yandex-research/rtdl-num-embeddings

Issues

• Improve _NLinear implementation to use less RAM

  #19 opened a month ago by dholzmueller
  2

• Sensemaking

  #17 opened 2 months ago by firmai
  2

• compute_bins falls with CUDA out of memory for big datasets

  #18 opened 2 months ago by daniilkk
  0

• How can I save model parameters? I want to load the pre-trained model by myself in other experiments

  #16 opened 10 months ago by Phoebe-ovo
  2

• How can I get the feature name of the data (X_num_train.npy)?

  #15 opened 10 months ago by Phoebe-ovo
  1

• training config file

  #12 opened 10 months ago by morgsmss7
  4

• Parameter values ​​in exp/.../xx.toml

  #13 opened a year ago by gy0425
  5

• About periodic config

  #14 opened a year ago by Minhvt34
  5

• data issue

  #9 opened a year ago by Jacob-X
  7

• How to use it to evaluate on other datasets and for other embedding algorithms?

  #11 opened a year ago by herdonyan
  2

• TypeError: decoding str is not supported

  #10 opened a year ago by herdonyan
  6

• How to inverse normalize the predictions ?

  #8 opened 2 years ago by manmeet3591
  2

• What if use "continuous one-hot" or "bi-PLE" to alternate PLE

  #7 opened 2 years ago by xumwen
  4

• Perform a simple prediction from the tuned and trained best model ?

  #6 opened 2 years ago by manmeet3591
  4

• Dataset link not working ？

  #5 opened 2 years ago by ZhaoShengjian
  2

• About the feature shapes of embeddings in MLP vs FT_Transformer

  #4 opened 2 years ago by yiqings
  2

• Where are PLE embeddings?

  #3 opened 2 years ago by Froskekongen
  1