Skip to content

GitLab

"...llamacpp/git@developer.sourcefind.cn:OpenDAS/dynamo.git" did not exist on "ceaeba3e3858432fd8b5e696eb6c2c8a64ffd632"
Commit bc240326 authored by Geoffrey Yu's avatar Geoffrey Yu Committed by Jennifer Wei
Browse files

added comments

parent 04d63782
Hide whitespace changes
Inline Side-by-side
Showing with 5 additions and 2 deletions
openfold/utils/multi_chain_permutation.py
View file @ bc240326
...@@ -424,8 +424,11 @@ def compute_permutation_alignment(out: Dict[str,torch.Tensor], ...@@ -424,8 +424,11 @@ def compute_permutation_alignment(out: Dict[str,torch.Tensor],
features: Dict[str,torch.Tensor], features: Dict[str,torch.Tensor],
ground_truth: List[Dict[str, torch.Tensor]]) -> Tuple[List[Tuple[int, int]], Dict[int, List[int]]]: ground_truth: List[Dict[str, torch.Tensor]]) -> Tuple[List[Tuple[int, int]], Dict[int, List[int]]]:
""" """
A method that permutes chains in ground truth A method that permutes chains in ground truth before calculating the loss
before calculating the loss. because the mapping between the predicted and ground-truth will become arbitrary.
The model cannot be assumed to predict chains in the same order as the ground truth.
Thus, this function pick the optimal permutaion of predicted chains that best matches the ground truth,
by minimising the RMSD.
Args: Args:
out: a dictionary of output tensors from model.forward() out: a dictionary of output tensors from model.forward()
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment