system print Args: s (str): the string to print

create cold-split where given one or multiple columns, it first splits based on entities in the columns and then maps all associated data points to t

create standard split: imbalanced training dataset and balanced validation and testing dataset Args: df (pd.DataFrame): dataset dataframe

fuzzy matching between the real dataset name and the input name Args: name (str): input dataset name given by users dataset_names (str):

Print basic data statistics.

It is a helper for get_parser. It calls argparse.ArgumentParser().add_argument. If the argument (custom parser name) already in parse

create random split class by class Args: df (pd.DataFrame): dataset dataframe fold_seed (int): the random seed frac (list): a list o

create splits based on time, class by class Args: df (pd.DataFrame): the dataset dataframe frac (list): list of train/valid/test fraction

print the statistics of the dataset

Reinitialize model parameters.

create cold-split where given one or multiple columns, it first splits based on entities in the columns and then maps all associated data points to t

generate data in some format, e.g., pandas.DataFrame Args: format (str, optional): format of data, the d

load a pandas dataframe from local file. Args: name (str): dataset name path (str): the path where the dataset is saved Returns:

Function for splitting drug combination dataset such that no combinations are shared across the split, class by class Args: df (pd.Dataframe

create random split Args: df (pd.DataFrame): dataset dataframe fold_seed (int): the random seed frac (list): a list of train/valid/t

label unit conversion helper function Args: y (list): a list of labels from_ (str): source units, 'nM'/'p' to_ (str): target units,

split whole dataset into closed set and open set (for testing only), class by class Args: frac (float): fraction of number of open classes Retur

wrapper for downloading a dataset given the name and path, for csv,pkl,tsv files Args: name (str): the rough dataset query name path (str

Arguments: format (str, optional): the dataset format Returns: pd.DataFrame/dict/np.array: w

install pip package Args: package (str): package name

Adopt cost-sensitive cross-entropy as the default Args: epoch: int. starting from 1.

input_dim (int) output_dim (int) hidden_dims_lst (list, each element is a integer, indicating the hidden size)

Get kwargs from the config. Use for designer but is not available for user. :param config: the parse_args() results :

Function for splitting drug combination dataset such that no combinations are shared across the split Args: df (pd.Dataframe): dataset to sp

create splits based on time Args: df (pd.DataFrame): the dataset dataframe frac (list): list of train/valid/test fractions date_colu

create scaffold split. it first generates molecular scaffold for each molecule and then split based on scaffolds, class by class reference: https://g

dataverse download helper with progress bar, for ImDrug datasets hosted on Harvard Dataverse Args: url (str): the url of the dataset path

update dict of dicts with new values

Negative Sampling for Binary Interaction Dataset Args: df (pandas.DataFrame): input dataset dataframe column_names (list): column names i

Initialize config Should be written by designer and used by user :param mapping: Dict. The mapping for parser (param name -->

Get args from the config. Use for designer but is not available for user. :param config: the parse_args() results :pa

The simple version of _parse_config. For internal use only. :param config: the target args generated by parse_args. :param ar

Extract property value from record using the given urn search filter.

Add path to PYTHONPATH environment variable for relative import

obtain evaluator function given the evaluator name

Assuming i.i.d of the samples in each class, to compensate for the effect of imbalanced distribution on f1 score, we calculate precision based on f

create scaffold split. it first generates molecular scaffold for each molecule and then split based on scaffolds reference: https://github.com/chempr

Get the closest match by Levenshtein Distance. Args: predefined_tokens (list): Predefined string tokens. test_token (str): User input tha

generate data in some format, e.g., pandas.DataFrame Args: format (str, optional): format of data, the d

Arguments: format (str, optional): the returning dataset format, defaults to 'df' Returns: pandas DataFrame/dict: a dataframe of a dataset

retrieve all .log files from training output

The function to get the parameter set :param func: The target function. :param except_list: the excepted parameter name.

split function, overwritten by single_pred/multi_pred/generation for more specific splits Arguments: method: splitting sc

a wrapper to download, process and load two-instance prediction task datasets Args: name (str): the rough dataset name path (str): the da