pyhazards.datasets package

Submodules

pyhazards.datasets.base module

class pyhazards.datasets.base.DataBundle(splits, feature_spec, label_spec, metadata=<factory>)

Bases: object

Bundle of train/val/test splits plus metadata. Keeps feature/label specs to make model construction easy.

feature_spec: FeatureSpec

get_split(name)

Return type:

DataSplit

label_spec: LabelSpec

metadata: Dict[str, Any]

splits: Dict[str, DataSplit]

class pyhazards.datasets.base.DataSplit(inputs, targets, metadata=<factory>)

Bases: object

Container for a single split.

inputs: Any

metadata: Dict[str, Any]

targets: Any

class pyhazards.datasets.base.Dataset(cache_dir=None)

Bases: object

Base class for hazard datasets. Subclasses should load data and return a DataBundle with splits ready for training.

_load()

Return type:

DataBundle

load(split=None, transforms=None)

Return a DataBundle. Optionally return a specific split if provided.

Return type:

DataBundle

name: str = 'base'

class pyhazards.datasets.base.FeatureSpec(input_dim=None, channels=None, description=None, extra=<factory>)

Bases: object

Describes input features (shapes, dtypes, normalization).

channels: Optional[int] = None

description: Optional[str] = None

extra: Dict[str, Any]

input_dim: Optional[int] = None

class pyhazards.datasets.base.LabelSpec(num_targets=None, task_type='regression', description=None, extra=<factory>)

Bases: object

Describes labels/targets for downstream tasks.

description: Optional[str] = None

extra: Dict[str, Any]

num_targets: Optional[int] = None

task_type: str = 'regression'

class pyhazards.datasets.base.Transform(*args, **kwargs)

Bases: Protocol

Callable data transform.

_abc_impl = <_abc._abc_data object>

_is_protocol = True

pyhazards.datasets.registry module

pyhazards.datasets.registry.available_datasets()

pyhazards.datasets.registry.load_dataset(name, **kwargs)

Return type:

Dataset

pyhazards.datasets.registry.register_dataset(name, builder)

Return type:

None

pyhazards.datasets.transforms package

Reusable transforms for preprocessing hazard datasets. Currently placeholders; implement normalization, index computation, temporal windowing, etc.

pyhazards.datasets.hazards package

Namespace for hazard-specific dataset loaders (earthquake, wildfire, flood, hurricane, landslide, etc.). Populate with concrete Dataset subclasses and register them in pyhazards.datasets.registry.

Module contents

class pyhazards.datasets.DataBundle(splits, feature_spec, label_spec, metadata=<factory>)

Bases: object

Bundle of train/val/test splits plus metadata. Keeps feature/label specs to make model construction easy.

feature_spec: FeatureSpec

get_split(name)

Return type:

DataSplit

label_spec: LabelSpec

metadata: Dict[str, Any]

splits: Dict[str, DataSplit]

class pyhazards.datasets.DataSplit(inputs, targets, metadata=<factory>)

Bases: object

Container for a single split.

inputs: Any

metadata: Dict[str, Any]

targets: Any

class pyhazards.datasets.Dataset(cache_dir=None)

Bases: object

Base class for hazard datasets. Subclasses should load data and return a DataBundle with splits ready for training.

_load()

Return type:

DataBundle

load(split=None, transforms=None)

Return a DataBundle. Optionally return a specific split if provided.

Return type:

DataBundle

name: str = 'base'

class pyhazards.datasets.FeatureSpec(input_dim=None, channels=None, description=None, extra=<factory>)

Bases: object

Describes input features (shapes, dtypes, normalization).

channels: Optional[int] = None

description: Optional[str] = None

extra: Dict[str, Any]

input_dim: Optional[int] = None

class pyhazards.datasets.GraphTemporalDataset(x, y, adjacency=None)

Bases: Dataset

Simple container for county/day style tensors with an optional adjacency.

Each sample is a window of shape (past_days, num_counties, num_features) and a label of shape (num_counties,).

class pyhazards.datasets.LabelSpec(num_targets=None, task_type='regression', description=None, extra=<factory>)

Bases: object

Describes labels/targets for downstream tasks.

description: Optional[str] = None

extra: Dict[str, Any]

num_targets: Optional[int] = None

task_type: str = 'regression'

pyhazards.datasets.available_datasets()

pyhazards.datasets.graph_collate(batch)

Collate function that stacks x and adjacency if provided.

pyhazards.datasets.load_dataset(name, **kwargs)

Return type:

Dataset

pyhazards.datasets.register_dataset(name, builder)

Return type:

None