchmark aims to set a new standard, guiding future innovations in time series foundation models.
To facilitate the effective pretraining and evaluation… See the full description on the dataset page: https://huggingface.co/datasets/Salesforce/GiftEval.GiftEvalPretrain
GIFT-Eval Pre-training Datasets
Pretraining dataset aligned with GIFT-Eval that has 71 univariate and 17 multivariate datasets, spanning seven domains and 13 frequencies, totaling 4.5 million time series and 230 billion data points. Notably this collection of data has no leakage issue with the train/test split and can be used to pretrain foundation models that can be fairly evaluated on GIFT-Eval.
📄 Paper
🖥️ Code
📔 Blog Post
🏎️ Leader Board
Ethical Considerations… See the full description on the dataset page: https://huggingface.co/datasets/yangsir101/GiftEvalPretrain.gifteval
GIFT-Eval - TsFile Conversion
This repository contains a TsFile conversion of
Salesforce/GiftEval.
GIFT-Eval is a benchmark for zero-shot time-series forecasting. The original
dataset contains 23 datasets, about 144,000 time series and 177 million data
points across seven domains and ten frequencies.
Conversion Layout
Each source Arrow file is converted separately and the original
dataset/frequency hierarchy is preserved:
data/
<dataset>.tsfile… See the full description on the dataset page: https://huggingface.co/datasets/THULab/gifteval.