ehoogeboom/e3_diffusion_for_molecules

Issues

• MAE

  #39 opened 3 months ago by xzliu6
  0

• Questions regarding dataset error

  #23 opened a year ago by ZaiyunLin
  2

• Instructions to turn off equivarinace?

  #38 opened 9 months ago by Ala1s
  2

• Training time for Geom Drug & Incorrect mask generation?

  #37 opened 9 months ago by Accelerator-thu
  2

• Turning off equivariance?

  #25 opened a year ago by MilesCranmer
  3

• Training time for QM9?

  #36 opened 10 months ago by sudarshan1994
  2

• Request for Release of Pre-trained Model Checkpoints

  #34 opened 10 months ago by yanliang3612
  1

• Use e3_diffusion for sphere packs generation

  #33 opened 10 months ago by charlenebruno
  1

• Inquiry Regarding the Creation of the qm9_second_half Dataset

  #35 opened 10 months ago by yanliang3612
  1

• How many samples are generated in the GEOM-drugs experiment?

  #19 opened a year ago by DongkiKim95
  1

• Question related to the equivariance

  #17 opened a year ago by Frankie123421
  1

• Train on custom dataset

  #20 opened a year ago by zhenzey
  3

• Question regarding property norms calculation

  #16 opened a year ago by amorehead
  0

• Path to pretrained QM9 property classifiers is missing

  #21 opened a year ago by amorehead
  2

• Missing argument for name of dataset

  #15 opened a year ago by amorehead
  0

• Problem is already solved.

  #18 opened a year ago by tryumanshow
  0

• About dataset_info in datasets_config.py

  #13 opened a year ago by lsnty5190
  2

• Question regarding the dimension of QM9 and input to property classifier

  #22 opened a year ago by yogeshverma1998
  2

• AssertionError: Variables not masked properly.

  #24 opened a year ago by sundevil0405
  1

• How to convert EDM results to SMILES form?

  #12 opened a year ago by WoohyunSonTMS
  1

• energies for the molecules generated

  #26 opened a year ago by pearl-rabbit
  1

• Question about the code

  #27 opened a year ago by 545487677
  1

• Training Split for G-SchNet

  #28 opened a year ago by ameya98
  2

• Differences between paper and code

  #30 opened a year ago by oliviertrottier
  1

• Does the code use Unet?

  #31 opened a year ago by Littletoyone
  1

• Explicit valence error in training

  #32 opened a year ago by lajictw
  1

• Validity and uniqueness of molecules generated by G-schent

  #29 opened a year ago by pearl-rabbit
  0

• Question about the function log_constants_p_x_given_z0() in EnVariationalDiffusion

  #14 opened 2 years ago by Humanoid-z
  2

• Question about the units of gap, homo and lumo in conditional gneration task

  #9 opened 2 years ago by Layne-Huang
  1

• question about error during training

  #11 opened 2 years ago by orgw
  0

• About the training loss

  #10 opened 2 years ago by jkwang93
  4

• Question about the paper

  #8 opened 2 years ago by Frankie123421
  2

• question regarding EGCL layer

  #7 opened 2 years ago by orgw
  0

• eval_sample.py looking for a model file that doesn't exist

  #6 opened 2 years ago by Dunni3
  1

• aromatic bonds??

  #5 opened 3 years ago by orgw
  1

• Why did you remove the input coordinates in EGNN?

  #4 opened 3 years ago by Layne-Huang
  1

• Trained model available?

  #2 opened 3 years ago by hjuinj
  3

• question regarding generation

  #3 opened 3 years ago by orgw
  2

• Why set the normalize value and bias?

  #1 opened 3 years ago by Layne-Huang
  1