egnn pytorch
0.2.8
** Un bug a été découvert avec la sélection des voisins en présence de masquage. Si vous avez effectué des expériences antérieures à la version 0.1.12 comportant un masquage, veuillez les réexécuter. **
Implémentation de réseaux de neurones à graphes équivariants E(n), dans Pytorch. Peut éventuellement être utilisé pour la réplication Alphafold2. Cette technique a opté pour des fonctionnalités invariantes simples et a fini par battre toutes les méthodes précédentes (y compris SE3 Transformer et Lie Conv) en termes de précision et de performances. SOTA dans les modèles de systèmes dynamiques, les tâches de prédiction de l'activité moléculaire, etc.
$ pip install egnn-pytorch import torch
from egnn_pytorch import EGNN
layer1 = EGNN ( dim = 512 )
layer2 = EGNN ( dim = 512 )
feats = torch . randn ( 1 , 16 , 512 )
coors = torch . randn ( 1 , 16 , 3 )
feats , coors = layer1 ( feats , coors )
feats , coors = layer2 ( feats , coors ) # (1, 16, 512), (1, 16, 3)Avec bords
import torch
from egnn_pytorch import EGNN
layer1 = EGNN ( dim = 512 , edge_dim = 4 )
layer2 = EGNN ( dim = 512 , edge_dim = 4 )
feats = torch . randn ( 1 , 16 , 512 )
coors = torch . randn ( 1 , 16 , 3 )
edges = torch . randn ( 1 , 16 , 16 , 4 )
feats , coors = layer1 ( feats , coors , edges )
feats , coors = layer2 ( feats , coors , edges ) # (1, 16, 512), (1, 16, 3)Un réseau EGNN complet
import torch
from egnn_pytorch import EGNN_Network
net = EGNN_Network (
num_tokens = 21 ,
num_positions = 1024 , # unless what you are passing in is an unordered set, set this to the maximum sequence length
dim = 32 ,
depth = 3 ,
num_nearest_neighbors = 8 ,
coor_weights_clamp_value = 2. # absolute clamped value for the coordinate weights, needed if you increase the num neareest neighbors
)
feats = torch . randint ( 0 , 21 , ( 1 , 1024 )) # (1, 1024)
coors = torch . randn ( 1 , 1024 , 3 ) # (1, 1024, 3)
mask = torch . ones_like ( feats ). bool () # (1, 1024)
feats_out , coors_out = net ( feats , coors , mask = mask ) # (1, 1024, 32), (1, 1024, 3)Ne vous occupez que des voisins clairsemés, donnés au réseau comme matrice de contiguïté.
import torch
from egnn_pytorch import EGNN_Network
net = EGNN_Network (
num_tokens = 21 ,
dim = 32 ,
depth = 3 ,
only_sparse_neighbors = True
)
feats = torch . randint ( 0 , 21 , ( 1 , 1024 ))
coors = torch . randn ( 1 , 1024 , 3 )
mask = torch . ones_like ( feats ). bool ()
# naive adjacency matrix
# assuming the sequence is connected as a chain, with at most 2 neighbors - (1024, 1024)
i = torch . arange ( 1024 )
adj_mat = ( i [:, None ] >= ( i [ None , :] - 1 )) & ( i [:, None ] <= ( i [ None , :] + 1 ))
feats_out , coors_out = net ( feats , coors , mask = mask , adj_mat = adj_mat ) # (1, 1024, 32), (1, 1024, 3)Vous pouvez également demander au réseau de déterminer automatiquement les voisins d'ordre N et de transmettre une intégration de contiguïté (en fonction de l'ordre) à utiliser comme bordure, avec deux arguments de mot-clé supplémentaires.
import torch
from egnn_pytorch import EGNN_Network
net = EGNN_Network (
num_tokens = 21 ,
dim = 32 ,
depth = 3 ,
num_adj_degrees = 3 , # fetch up to 3rd degree neighbors
adj_dim = 8 , # pass an adjacency degree embedding to the EGNN layer, to be used in the edge MLP
only_sparse_neighbors = True
)
feats = torch . randint ( 0 , 21 , ( 1 , 1024 ))
coors = torch . randn ( 1 , 1024 , 3 )
mask = torch . ones_like ( feats ). bool ()
# naive adjacency matrix
# assuming the sequence is connected as a chain, with at most 2 neighbors - (1024, 1024)
i = torch . arange ( 1024 )
adj_mat = ( i [:, None ] >= ( i [ None , :] - 1 )) & ( i [:, None ] <= ( i [ None , :] + 1 ))
feats_out , coors_out = net ( feats , coors , mask = mask , adj_mat = adj_mat ) # (1, 1024, 32), (1, 1024, 3) Si vous devez passer par des bords continus
import torch
from egnn_pytorch import EGNN_Network
net = EGNN_Network (
num_tokens = 21 ,
dim = 32 ,
depth = 3 ,
edge_dim = 4 ,
num_nearest_neighbors = 3
)
feats = torch . randint ( 0 , 21 , ( 1 , 1024 ))
coors = torch . randn ( 1 , 1024 , 3 )
mask = torch . ones_like ( feats ). bool ()
continuous_edges = torch . randn ( 1 , 1024 , 1024 , 4 )
# naive adjacency matrix
# assuming the sequence is connected as a chain, with at most 2 neighbors - (1024, 1024)
i = torch . arange ( 1024 )
adj_mat = ( i [:, None ] >= ( i [ None , :] - 1 )) & ( i [:, None ] <= ( i [ None , :] + 1 ))
feats_out , coors_out = net ( feats , coors , edges = continuous_edges , mask = mask , adj_mat = adj_mat ) # (1, 1024, 32), (1, 1024, 3) L'architecture initiale d'EGNN souffrait d'instabilité lorsqu'il y avait un nombre élevé de voisins. Heureusement, il semble y avoir deux solutions qui atténuent largement ce problème.
import torch
from egnn_pytorch import EGNN_Network
net = EGNN_Network (
num_tokens = 21 ,
dim = 32 ,
depth = 3 ,
num_nearest_neighbors = 32 ,
norm_coors = True , # normalize the relative coordinates
coor_weights_clamp_value = 2. # absolute clamped value for the coordinate weights, needed if you increase the num neareest neighbors
)
feats = torch . randint ( 0 , 21 , ( 1 , 1024 )) # (1, 1024)
coors = torch . randn ( 1 , 1024 , 3 ) # (1, 1024, 3)
mask = torch . ones_like ( feats ). bool () # (1, 1024)
feats_out , coors_out = net ( feats , coors , mask = mask ) # (1, 1024, 32), (1, 1024, 3) import torch
from egnn_pytorch import EGNN
model = EGNN (
dim = dim , # input dimension
edge_dim = 0 , # dimension of the edges, if exists, should be > 0
m_dim = 16 , # hidden model dimension
fourier_features = 0 , # number of fourier features for encoding of relative distance - defaults to none as in paper
num_nearest_neighbors = 0 , # cap the number of neighbors doing message passing by relative distance
dropout = 0.0 , # dropout
norm_feats = False , # whether to layernorm the features
norm_coors = False , # whether to normalize the coordinates, using a strategy from the SE(3) Transformers paper
update_feats = True , # whether to update features - you can build a layer that only updates one or the other
update_coors = True , # whether ot update coordinates
only_sparse_neighbors = False , # using this would only allow message passing along adjacent neighbors, using the adjacency matrix passed in
valid_radius = float ( 'inf' ), # the valid radius each node considers for message passing
m_pool_method = 'sum' , # whether to mean or sum pool for output node representation
soft_edges = False , # extra GLU on the edges, purportedly helps stabilize the network in updated version of the paper
coor_weights_clamp_value = None # clamping of the coordinate updates, again, for stabilization purposes
) Pour exécuter l'exemple de débruitage du squelette protéique, installez d'abord sidechainnet
$ pip install sidechainnetAlors
$ python denoise_sparse.pyAssurez-vous que pytorch géométrique est installé localement
$ python setup.py test @misc { satorras2021en ,
title = { E(n) Equivariant Graph Neural Networks } ,
author = { Victor Garcia Satorras and Emiel Hoogeboom and Max Welling } ,
year = { 2021 } ,
eprint = { 2102.09844 } ,
archivePrefix = { arXiv } ,
primaryClass = { cs.LG }
}
