rztdl.dl.dataset package

Submodules

rztdl.dl.dataset.dataset module

@created on: 12/15/19, @author: Himaprasoon, @version: v0.0.1

Description:

Sphinx Documentation Status:

class rztdl.dl.dataset.dataset.Dataset(buffer_names: typing.List[str], name: str)[source]

Bases: object

check_splits(split_handler=None)[source]
prepare_dataset(required_inputs, split_handler)[source]
set_split_buffer_name_mapping(split_names)[source]

rztdl.dl.dataset.dataset_handler module

@created on: 12/15/19, @author: Himaprasoon, @version: v0.0.1

Description:

Sphinx Documentation Status:

class rztdl.dl.dataset.dataset_handler.DatasetHandler(name='DatasetHandler')[source]

Bases: object

add(dataset)[source]

Adds Dataset to Dataset Handler :type dataset: Dataset :param dataset: :return:

check_split_buffer_duplicate(dataset)[source]

Checks if any two datasets has same buffers for the same split :param dataset: :return:

initialize(split_required_buffer_mapping, batch_size, as_dict=True, split_handler=None)[source]
Parameters:
  • split_handler (Optional[TrainSplitHandler]) –
  • batch_size (int) – Batch size of data to be read
  • split_required_buffer_mapping (Dict[str, List[str]]) – Input Layer names to read data for.
  • as_dict (bool) – Boolean Indicates if output should be a dictionary to list
Returns:

rztdl.dl.dataset.generator_dataset module

@created on: 1/8/20, @author: Himaprasoon, @version: v0.0.1

Description:

Sphinx Documentation Status:

class rztdl.dl.dataset.generator_dataset.GeneratorDataset(name: str, buffer_names: typing.List[str], gen_function: typing.Union[function, typing.Dict[str, function]])[source]

Bases: rztdl.dl.dataset.dataset.Dataset

prepare_dataset(required_inputs, split_handler)[source]

rztdl.dl.dataset.splits module

class rztdl.dl.dataset.splits.DataSplit(name: str, split_ratio: int, metrics: typing.List[str], log_frequency: typing.Union[int, NoneType] = 1)[source]

Bases: object

Parameters:
  • name (str) – Name of split
  • split_ratio (int) – split_ratio
  • metrics (List[str]) –
  • log_frequency (Optional[int]) –
class rztdl.dl.dataset.splits.TestDataSplit(name: str, split_ratio: int, metrics: typing.List[str], log_frequency: typing.Union[int, NoneType] = None)[source]

Bases: rztdl.dl.dataset.splits.DataSplit

initialize(model)[source]
reset_metrics()[source]
class rztdl.dl.dataset.splits.TrainDataSplit(name: str, split_ratio: int, metrics: typing.List[str], log_frequency: typing.Union[int, NoneType] = None)[source]

Bases: rztdl.dl.dataset.splits.DataSplit

initialize(model)[source]
reset_metrics()[source]
class rztdl.dl.dataset.splits.TrainSplitHandler(name: str = 'TrainSplitHandler')[source]

Bases: object

add(split)[source]
check_closed()[source]
close()[source]
get_split(sp_name)[source]
validate()[source]
class rztdl.dl.dataset.splits.ValidationDataSplit(name: str, split_ratio: int, metrics: typing.List[str], interval, frequency: int, log_frequency: typing.Union[int, NoneType] = None)[source]

Bases: rztdl.dl.dataset.splits.DataSplit

” Split for Validation

Parameters:
  • name (str) –
  • split_ratio (int) –
  • metrics (List[str]) –
  • interval
  • frequency (int) –
  • log_frequency (Optional[int]) –
initialize(model)[source]
reset_metrics()[source]

Module contents

@created on: 12/15/19, @author: Himaprasoon, @version: v0.0.1

Description:

Sphinx Documentation Status: