Types & classes — github.com/apple/ml-simplefold

CodeHub MCPcopy Create free account

hub / github.com/apple/ml-simplefold / types & classes

Types & classes102 in github.com/apple/ml-simplefold

⨍ Functions 396◇ Types & classes 102

↓ 5 callersClassRecord

Record datatype.

src/simplefold/boltz_data_pipeline/types.py:382

↓ 5 callersClassStructure

Structure datatype.

src/simplefold/boltz_data_pipeline/types.py:148

↓ 4 callersClassBoltzTokenizer

Tokenize an input structure for training.

src/simplefold/boltz_data_pipeline/tokenize/boltz_protein.py:39

↓ 4 callersClassManifest

Manifest datatype.

src/simplefold/boltz_data_pipeline/types.py:407

↓ 4 callersClassRankedLogger

A multi-GPU-friendly python command line logger.

src/simplefold/utils/pylogger.py:12

↓ 3 callersClassBoltzFeaturizer

Boltz featurizer.

src/simplefold/boltz_data_pipeline/feature/featurizer.py:382

↓ 3 callersClassChainInfo

ChainInfo datatype.

src/simplefold/boltz_data_pipeline/types.py:353

↓ 3 callersClassInput

Input datatype.

src/simplefold/boltz_data_pipeline/types.py:452

↓ 3 callersClassParsedAtom

A parsed atom object.

src/simplefold/utils/mmcif_utils.py:38

↓ 3 callersClassParsedAtom

A parsed atom object.

src/simplefold/boltz_data_pipeline/parse/schema.py:41

↓ 3 callersClassParsedChain

A parsed chain object.

src/simplefold/boltz_data_pipeline/parse/schema.py:79

↓ 3 callersClassParsedResidue

A parsed residue object.

src/simplefold/utils/mmcif_utils.py:60

↓ 3 callersClassParsedResidue

A parsed residue object.

src/simplefold/boltz_data_pipeline/parse/schema.py:63

↓ 2 callersClassClash

A clash between two chains.

src/simplefold/boltz_data_pipeline/filter/static/polymer.py:173

↓ 2 callersClassDataset

src/simplefold/utils/datamodule_utils.py:52

↓ 2 callersClassLinearPath

Linear flow process: x0: noise, x1: data In inference, we sample data from 0 -> 1

src/simplefold/model/flow.py:81

↓ 2 callersClassMSA

src/simplefold/boltz_data_pipeline/types.py:326

↓ 2 callersClassProteinDataProcessor

src/simplefold/processor/protein_processor.py:22

↓ 2 callersClassRMSNorm

src/simplefold/model/torch/layers.py:271

↓ 2 callersClassSimpleFoldTrainingDataset

Base iterable dataset.

src/simplefold/datasets/train_datamodule.py:53

↓ 2 callersClassStructureInfo

StructureInfo datatype.

src/simplefold/boltz_data_pipeline/types.py:340

↓ 2 callersClassSwiGLUFeedForward

src/simplefold/model/torch/layers.py:118

↓ 2 callersClassSwiGLUFeedForward

src/simplefold/model/mlx/layers.py:143

↓ 2 callersClassTarget

Target datatype.

src/simplefold/boltz_data_pipeline/types.py:398

↓ 2 callersClassTokenData

TokenData datatype.

src/simplefold/boltz_data_pipeline/tokenize/boltz_protein.py:19

↓ 1 callersClassClashingChainsFilter

A filter that filters clashing chains. Clashing chains are defined as those with >30% of atoms within 1.7 Å of an atom in another chain. If t

src/simplefold/boltz_data_pipeline/filter/static/polymer.py:182

↓ 1 callersClassConditionEmbedder

Embeds class labels into vector representations. Also handles label dropout for classifier-free guidance.

src/simplefold/model/torch/layers.py:200

↓ 1 callersClassConditionEmbedder

Embeds class labels into vector representations. Also handles label dropout for classifier-free guidance.

src/simplefold/model/mlx/layers.py:212

↓ 1 callersClassConsecutiveCA

Filter proteins with consecutive CA atoms above a threshold.

src/simplefold/boltz_data_pipeline/filter/static/polymer.py:111

↓ 1 callersClassContactPredictionHead

Performs symmetrization, apc, and computes a logistic regression on the output features

src/simplefold/model/mlx/esm_modules.py:145

↓ 1 callersClassExcludedLigands

Filter excluded ligands.

src/simplefold/boltz_data_pipeline/filter/static/ligand.py:150

↓ 1 callersClassFinalLayer

The final layer of DiT.

src/simplefold/model/torch/layers.py:234

↓ 1 callersClassFinalLayer

The final layer of DiT.

src/simplefold/model/mlx/layers.py:246

↓ 1 callersClassInferenceOptions

src/simplefold/boltz_data_pipeline/types.py:376

↓ 1 callersClassMinimumLengthFilter

Filter polymers based on their length. We use the number of resolved residues when considering the minimum, and the sequence length for the m

src/simplefold/boltz_data_pipeline/filter/static/polymer.py:19

↓ 1 callersClassMultiheadAttention

Multi-headed attention. See "Attention Is All You Need" for more details.

src/simplefold/model/mlx/esm_multihead_attention.py:22

↓ 1 callersClassPDB

A raw MMCIF PDB file.

src/simplefold/process_mmcif.py:46

↓ 1 callersClassParsedBond

A parsed bond object.

src/simplefold/utils/mmcif_utils.py:51

↓ 1 callersClassParsedBond

A parsed bond object.

src/simplefold/boltz_data_pipeline/parse/schema.py:54

↓ 1 callersClassParsedChain

A parsed chain object.

src/simplefold/utils/mmcif_utils.py:76

↓ 1 callersClassParsedConnection

A parsed connection object.

src/simplefold/utils/mmcif_utils.py:87

↓ 1 callersClassParsedStructure

A parsed structure object.

src/simplefold/utils/mmcif_utils.py:99

↓ 1 callersClassResource

A shared resource for processing.

src/simplefold/process_mmcif.py:53

↓ 1 callersClassRobertaLMHead

Head for masked language modeling.

src/simplefold/model/mlx/esm_modules.py:126

↓ 1 callersClassRotaryEmbedding

The rotary position embeddings from RoFormer_ (Su et. al). A crucial insight from the method is that the query and keys are transformed b

src/simplefold/model/mlx/esm_rotary_embedding.py:25

↓ 1 callersClassSample

src/simplefold/process_structure.py:24

↓ 1 callersClassSimpleFoldPredictionDataset

Base iterable dataset.

src/simplefold/datasets/test_datamodule.py:27

↓ 1 callersClassTokenized

Tokenized datatype.

src/simplefold/boltz_data_pipeline/types.py:489

↓ 1 callersClassTransformerLayer

src/simplefold/model/mlx/esm_modules.py:62

↓ 1 callersClassUnknownFilter

Filter proteins with all unknown residues.

src/simplefold/boltz_data_pipeline/filter/static/polymer.py:72

↓ 1 callersClass_LocalPLDDT

src/simplefold/boltz_data_pipeline/write/mmcif.py:122

↓ 1 callersClass_MyModel

src/simplefold/boltz_data_pipeline/write/mmcif.py:127

ClassAbsolutePositionEncoding

src/simplefold/model/torch/pos_embed.py:11

ClassAbsolutePositionEncoding

src/simplefold/model/mlx/pos_embed.py:14

ClassAxialRotaryPositionEncoding

src/simplefold/model/torch/pos_embed.py:153

ClassAxialRotaryPositionEncoding

src/simplefold/model/mlx/pos_embed.py:179

base class for flow matching path

src/simplefold/model/flow.py:13

ClassBoltzCropper

Interpolate between contiguous and spatial crops.

src/simplefold/boltz_data_pipeline/crop/boltz.py:135

ClassBoltzWriter

Custom writer for predictions.

src/simplefold/boltz_data_pipeline/write/writer.py:29

ClassConfidenceModule

src/simplefold/model/torch/confidence_module.py:39

ClassConfidenceModule

src/simplefold/model/mlx/confidence_module.py:37

Abstract base class for cropper.

src/simplefold/boltz_data_pipeline/crop/cropper.py:17

ClassDatasetConfig

Dataset configuration.

src/simplefold/utils/datamodule_utils.py:67

ClassDateFilter

A filter that filters complexes based on their date. The date can be the deposition, release, or revision date. If the date is not available,

src/simplefold/boltz_data_pipeline/filter/dynamic/date.py:16

A DiT block with adaptive layer norm zero (adaLN-Zero) conditioning.

src/simplefold/model/torch/blocks.py:12

A DiT block with adaptive layer norm zero (adaLN-Zero) conditioning.

src/simplefold/model/mlx/blocks.py:12

ClassDynamicFilter

Base class for data filters.

src/simplefold/boltz_data_pipeline/filter/dynamic/filter.py:14

A Euler-Maruyama solver for SDEs.

src/simplefold/model/torch/sampler.py:12

A Euler-Maruyama solver for SDEs.

src/simplefold/model/mlx/sampler.py:19

ClassESM1LayerNorm

src/simplefold/model/mlx/esm_modules.py:36

src/simplefold/model/mlx/esm_network.py:26

ClassEfficientSelfAttentionLayer

Started from https://github.com/facebookresearch/dinov2/blob/main/dinov2/layers/attention.py

src/simplefold/model/torch/layers.py:78

ClassEfficientSelfAttentionLayer

Adapted from https://github.com/facebookresearch/dinov2/blob/main/dinov2/layers/attention.py

src/simplefold/model/mlx/layers.py:80

ClassFoldingDiT

src/simplefold/model/torch/architecture.py:15

ClassFoldingDiT

src/simplefold/model/mlx/architecture.py:40

ClassFourierPositionEncoding

src/simplefold/model/torch/pos_embed.py:49

ClassFourierPositionEncoding

src/simplefold/model/mlx/pos_embed.py:52

ClassHomogenTrunk

src/simplefold/model/torch/blocks.py:116

ClassHomogenTrunk

src/simplefold/model/mlx/blocks.py:91

ClassInferenceWrapper

src/simplefold/wrapper.py:199

ClassInterfaceInfo

InterfaceInfo datatype.

src/simplefold/boltz_data_pipeline/types.py:367

ClassJSONSerializable

Serializable datatype.

src/simplefold/boltz_data_pipeline/types.py:54

ClassLinearWarmup

src/simplefold/utils/lr_scheduler.py:11

ClassMaxResiduesFilter

A filter that filters structures based on their size.

src/simplefold/boltz_data_pipeline/filter/dynamic/max_residues.py:13

ClassModelWrapper

src/simplefold/wrapper.py:50

ClassNumpySerializable

Serializable datatype.

src/simplefold/boltz_data_pipeline/types.py:22

src/simplefold/evaluation/analyze_two_state.py:97

ClassResolutionFilter

A filter that filters complexes based on their resolution.

src/simplefold/boltz_data_pipeline/filter/dynamic/resolution.py:13

ClassSelfAttentionLayer

src/simplefold/model/torch/layers.py:22

ClassSelfAttentionLayer

src/simplefold/model/mlx/layers.py:22

ClassSimpleFold

src/simplefold/model/simplefold.py:69

ClassSimpleFoldInferenceDataModule

DataModule for SimpleFold Inference.

src/simplefold/datasets/test_datamodule.py:127

ClassSimpleFoldTrainingDataModule

DataModule for SimpleFold Training.

src/simplefold/datasets/train_datamodule.py:211

ClassSizeFilter

A filter that filters structures based on their size.

src/simplefold/boltz_data_pipeline/filter/dynamic/size.py:13

ClassSliceCropper

Interpolate between contiguous and spatial crops.

src/simplefold/boltz_data_pipeline/crop/slice.py:40

ClassStaticFilter

Base class for structure filters.

src/simplefold/boltz_data_pipeline/filter/static/filter.py:16

ClassSubsetFilter

Filter a data record based on a subset of the data.

src/simplefold/boltz_data_pipeline/filter/dynamic/subset.py:15

ClassTimestepEmbedder

Embeds scalar timesteps into vector representations.

src/simplefold/model/torch/layers.py:150

ClassTimestepEmbedder

Embeds scalar timesteps into vector representations.

src/simplefold/model/mlx/layers.py:162

Tokenize an input structure for training.

src/simplefold/boltz_data_pipeline/tokenize/tokenizer.py:14

next →1–100 of 102, ranked by callers