search

emadeldeen24 / TS-TCC

[IJCAI-21] "Time-Series Representation Learning via Temporal and Contextual Contrasting"
MIT License
357 stars 99 forks source link

time series forecasting #7

Closed zxhjames closed 2 years ago

zxhjames commented 2 years ago

Hi, great job! can this model be used for time series forecasting?How's the effect

emadeldeen24 commented 2 years ago

I didn't test on time series forecasting applications. However, self-supervised contrastive learning is used here to pretrain the encoder network from the unlabeled data, regardless the downstream task. You can apply the downstream task you want (i.e., classification, forecasting, anomaly detection, etc.) afterwards just by adding the suitable module .

zxhjames commented 2 years ago

thanks a lot, I will try to test it