Hey, I'm interested in reproducing the results of the paper on the DES15K dataset (Table 8), but I'm unable to find the training configuration/hyperparamters that was used. Also it seems that the dataset is not actually implemented in torchmdnet/datatsets. (I assume no specific processing is necessary?)
It would be great if the config (similar to those for e.g. QM9) could be shared here as well.
Hey, I'm interested in reproducing the results of the paper on the DES15K dataset (Table 8), but I'm unable to find the training configuration/hyperparamters that was used. Also it seems that the dataset is not actually implemented in
torchmdnet/datatsets. (I assume no specific processing is necessary?)It would be great if the config (similar to those for e.g. QM9) could be shared here as well.