Contents Menu Expand Light mode Dark mode Auto light/dark, in light mode Auto light/dark, in dark mode Skip to content
IITP PPI 0.0.2 documentation
IITP PPI 0.0.2 documentation
  • API
    • API/Backbone
      • deepfold
        • deepfold.common
          • deepfold.common.protein
          • deepfold.common.residue_constants
        • deepfold.config
        • deepfold.data
          • deepfold.data.errors
          • deepfold.data.modules
            • deepfold.data.modules.dataloaders
            • deepfold.data.modules.mmcif_dataset
            • deepfold.data.modules.samplers
            • deepfold.data.modules.single_dataset
          • deepfold.data.monomer
          • deepfold.data.multimer
            • deepfold.data.multimer.input_features
            • deepfold.data.multimer.msa_pairing
          • deepfold.data.pdbio
          • deepfold.data.pdbx_parsing
          • deepfold.data.search
            • deepfold.data.search.crfalign
            • deepfold.data.search.input_features
            • deepfold.data.search.mmcif
            • deepfold.data.search.msa_identifiers
            • deepfold.data.search.parsers
            • deepfold.data.search.templates
          • deepfold.data.tools
            • deepfold.data.tools.hhblits
            • deepfold.data.tools.hhsearch
            • deepfold.data.tools.hmmbuild
            • deepfold.data.tools.hmmsearch
            • deepfold.data.tools.jackhmmer
            • deepfold.data.tools.kalign
            • deepfold.data.tools.utils
        • deepfold.distributed
          • deepfold.distributed.model_parallel
          • deepfold.distributed.utils
        • deepfold.eval
          • deepfold.eval.cell_lists
          • deepfold.eval.contact
          • deepfold.eval.distogram
          • deepfold.eval.msa
          • deepfold.eval.plot
          • deepfold.eval.pseudo_3d
        • deepfold.loss
        • deepfold.losses
          • deepfold.losses.auxillary
          • deepfold.losses.confidence
          • deepfold.losses.geometry
          • deepfold.losses.masked_msa
          • deepfold.losses.procrustes
          • deepfold.losses.utils
          • deepfold.losses.violation
        • deepfold.modules
          • deepfold.modules.alphafold
          • deepfold.modules.angle_resnet
          • deepfold.modules.attention
          • deepfold.modules.auxiliary_heads
          • deepfold.modules.backbone_update
          • deepfold.modules.dropout
          • deepfold.modules.evoformer_block
          • deepfold.modules.evoformer_block_pair_core
          • deepfold.modules.evoformer_stack
          • deepfold.modules.extra_msa_block
          • deepfold.modules.extra_msa_embedder
          • deepfold.modules.extra_msa_stack
          • deepfold.modules.global_attention
          • deepfold.modules.inductor
          • deepfold.modules.input_embedder
          • deepfold.modules.invariant_point_attention
          • deepfold.modules.layer_norm
          • deepfold.modules.linear
          • deepfold.modules.msa_column_attention
          • deepfold.modules.msa_column_global_attention
          • deepfold.modules.msa_row_attention_with_pair_bias
          • deepfold.modules.msa_transition
          • deepfold.modules.outer_product_mean
          • deepfold.modules.pair_transition
          • deepfold.modules.recycling_embedder
          • deepfold.modules.single_transition
          • deepfold.modules.structure_module
          • deepfold.modules.template_angle_embedder
          • deepfold.modules.template_pair_block
          • deepfold.modules.template_pair_embedder
          • deepfold.modules.template_pair_stack
          • deepfold.modules.template_pointwise_attention
          • deepfold.modules.template_projection
          • deepfold.modules.triangular_attention
          • deepfold.modules.triangular_multiplicative_update
          • deepfold.modules.tweaks
            • deepfold.modules.tweaks.evo_attn
        • deepfold.ops
          • deepfold.ops.evoformer_attention
        • deepfold.package_info
        • deepfold.relax
          • deepfold.relax.amber_minimize
          • deepfold.relax.cleanup
          • deepfold.relax.relax
          • deepfold.relax.utils
        • deepfold.runner
          • deepfold.runner.config_utils
          • deepfold.runner.parser
          • deepfold.runner.utils
        • deepfold.testing
          • deepfold.testing.common
          • deepfold.testing.utils
        • deepfold.train
          • deepfold.train.gradient_clipping
          • deepfold.train.lr_scheduler
          • deepfold.train.validation_metrics
        • deepfold.utils
          • deepfold.utils.checkpoint_utils
          • deepfold.utils.chunk_utils
          • deepfold.utils.crop_utils
          • deepfold.utils.datetime_utils
          • deepfold.utils.dist_utils
          • deepfold.utils.feats_utils
          • deepfold.utils.file_utils
          • deepfold.utils.geometry
            • deepfold.utils.geometry.quat_rigid
            • deepfold.utils.geometry.rigid_matrix_vector
            • deepfold.utils.geometry.rotation_matrix
            • deepfold.utils.geometry.utils
            • deepfold.utils.geometry.vector
          • deepfold.utils.hash_utils
          • deepfold.utils.import_utils
          • deepfold.utils.iter_utils
          • deepfold.utils.log_utils
          • deepfold.utils.precision
          • deepfold.utils.random
          • deepfold.utils.rigid_utils
          • deepfold.utils.script_utils
          • deepfold.utils.swa_utils
          • deepfold.utils.tensor_utils
    • API/Tool
      • bis_protein_structure
        • bis_protein_structure.CROSSLINK
        • bis_protein_structure.DATAGEN
        • bis_protein_structure.DISULFIDE
        • bis_protein_structure.MMEVAL
        • bis_protein_structure.SOLVENTACC
        • bis_protein_structure.TORSION2
        • bis_protein_structure.residue_constants
        • bis_protein_structure.split_pdb
      • gd_dl
        • gd_dl.data
          • gd_dl.data.ML_inference
          • gd_dl.data.ML_rerank
          • gd_dl.data.all_ML_rerank
          • gd_dl.data.single_ML_rerank
        • gd_dl.lib_pdb_mol2
        • gd_dl.models
        • gd_dl.path_setting
        • gd_dl.rerank_model
        • gd_dl.utils
      • promptbind
        • promptbind.data
          • promptbind.data.data
        • promptbind.models
          • promptbind.models.att_model
          • promptbind.models.cross_att
          • promptbind.models.egnn
          • promptbind.models.model
          • promptbind.models.model_utils
        • promptbind.test_promptbind
        • promptbind.utils
          • promptbind.utils.fabind_inference_dataset
          • promptbind.utils.feature_utils
          • promptbind.utils.generation_utils
          • promptbind.utils.inference_mol_utils
          • promptbind.utils.inference_pdb_utils
          • promptbind.utils.logging_utils
          • promptbind.utils.metrics
          • promptbind.utils.metrics_to_tsb
          • promptbind.utils.post_optim_utils
          • promptbind.utils.utils
        • promptbind.visualize_prompt_components
      • diffalign
        • diffalign.models
          • diffalign.models.common
          • diffalign.models.encoder
            • diffalign.models.encoder.edge
            • diffalign.models.encoder.egnn
          • diffalign.models.epsnet
            • diffalign.models.epsnet.diffusion
        • diffalign.utils
          • diffalign.utils.chem
          • diffalign.utils.common
          • diffalign.utils.dataloader
          • diffalign.utils.datasets
          • diffalign.utils.evaluation
            • diffalign.utils.evaluation.covmat
          • diffalign.utils.misc
          • diffalign.utils.transforms
          • diffalign.utils.visualize
Back to top
View this page

deepfold.modules.invariant_point_attentionΒΆ

Functions

ipa_point_weights_init_(weights_data)

Classes

InvariantPointAttention(c_s, c_z, c_hidden, ...)

Invariant Point Attention (IPA) module.

InvariantPointAttentionMultimer(c_s, c_z, ...)

PointProjection(c_hidden, num_points, ...[, ...])

Next
deepfold.modules.layer_norm
Previous
deepfold.modules.input_embedder
Copyright © 2024, Paek et al.
Made with Sphinx and @pradyunsg's Furo