DeepChem | PyTorch中用自定义层实现DeepChem的GraphConvLayer

PyTorch中用自定义层实现DeepChem的GraphConvLayer

环境

• DeepChem 2.4
• PyTorch 1.7.0
• Python3.7.9

PyTorch中用自定义层实现DeepChem的GraphConvLayer

导入库

  

   

    

     

    
    

     

      import torch
     
    
   

    

     

    
    

     

      from torch.utils import data
     
    
   

    

     

    
    

     

      from deepchem.feat.graph_features import ConvMolFeaturizer
     
    
   

    

     

    
    

     

      from deepchem.feat.mol_graphs import ConvMol
     
    
   

    

     

    
    

     

      import torch.nn as nn
     
    
   

    

     

    
    

     

      import numpy as np
     
    
  
 

PyTorch中定义GraphConvLayer

  

   

    

     

    
    

     

      class GraphConv(nn.Module):
     
    
   

    

     

    
    

     
 
     
    
   

    

     

    
    

     
 def __init__(self,
     
    
   

    

     

    
    

     

       in_channel,
     
    
   

    

     

    
    

     

       out_channel,
     
    
   

    

     

    
    

     

       min_deg=0,
     
    
   

    

     

    
    

     

       max_deg=10,
     
    
   

    

     

    
    

     

       activation=lambda x: x
     
    
   

    

     

    
    

     

       ):
     
    
   

    

     

    
    

     
 
     
    
   

    

     

    
    

     

       super().__ini
     
    
  
 

文章来源: drugai.blog.csdn.net，作者：DrugAI，版权归原作者所有，如需转载，请联系作者。

原文链接：drugai.blog.csdn.net/article/details/112250343