Does this support tensor? Meaning each sample is MxNxT

dmmiller612 / sparktorch

Train and run Pytorch models on Apache Spark.

MIT License

339 stars 44 forks source link

Does this support tensor? Meaning each sample is MxNxT #23

Open ghost opened 3 years ago

ghost commented 3 years ago

I have M users in a batch each one have T time points at each N measurement are saved. Does spark/spark torch supports feeding the model with tensor rather than a flat vector of vectorAssembler?