MCPcopy Index your code

hub / github.com/bytedance/dplm / functions

Functions1,008 in github.com/bytedance/dplm

↓ 3 callersMethodcompute_loss
(self, batch, weighting="linear")
src/byprot/models/dplm2/dplm2_bit.py:149
↓ 3 callersFunctiond
(i, j)
src/byprot/modules/metrics.py:21
↓ 3 callersMethoddecode
(self, quant, aatype, mask, decoder_kwargs={})
src/byprot/models/structok/structok_lfq.py:161
↓ 3 callersFunctiondetailed_search_standard
analysis/TMalign.cpp:2156
↓ 3 callersFunctiondot
analysis/TMalign.cpp:882
↓ 3 callersFunctiondot
analysis/TMscore.cpp:3030
↓ 3 callersFunctionexists
(o)
src/byprot/models/structok/structok_lfq.py:28
↓ 3 callersMethodfeed_forward_chunk
(self, attention_output)
src/byprot/models/dplm/modules/dplm_adapter.py:276
↓ 3 callersMethodfold_fasta
(self, fasta_path, output_dir)
src/byprot/utils/protein/folding_model.py:45
↓ 3 callersFunctionget_net
(cfg)
src/byprot/models/utils.py:51
↓ 3 callersMethodget_non_special_sym_mask
(self, output_tokens, partial_masks=None)
src/byprot/models/dplm/modules/dplm_modeling_esm.py:371
↓ 3 callersMethodget_pdb_from_struct_fasta
(self, struct_fasta_path)
src/byprot/utils/protein/evaluator_dplm2.py:252
↓ 3 callersFunctionget_struct_tokenizer
( model_name_or_path="airkingbd/struct_tokenizer", eval_mode=True )
src/byprot/models/utils.py:387
↓ 3 callersFunctioninstantiate_from_config
(cfg: OmegaConf, group=None, **override_kwargs)
src/byprot/utils/config.py:117
↓ 3 callersMethodis_open
analysis/TMscore.cpp:1919
↓ 3 callersFunctionmake_bond_key
Unique key to lookup bonds.
src/byprot/utils/protein/residue_constants.py:497
↓ 3 callersFunctionmake_bond_key
Unique key to lookup bonds.
src/byprot/datamodules/pdb_dataset/residue_constants.py:497
↓ 3 callersFunctionparse_chain_feats
(chain_feats, scale_factor=1.0)
src/byprot/datamodules/pdb_dataset/utils.py:378
↓ 3 callersFunctionprint_version
analysis/TMalign.cpp:86
↓ 3 callersMethodprocess_chain
(chain_feats: dict, random_crop=False, crop_size=256)
src/byprot/datamodules/pdb_dataset/pdb_datamodule.py:387
↓ 3 callersFunctionsetup_dataloader
( ds: UniRefDataset, max_tokens=6000, bucket_size=1000, max_batch_size=800, num_workers=8,
src/byprot/datamodules/dataset/uniref.py:289
↓ 3 callersFunctionsetup_dataloader
( ds: TokenizedProteinDataset, max_tokens=6000, bucket_size=1000, max_batch_size=100, num_
src/byprot/datamodules/dataset/tokenized_protein.py:610
↓ 3 callersFunctionsetup_dataloader
( ds: UniRefHFDataset, max_tokens=6000, bucket_size=1000, max_batch_size=800, num_workers=
src/byprot/datamodules/dataset/uniref_hf.py:300
↓ 3 callersFunctionsplit
split a long string into vectors by whitespace * line - input string * line_vec - output vector * delimiter - delimiter */
analysis/TMalign.cpp:333
↓ 3 callersFunctiontop_k_top_p_filtering
Filter a distribution of logits using top-k and/or nucleus (top-p) filtering Args: logits: logits distribution shape (vocabulary size)
src/byprot/models/utils.py:347
↓ 3 callersFunctionwrite_prot_to_pdb
( prot_pos: np.ndarray, file_path: str, aatype: np.ndarray = None, overwrite=False, no_ind
src/byprot/utils/protein/utils.py:425
↓ 2 callersFunctionAAmap
analysis/TMalign.cpp:270
↓ 2 callersFunctionAAmap
analysis/TMscore.cpp:2487
↓ 2 callersFunctionCATH
( root=".data", chain_set_jsonl="chain_set.jsonl", chain_set_splits_json="chain_set_splits.json",
src/byprot/datamodules/dataset/cath.py:24
↓ 2 callersFunctionTMscore8_search_standard
analysis/TMalign.cpp:1962
↓ 2 callersMethod__init__
( self, optimizer: Optimizer, lr, warmup_init_lr, model_size: int = 12
src/byprot/utils/lr_scheduler.py:126
↓ 2 callersMethod__init__
(self, dataset: Dataset[T_co], indices: Sequence[int])
src/byprot/datamodules/dataset/tokenized_protein.py:335
↓ 2 callersMethod__init__
( self, *, dataset_cfg, split, is_training, )
src/byprot/datamodules/pdb_dataset/pdb_datamodule.py:190
↓ 2 callersMethod_build_batches
(self)
src/byprot/datamodules/dataset/uniref.py:121
↓ 2 callersFunction_calc_bb_tmscore
(mask, sample_bb_pos, folded_bb_pos, sample_seq)
src/byprot/utils/protein/utils.py:679
↓ 2 callersFunction_calc_ca_rmsd
(mask, sample_ca_pos, folded_ca_pos)
src/byprot/utils/protein/utils.py:649
↓ 2 callersFunction_calculate_bin_centers
(boundaries: torch.Tensor)
src/byprot/models/structok/modules/loss.py:660
↓ 2 callersFunction_chain_end
(atom_index, end_resname, chain_name, residue_index)
src/byprot/datamodules/pdb_dataset/protein.py:154
↓ 2 callersMethod_create_batches
(self)
src/byprot/datamodules/pdb_dataset/pdb_datamodule.py:650
↓ 2 callersFunction_full_mask
(target_tokens, coord_mask, alphabet)
src/byprot/utils/scaffold_utils.py:279
↓ 2 callersMethod_init_residue_constants
(self, float_dtype, device)
src/byprot/models/structok/modules/folding_utils/structure_module.py:801
↓ 2 callersMethod_log
(self, loss_breakdown, batch, outputs, train=True)
src/byprot/tasks/struct_tokenizer/structok.py:226
↓ 2 callersFunction_points
(affine)
src/byprot/models/structok/modules/loss.py:770
↓ 2 callersMethod_reparam_decoding
( self, output_tokens, output_scores, cur_tokens, cur_scores,
src/byprot/models/dplm2/dplm2.py:566
↓ 2 callersMethod_update_cfg
(self, cfg)
src/byprot/models/dplm2/dplm2.py:96
↓ 2 callersFunctionalign_structures
Align structures in a ChemGraph batch to a reference, e.g. for RMSD computation. This uses the sparse formulation of pytorch geometric. If the
src/byprot/datamodules/pdb_dataset/utils.py:446
↓ 2 callersFunctionbatch_encode_sequences
( sequences: T.Sequence[str], residue_index_offset: T.Optional[int] = 512, chain_linker: T.Optiona
src/byprot/models/structok/modules/folding_utils/misc.py:73
↓ 2 callersFunctionbuild_batch
(input_data_aatype, input_data_struct_tokens)
generate_dplm2.py:63
↓ 2 callersFunctioncategorical_lddt
(logits, bins=50)
src/byprot/models/structok/modules/folding_utils/categorical_mixture.py:51
↓ 2 callersFunctioncenter_zero
Move the molecule center to zero for sparse position tensors. Args: pos: [N, 3] batch positions of atoms in the molecule in sparse batch
src/byprot/datamodules/pdb_dataset/utils.py:425
↓ 2 callersFunctionchain_str_to_int
(chain_str: str)
src/byprot/datamodules/pdb_dataset/utils.py:369
↓ 2 callersFunctionchi_angle_atom
Define chi-angle rigid groups via one-hot representations.
src/byprot/utils/protein/residue_constants.py:1058
↓ 2 callersFunctionchi_angle_atom
Define chi-angle rigid groups via one-hot representations.
src/byprot/datamodules/pdb_dataset/residue_constants.py:1058
↓ 2 callersFunctionclean_up_after_approx_TM
analysis/TMscore.cpp:5781
↓ 2 callersMethodcollate_dense_tensors
Takes a list of tensors with the following dimensions: [(d_11, ..., d_1K), (d_21, ..., d_2K), ..., (d_N1,
src/byprot/datamodules/dataset/cath.py:299
↓ 2 callersFunctionconcat_np_features
Performs a nested concatenation of feature dicts. Args: np_dicts: list of dicts with the same structure. Each dict must have
src/byprot/datamodules/pdb_dataset/utils.py:400
↓ 2 callersMethodconstruct_x_t
(self, struct_target, aatype_target)
src/byprot/models/dplm2/dplm2.py:317
↓ 2 callersFunctioncreate_full_prot
( atom37: np.ndarray, atom37_mask: np.ndarray, aatype=None, b_factors=None, )
src/byprot/utils/protein/utils.py:399
↓ 2 callersFunctiondecode
(batch_ids, alphabet, remove_special=False, replace_X=True)
src/byprot/tasks/lm/dplm_invfold.py:605
↓ 2 callersFunctiondefault
(*args)
src/byprot/models/structok/modules/lfq.py:44
↓ 2 callersFunctiondetailed_search_standard
analysis/TMscore.cpp:4585
↓ 2 callersMethoddevice
(self)
src/byprot/utils/protein/folding_model.py:40
↓ 2 callersFunctiondrmsd
(structure_1, structure_2, mask=None)
src/byprot/models/structok/modules/loss.py:53
↓ 2 callersMethodencode
(self, atom_positions, mask, seq_length=None, gvp_feat=None)
src/byprot/models/structok/structok_lfq.py:138
↓ 2 callersFunctionexists
(o)
src/byprot/tasks/struct_tokenizer/structok.py:36
↓ 2 callersFunctionfile2chainlist
read list of entries from 'name' to 'chain_list'. * dir_opt is the folder name (prefix). * suffix_opt is the file name extension (suffix_opt). *
analysis/TMalign.cpp:953
↓ 2 callersFunctionfile2chainlist
read list of entries from 'name' to 'chain_list'. * dir_opt is the folder name (prefix). * suffix_opt is the file name extension (suffix_opt). *
analysis/TMscore.cpp:3056
↓ 2 callersFunctionfind_max_frag
analysis/TMalign.cpp:2678
↓ 2 callersFunctionfind_structural_violations
Computes several checks for structural violations.
src/byprot/models/structok/modules/loss.py:1210
↓ 2 callersMethodfrom_lists
Args: coords_list: list of length batch_size, each item is a list of floats in shape L x 3 x 3 to describe a backbone
src/byprot/datamodules/dataset/cath.py:269
↓ 2 callersMethodfrom_pretrained
( cls, net_name, cfg_override={}, net_override={}, from_huggingface=True )
src/byprot/models/dplm2/dplm2.py:119
↓ 2 callersFunctionget_PDB_lines
analysis/TMalign.cpp:397
↓ 2 callersFunctionget_PDB_lines
analysis/TMscore.cpp:2584
↓ 2 callersFunctionget_intervals
Given a list (Tensor) of non-masked residues get new start and end index for motif placed in scaffold.
src/byprot/utils/scaffold_utils.py:192
↓ 2 callersFunctionget_logger
Initializes multi-GPU-friendly python command line logger.
src/byprot/utils/config.py:14
↓ 2 callersMethodget_mixup_xt
(self, input_ids, model_pred, non_special_sym_mask=None)
src/byprot/models/dplm2/dplm2.py:297
↓ 2 callersFunctionget_motif_dplm
(pdb, ori_pdb)
src/byprot/utils/scaffold_utils.py:212
↓ 2 callersFunctionget_motif_dplm2
(pdb_name, ori_pdb_name, motif_seq, mask_token, spacer_list=None)
src/byprot/utils/scaffold_utils.py:247
↓ 2 callersFunctionget_net_class
(dplm_type)
src/byprot/models/utils.py:42
↓ 2 callersFunctionget_optimizer
(cfg, params)
src/byprot/utils/optim.py:15
↓ 2 callersFunctionget_scheduler
(cfg, optimizer)
src/byprot/utils/lr_scheduler.py:10
↓ 2 callersFunctionimport_modules
(models_dir, namespace, excludes=[])
src/byprot/utils/__init__.py:335
↓ 2 callersMethodindices_to_bits
x: long tensor of indices for constructing codebook, but actually not utilized in all the experiments. returns big endian bits
src/byprot/models/structok/modules/lfq.py:205
↓ 2 callersMethodinitialize_output_tokens
( self, input_tokens, partial_masks=None, **kwargs )
src/byprot/models/dplm2/dplm2.py:459
↓ 2 callersMethodinject_noise
(self, tokens)
src/byprot/tasks/lm/mlm.py:109
↓ 2 callersFunctionlddt
( all_atom_pred_pos: torch.Tensor, all_atom_positions: torch.Tensor, all_atom_mask: torch.Tensor,
src/byprot/models/structok/modules/loss.py:481
↓ 2 callersFunctionlddt_ca
( all_atom_pred_pos: torch.Tensor, all_atom_positions: torch.Tensor, all_atom_mask: torch.Tensor,
src/byprot/models/structok/modules/loss.py:537
↓ 2 callersFunctionload_pdb_by_name
(pdb_name, metadata_df)
src/byprot/utils/protein/evaluator_dplm2.py:60
↓ 2 callersMethodlog_prob
(self, true)
src/byprot/models/structok/modules/folding_utils/categorical_mixture.py:25
↓ 2 callersFunctionmake_sec
secondary stucture assignment for protein: * 1->coil, 2->helix, 3->turn, 4->strand */
analysis/TMalign.cpp:2466
↓ 2 callersFunctionoutput_to_pdb
Returns the pbd (file) string from the model given the model output.
src/byprot/models/structok/modules/folding_utils/misc.py:111
↓ 2 callersFunctionpad
Right pads dimension of numpy array. Args: x: numpy like array to pad. max_len: desired length after padding pad_idx: dim
src/byprot/datamodules/pdb_dataset/utils.py:119
↓ 2 callersFunctionparse
(s)
analysis/cal_plddt_dir.py:73
↓ 2 callersFunctionprep_d
(structure)
src/byprot/models/structok/modules/loss.py:54
↓ 2 callersFunctionprint_help
analysis/TMalign.cpp:176
↓ 2 callersFunctionprint_help
analysis/TMscore.cpp:159
↓ 2 callersFunctionprint_version
analysis/TMscore.cpp:72
↓ 2 callersMethodq_sample
(self, x_0, t, type_ids, maskable_mask)
src/byprot/models/dplm2/dplm2.py:206
↓ 2 callersFunctionread_PDB
analysis/TMalign.cpp:854
↓ 2 callersFunctionread_PDB
analysis/TMscore.cpp:3002
← previousnext →101–200 of 1,008, ranked by callers