Unverified Commit af54fae1 authored by Bharath Ramsundar's avatar Bharath Ramsundar Committed by GitHub
Browse files

Merge pull request #1624 from VIGS25/split-transform-order 

Swap Split-Transform order - II
parents 42fbae1d b17c7467

deepchem/molnet/load_function/nci_datasets.py

+18 −10
Original line number Diff line number Diff line
@@ -62,16 +62,14 @@ def load_nci(featurizer='ECFP', shard_size=1000, split='random', reload=True): 


  dataset = loader.featurize(dataset_file, shard_size=shard_size)



  # Initialize transformers

  logger.info("About to transform data")

  if split == None:

    logger.info("Split is None, about to transform data")

    transformers = [

        deepchem.trans.NormalizationTransformer(

            transform_y=True, dataset=dataset)

    ]

    for transformer in transformers:

      dataset = transformer.transform(dataset)



  if split == None:

    return all_nci_tasks, (dataset, None, None), transformers



  splitters = {
@@ -80,9 +78,19 @@ def load_nci(featurizer='ECFP', shard_size=1000, split='random', reload=True): 
      'scaffold': deepchem.splits.ScaffoldSplitter()

  }

  splitter = splitters[split]

  logger.info("Performing new split.")

  logger.info("About to split data with {} splitter.".format(splitter))

  train, valid, test = splitter.train_valid_test_split(dataset)



  transformers = [

      deepchem.trans.NormalizationTransformer(transform_y=True, dataset=train)

  ]



  logger.info("About to transform dataset.")

  for transformer in transformers:

    train = transformer.transform(train)

    valid = transformer.transform(valid)

    test = transformer.transform(test)



  if reload:

    deepchem.utils.save.save_dataset_to_disk(save_dir, train, valid, test,

                                             transformers)

deepchem/molnet/load_function/pcba_datasets.py

+19 −9
Original line number Diff line number Diff line
@@ -82,16 +82,16 @@ def load_pcba_dataset(featurizer='ECFP', 
      tasks=PCBA_tasks, smiles_field="smiles", featurizer=featurizer)



  dataset = loader.featurize(dataset_file)

  # Initialize transformers



  if split == None:

    transformers = [

        deepchem.trans.BalancingTransformer(transform_w=True, dataset=dataset)

    ]



  logger.info("About to transform data")

    logger.info("Split is None, about to transform data")

    for transformer in transformers:

      dataset = transformer.transform(dataset)



  if split == None:

    return PCBA_tasks, (dataset, None, None), transformers



  splitters = {
@@ -100,9 +100,19 @@ def load_pcba_dataset(featurizer='ECFP', 
      'scaffold': deepchem.splits.ScaffoldSplitter()

  }

  splitter = splitters[split]

  logger.info("Performing new split.")

  logger.info("About to split dataset using {} splitter.".format(split))

  train, valid, test = splitter.train_valid_test_split(dataset)



  transformers = [

      deepchem.trans.BalancingTransformer(transform_w=True, dataset=train)

  ]



  logger.info("About to transform dataset.")

  for transformer in transformers:

    train = transformer.transform(train)

    valid = transformer.transform(valid)

    test = transformer.transform(test)



  if reload:

    deepchem.utils.save.save_dataset_to_disk(save_dir, train, valid, test,

                                             transformers)

deepchem/molnet/load_function/pdbbind_datasets.py

+20 −7
Original line number Diff line number Diff line
@@ -111,16 +111,17 @@ def load_pdbbind_grid(split="random", 
    loader = deepchem.data.CSVLoader(

        tasks=tasks, smiles_field="smiles", featurizer=featurizer)

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

    df = pd.read_csv(dataset_file)



    if split == None:

      transformers = [

          deepchem.trans.NormalizationTransformer(

              transform_y=True, dataset=dataset)

      ]



      logger.info("Split is None, about to transform data.")

      for transformer in transformers:

        dataset = transformer.transform(dataset)

    df = pd.read_csv(dataset_file)



    if split == None:

      return tasks, (dataset, None, None), transformers



    splitters = {
@@ -130,8 +131,20 @@ def load_pdbbind_grid(split="random", 
        'time': deepchem.splits.TimeSplitterPDBbind(np.array(df['id']))

    }

    splitter = splitters[split]

    logger.info("About to split dataset with {} splitter.".format(split))

    train, valid, test = splitter.train_valid_test_split(dataset)



    transformers = [

        deepchem.trans.NormalizationTransformer(

            transform_y=True, dataset=train)

    ]



    logger.info("About to transform dataset.")

    for transformer in transformers:

      train = transformer.transform(train)

      valid = transformer.transform(valid)

      test = transformer.transform(test)



    if reload:

      deepchem.utils.save.save_dataset_to_disk(save_dir, train, valid, test,

                                               transformers)

deepchem/molnet/load_function/ppb_datasets.py

+19 −9
Original line number Diff line number Diff line
@@ -47,17 +47,16 @@ def load_ppb(featurizer='ECFP', split='index', reload=True): 
      tasks=PPB_tasks, smiles_field="smiles", featurizer=featurizer)

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



  # Initialize transformers

  if split == None:

    transformers = [

        deepchem.trans.NormalizationTransformer(

            transform_y=True, dataset=dataset)

    ]



  logger.info("About to transform data")

    logger.info("Split is None, about to transform data")

    for transformer in transformers:

      dataset = transformer.transform(dataset)



  if split == None:

    return PPB_tasks, (dataset, None, None), transformers



  splitters = {
@@ -66,8 +65,19 @@ def load_ppb(featurizer='ECFP', split='index', reload=True): 
      'scaffold': deepchem.splits.ScaffoldSplitter()

  }

  splitter = splitters[split]

  logger.info("About to split dataset with {} splitter.".format(split))

  train, valid, test = splitter.train_valid_test_split(dataset)



  transformers = [

      deepchem.trans.NormalizationTransformer(transform_y=True, dataset=train)

  ]



  logger.info("About to transform dataset.")

  for transformer in transformers:

    train = transformer.transform(train)

    valid = transformer.transform(valid)

    test = transformer.transform(test)



  if reload:

    deepchem.utils.save.save_dataset_to_disk(save_dir, train, valid, test,

                                             transformers)

deepchem/molnet/load_function/sampl_datasets.py

+21 −9
Original line number Diff line number Diff line
@@ -51,17 +51,16 @@ def load_sampl(featurizer='ECFP', split='index', reload=True, move_mean=True): 
      tasks=SAMPL_tasks, smiles_field="smiles", featurizer=featurizer)

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



  # Initialize transformers

  if split == None:

    transformers = [

        deepchem.trans.NormalizationTransformer(

            transform_y=True, dataset=dataset, move_mean=move_mean)

    ]



  logger.info("About to transform data")

    logger.info("Split is None, about to transform data")

    for transformer in transformers:

      dataset = transformer.transform(dataset)



  if split == None:

    return SAMPL_tasks, (dataset, None, None), transformers



  splitters = {
@@ -70,7 +69,20 @@ def load_sampl(featurizer='ECFP', split='index', reload=True, move_mean=True): 
      'scaffold': deepchem.splits.ScaffoldSplitter()

  }

  splitter = splitters[split]

  logger.info("About to split dataset with {} splitter.".format(split))

  train, valid, test = splitter.train_valid_test_split(dataset)



  transformers = [

      deepchem.trans.NormalizationTransformer(

          transform_y=True, dataset=train, move_mean=move_mean)

  ]



  logger.info("About to transform dataset.")

  for transformer in transformers:

    train = transformer.transform(train)

    valid = transformer.transform(valid)

    test = transformer.transform(test)



  if reload:

    deepchem.utils.save.save_dataset_to_disk(save_dir, train, valid, test,

                                             transformers)

CRA Git | Maintained and supported by SUSTech CRA and CCSE