Rocke Dong

Dear Alice and reymond-group, thanks for very much for your share of codes. I am trying to reproduce your codes, in the start of '09-Property-analysis.ipynb', df_all_nohem = pd.read_pickle(folder +"/pickles/all_sequences_with_NN_prop_helicity.pkl") But...

### Bug description pytorch nn.Module class, model_state_dict=model.state_dict(), assignmentPylintE1137:unsupported-assignment-operation ```python import torch import torch.nn as nn class Test(nn.Module): def __init__( self, embed_dim=3, ) -> None: super().__init__() self.embedding = torch.nn.Embedding(100, embed_dim) def...

support lightning 2.* and pytorch 2.*

2

## 🚀 Feature Support lightning 2.* and pytorch 2.* Is there a time plan? ### Motivation Pytorch lightning 2. and pytorch 2.* are both available and with nice new and...

run file: af/examples/peptide_binder_design.ipynb hope: the binder binds at the hotspot, e.g. 32-41 index in the target chain. results: the binder not binds at the hotspots at the target chain at...

Could partial_ratio support Levenshtein.normalized_similarity?

2

rapidfuzz 3.9.6 and python 3.10. I have carefully read the rapidfuzz docs and made tests. I find: partial_ratio use Indel.normalized_similarity. Could partial_ratio also support Levenshtein.normalized_similarity? Maybe default to use Indel.normalized_similarity,...

Dear, In my usage, it seems 70B is a crucial threshold for LM, >= 70B give a much stable and good performance. could you release Qwen3 VL 72B? thanks!!