DeepChem | 基于图卷积预测分子的溶解度

如何使用DeepChem库将图卷积用于类似问题的回归分析。

  

   

    

     

    
    

     

      from deepchem.models.tensorgraph.layers import GraphPool, GraphGather
     
    
   

    

     

    
    

     

      from deepchem.models.tensorgraph.layers import Dense, L2Loss, WeightedError, Stack
     
    
   

    

     

    
    

     

      from deepchem.models.tensorgraph.layers import Label, Weights
     
    
   

    

     

    
    

     

      import numpy as np
     
    
   

    

     

    
    

     

      import tensorflow as tf
     
    
   

    

     

    
    

     

      import os
     
    
  
 

  

   

    

     

    
    

     

      num_epochs = 50
     
    
   

    

     

    
    

     

      batch_size = 200
     
    
   

    

     

    
    

     

      pad_batches = True
     
    
   

    

     

    
    

     
 
     
    
   

    

     

    
    

     

      tg = TensorGraph(batch_size=batch_size,learning_rate=0.0005,use_queue=False)
     
    
   

    

     

    
    

     

      prediction_tasks = ['cLogP','cLogS']
     
    
   

    

     

    
    

     
 
     
    
   

    

     

    
    

     

      def read_data(input_file_path):
     
    
   

    

     

    
    

     

       featurizer = dc.feat.ConvMolFeaturizer()
     
    
   

    

     

    
    

     

       loader = dc.data.CSVLoader(tasks=prediction_tasks, smiles_field="Smiles", featurizer=featurizer)
     
    
   

    

     

    
    

     

       dataset = loader.featurize(input_file_path, shard_size=8192)
     
    
   

    

     

    
    

     
 # Initialize transformers
     
    
   

    

     

    
    

     

       transformer = dc.trans.NormalizationTransformer(transform_w&
     
    
  
 

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/106069191