Time series Timeseries Deep Learning Machine Learning Python Pytorch fastai | State-of-the-art Deep Learning library for Time Series and Sequences in Pytorch / fastai
I'm curious whether timeseriesAI supports multilabel regression. I have a dataset with the shape of (samples, features, time steps), and the labels have a shape of (samples, label_features). Can I use timeseriesAI to regress the data directly to these labels?
I noticed that there is a TSMultiLabelClassification() transform for multilabel classification, but I haven't found a similar transform for regression tasks. Are there any specific transforms for multilabel regression in timeseriesAI?
Also, I wanted to bring to your attention that the Colab link provided in the 01a_MultiClass_MultiLabel_TSClassification.ipynb tutorial seems to be incorrectly connected to 01_Intro_to_Time_Series_Classification.ipynb. Would you please fix this issue?
Hi there,
I'm curious whether timeseriesAI supports multilabel regression. I have a dataset with the shape of
(samples, features, time steps)
, and the labels have a shape of(samples, label_features)
. Can I use timeseriesAI to regress the data directly to these labels?I noticed that there is a
TSMultiLabelClassification()
transform for multilabel classification, but I haven't found a similar transform for regression tasks. Are there any specific transforms for multilabel regression in timeseriesAI?Also, I wanted to bring to your attention that the Colab link provided in the 01a_MultiClass_MultiLabel_TSClassification.ipynb tutorial seems to be incorrectly connected to 01_Intro_to_Time_Series_Classification.ipynb. Would you please fix this issue?
Thank you!