Unverified Commit 95f43b86 authored by Suzukazole's avatar Suzukazole
Browse files

Removed toggle, added support for 50K

parent 7e936532

deepchem/molnet/load_function/molnet_loader.py

+1 −14
Original line number Diff line number Diff line
@@ -88,8 +88,7 @@ class _MolnetLoader(object): 
               splitter: Union[dc.splits.Splitter, str, None],

               transformer_generators: List[Union[TransformerGenerator, str]],

               tasks: List[str], data_dir: Optional[str],

               save_dir: Optional[str],

               precomputed_splits: Optional[bool], **kwargs):

               save_dir: Optional[str], **kwargs):

    """Construct an object for loading a dataset.



    Parameters
@@ -133,7 +132,6 @@ class _MolnetLoader(object): 
    self.tasks = list(tasks)

    self.data_dir = data_dir

    self.save_dir = save_dir

    self.precomputed_splits = precomputed_splits

    self.args = kwargs



  def load_dataset(
@@ -172,13 +170,6 @@ class _MolnetLoader(object): 
            save_folder)

        if all_dataset is not None:

          return self.tasks, all_dataset, transformers

    

    # Try to load precomputed splits.



    if self.precomputed_splits is True:

      train, valid, test = self.load_precomputed_splits()

      return (train, valid, test)



    # Create the dataset



    logger.info("About to featurize %s dataset." % name)
@@ -218,7 +209,3 @@ class _MolnetLoader(object): 
  def create_dataset(self) -> Dataset:

    """Subclasses must implement this to load the dataset."""

    raise NotImplementedError()
 


  def load_precomputed_splits(self) -> Tuple[Dataset, ...]:

    """Subclasses must implement this to load precomputed train/test/eval splits."""

    raise NotImplementedError()

deepchem/molnet/load_function/uspto_datasets.py

+18 −61
Original line number Diff line number Diff line
@@ -16,13 +16,9 @@ logger = logging.getLogger(__name__) 


DEFAULT_DIR = deepchem.utils.data_utils.get_data_dir()



USPTO_MIT_TRAIN = "https://deepchemdata.s3-us-west-1.amazonaws.com/datasets/USPTO_MIT_train.csv"

USPTO_MIT_TEST = "https://deepchemdata.s3-us-west-1.amazonaws.com/datasets/USPTO_MIT_test.csv"

USPTO_MIT_VALID = "https://deepchemdata.s3-us-west-1.amazonaws.com/datasets/USPTO_MIT_val.csv"



USPTO_STEREO_TRAIN = "https://deepchemdata.s3-us-west-1.amazonaws.com/datasets/USPTO_STEREO_train.csv"

USPTO_STEREO_TEST = "https://deepchemdata.s3-us-west-1.amazonaws.com/datasets/USPTO_STEREO_test.csv"

USPTO_STEREO_VALID = "https://deepchemdata.s3-us-west-1.amazonaws.com/datasets/USPTO_STEREO_val.csv"

USPTO_MIT_URL = "https://deepchemdata.s3.us-west-1.amazonaws.com/datasets/USPTO_MIT.csv"

USPTO_STEREO_URL = "https://deepchemdata.s3.us-west-1.amazonaws.com/datasets/USPTO_STEREO.csv"

USPTO_50K_URL = "https://deepchemdata.s3.us-west-1.amazonaws.com/datasets/USPTO_50K.csv"





class _USPTOLoader(_MolnetLoader):
@@ -34,72 +30,32 @@ class _USPTOLoader(_MolnetLoader): 
    self.name = 'USPTO_' + subset



  def create_dataset(self) -> Dataset:

    #####INCOMPLETE/INCORRECT: I don't think this is the right way to bypass the splitter!

    if self.subset not in ['MIT', 'STEREO']:

      raise ValueError("Valid Subset names are MIT and STEREO.")

    if self.subset not in ['MIT', 'STEREO', '50K']:

      raise ValueError("Valid Subset names are MIT, STEREO and 50K.")



    if self.subset == 'MIT':

      train_file = os.path.join(self.data_dir, USPTO_MIT_TRAIN)

      test_file = os.path.join(self.data_dir, USPTO_MIT_TEST)

      valid_file = os.path.join(self.data_dir, USPTO_MIT_VALID)



      if not os.path.exists(train_file):

        logger.info("Downloading training file...")

        dc.utils.data_utils.download_url(

            url=USPTO_MIT_TRAIN, dest_dir=self.data_dir)

        logger.info("Training file download complete.")



        logger.info("Downloading test file...")

        dc.utils.data_utils.download_url(

            url=USPTO_MIT_TEST, dest_dir=self.data_dir)

        logger.info("Test file download complete.")



        logger.info("Downloading validation file...")

        dc.utils.data_utils.download_url(

            url=USPTO_MIT_VALID, dest_dir=self.data_dir)

        logger.info("Validation file download complete.")

      dataset_url = USPTO_MIT_URL



    if self.subset == 'STEREO':

      train_file = os.path.join(self.data_dir, USPTO_STEREO_TRAIN)

      test_file = os.path.join(self.data_dir, USPTO_STEREO_TEST)

      valid_file = os.path.join(self.data_dir, USPTO_STEREO_VALID)



      if not os.path.exists(train_file):

        logger.info("Downloading training file...")

        dc.utils.data_utils.download_url(

            url=USPTO_STEREO_TRAIN, dest_dir=self.data_dir)

        logger.info("Training file download complete.")



        logger.info("Downloading test file...")

        dc.utils.data_utils.download_url(

            url=USPTO_STEREO_TEST, dest_dir=self.data_dir)

        logger.info("Test file download complete.")

      dataset_url = USPTO_STEREO_URL



        logger.info("Downloading validation file...")

        dc.utils.data_utils.download_url(

            url=USPTO_STEREO_VALID, dest_dir=self.data_dir)

        logger.info("Validation file download complete.")

    if self.subset == '50K':

      dataset_url = USPTO_50K_URL



    loader = dc.data.CSVLoader(

        tasks=self.tasks, feature_field="smiles", featurizer=self.featurizer)

    dataset_file = os.path.join(self.data_dir, self.name + '.csv')



    logger.info("Loading train dataset..")

    train_file = loader.create_dataset(train_file, shard_size=8192)

    logger.info("Loading test dataset..")

    test_file = loader.create_dataset(test_file, shard_size=8192)

    logger.info("Loading validation dataset..")

    valid_file = loader.create_dataset(valid_file, shard_size=8192)

    logger.info("Loading successful!")

    if not os.path.exists(dataset_file):

      logger.info("Downloading dataset...")

      dc.utils.data_utils.download_url(url=dataset_url, dest_dir=self.data_dir)

      logger.info("Dataset download complete.")



    #need to figure out how to return the train, test and valid files!

    return (train_file, test_file, valid_file)

    loader = dc.data.CSVLoader(tasks=[], featurizer=self.featurizer)



    def load_precomputed_splits(self) -> List[Dataset, ...]:

      pass

    return loader.create_dataset(dataset_file, shard_size=8192)





def load_uspto(

    featurizer=None,

    featurizer=None,  # should I remove this?

    splitter=None,

    transformers=None,

    reload: bool = True,
@@ -110,6 +66,7 @@ def load_uspto( 
    **kwargs

) -> Tuple[List[str], Tuple[Dataset, ...], List[dc.trans.Transformer]]:



  featurizer = dc.feat.UserDefinedFeaturizer([])

  loader = _USPTOLoader(

      featurizer,

      splitter,

CRA Git | Maintained and supported by SUSTech CRA and CCSE