gt4sd.algorithms.generation.diffusion.core module¶

HuggingFace Diffusers generation algorithm. Code and models adapted from https://github.com/huggingface/diffusers.

Summary¶

Classes:

`DDIMGenerator`	DDIM - Configuration to generate using a denoising diffusion implicit model.
`DDPMGenerator`	DDPM - Configuration to generate using unconditional denoising diffusion models.
`DiffusersConfiguration`	Basic configuration for a diffusion algorithm.
`DiffusersGenerationAlgorithm`
`GeoDiffGenerator`	GeoDiff Diffusion Model - Configuration for conditional 3D molecule structure generation given 2D information using a GeoDiff diffusion model.
`LDMGenerator`	Unconditional Latent Diffusion Model - Configuration to generate using a latent diffusion model.
`LDMTextToImageGenerator`	Conditional Latent Diffusion Model - Configuration for conditional text2image generation using a latent diffusion model.
`ScoreSdeGenerator`	Score SDE Generative Model - Configuration to generate using a score-based diffusion generative model.
`StableDiffusionGenerator`	Stable Diffusion Model - Configuration for conditional text2image generation using a stable diffusion model.

Reference¶

class DiffusersGenerationAlgorithm(configuration, target=None)[source]¶

Bases: GeneratorAlgorithm[S, None]

__init__(configuration, target=None)[source]¶

Diffusers generation algorithm.

Parameters

configuration (AlgorithmConfiguration) – domain and application specification, defining types and validations.
target (Optional[~S, None]) – none for untargeted generation.

Example

An example for using a generative algorithm from Diffusers:

configuration = GeneratorConfiguration()
algorithm = DiffusersGenerationAlgorithm(configuration=configuration)
items = list(algorithm.sample(1))
print(items)

get_generator(configuration, target)[source]¶

Get the function to sample batches.

Parameters

configuration (AlgorithmConfiguration[~S, None]) – helps to set up the application.
target (None) – context or condition for the generation.

Return type

Callable[[], Iterable[Any]]

Returns

callable generating a batch of items.

validate_configuration(configuration)[source]¶

Overload to validate the a configuration for the algorithm.

Parameters: configuration (AlgorithmConfiguration) – the algorithm configuration.
Raises: InvalidAlgorithmConfiguration – in case the configuration for the algorithm is invalid.
Return type: AlgorithmConfiguration
Returns: the validated configuration.

__abstractmethods__ = frozenset({})¶

__annotations__ = {'generate': 'Untargeted', 'generator': 'Union[Untargeted, Targeted[T]]', 'max_runtime': 'int', 'max_samples': 'int', 'target': 'Optional[T]'}¶

__doc__ = None¶

__module__ = 'gt4sd.algorithms.generation.diffusion.core'¶

__orig_bases__ = (gt4sd.algorithms.core.GeneratorAlgorithm[~S, NoneType],)¶

__parameters__ = (~S,)¶

_abc_impl = <_abc._abc_data object>¶

class DiffusersConfiguration(*args, **kwargs)[source]¶

Bases: DiffusersConfiguration, Generic[T]

Basic configuration for a diffusion algorithm.

algorithm_type: ClassVar[str] = 'generation'¶: General type of generative algorithm.

domain: ClassVar[str] = 'vision'¶: General application domain. Hints at input/output types.

modality: str = 'image'¶

model_type: str = 'diffusion'¶

scheduler_type: str = 'discrete'¶

prompt: Union[str, Dict[str, Any]] = None¶

get_conditional_generator(resources_path, **kwargs)[source]¶

Return type: Generator

__annotations__ = {'algorithm_application': 'ClassVar[str]', 'algorithm_name': 'ClassVar[str]', 'algorithm_type': typing.ClassVar[str], 'algorithm_version': 'str', 'domain': typing.ClassVar[str], 'modality': <class 'str'>, 'model_type': <class 'str'>, 'prompt': typing.Union[str, typing.Dict[str, typing.Any]], 'scheduler_type': <class 'str'>}¶

__dataclass_fields__ = {'algorithm_application': Field(name='algorithm_application',type=typing.ClassVar[str],default='DiffusersConfiguration',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_name': Field(name='algorithm_name',type=typing.ClassVar[str],default='DiffusersGenerationAlgorithm',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_type': Field(name='algorithm_type',type=typing.ClassVar[str],default='generation',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_version': Field(name='algorithm_version',type='str',default='',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'domain': Field(name='domain',type=typing.ClassVar[str],default='vision',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'modality': Field(name='modality',type=<class 'str'>,default='image',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({'description': "Modality. Supported: 'image', 'text', 'audio', 'molecule'."}),kw_only=False,_field_type=_FIELD), 'model_type': Field(name='model_type',type=<class 'str'>,default='diffusion',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({'description': 'Type of the model. Supported: diffusion, diffusion_implicit, latent_diffusion, latent_diffusion_conditional, stable_diffusion, score_sde, geodiff'}),kw_only=False,_field_type=_FIELD), 'prompt': Field(name='prompt',type=typing.Union[str, typing.Dict[str, typing.Any]],default=None,default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({'description': 'Prompt for conditional generation.'}),kw_only=False,_field_type=_FIELD), 'scheduler_type': Field(name='scheduler_type',type=<class 'str'>,default='discrete',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({'description': 'Type of the noise scheduler. Supported: ddpm, ddim, discrete, continuous'}),kw_only=False,_field_type=_FIELD)}¶

__dataclass_params__ = _DataclassParams(init=True,repr=True,eq=True,order=False,unsafe_hash=False,frozen=False)¶

__doc__ = 'Basic configuration for a diffusion algorithm.'¶

__eq__(other)¶: Return self==value.

__hash__ = None¶

__init__(*args, **kwargs)¶

__is_pydantic_dataclass__ = True¶

__match_args__ = ('algorithm_version', 'modality', 'model_type', 'scheduler_type', 'prompt')¶

__module__ = 'gt4sd.algorithms.generation.diffusion.core'¶

__orig_bases__ = (<class 'types.DiffusersConfiguration'>, typing.Generic[~T])¶

__parameters__ = (~T,)¶

__pydantic_complete__ = True¶

__pydantic_config__ = {}¶

__pydantic_core_schema__ = {'cls': <class 'gt4sd.algorithms.generation.diffusion.core.DiffusersConfiguration'>, 'config': {'title': 'DiffusersConfiguration'}, 'fields': ['algorithm_version', 'modality', 'model_type', 'scheduler_type', 'prompt'], 'frozen': False, 'post_init': False, 'ref': 'types.DiffusersConfiguration:94818326168528', 'schema': {'collect_init_only': False, 'computed_fields': [], 'dataclass_name': 'DiffusersConfiguration', 'fields': [{'type': 'dataclass-field', 'name': 'algorithm_version', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': ''}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'modality', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'image'}, 'kw_only': False, 'init': True, 'metadata': {'pydantic_js_updates': {'description': "Modality. Supported: 'image', 'text', 'audio', 'molecule'."}}}, {'type': 'dataclass-field', 'name': 'model_type', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'diffusion'}, 'kw_only': False, 'init': True, 'metadata': {'pydantic_js_updates': {'description': 'Type of the model. Supported: diffusion, diffusion_implicit, latent_diffusion, latent_diffusion_conditional, stable_diffusion, score_sde, geodiff'}}}, {'type': 'dataclass-field', 'name': 'scheduler_type', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'discrete'}, 'kw_only': False, 'init': True, 'metadata': {'pydantic_js_updates': {'description': 'Type of the noise scheduler. Supported: ddpm, ddim, discrete, continuous'}}}, {'type': 'dataclass-field', 'name': 'prompt', 'schema': {'type': 'default', 'schema': {'type': 'union', 'choices': [{'type': 'str'}, {'type': 'dict', 'keys_schema': {'type': 'str'}, 'values_schema': {'type': 'any'}}]}, 'default': None}, 'kw_only': False, 'init': True, 'metadata': {'pydantic_js_updates': {'description': 'Prompt for conditional generation.'}}}], 'type': 'dataclass-args'}, 'slots': True, 'type': 'dataclass'}¶

__pydantic_decorators__ = DecoratorInfos(validators={}, field_validators={}, root_validators={}, field_serializers={}, model_serializers={}, model_validators={}, computed_fields={})¶

__pydantic_fields__ = {'algorithm_version': FieldInfo(annotation=str, required=False, default='', init=True, kw_only=False), 'modality': FieldInfo(annotation=str, required=False, default='image', description="Modality. Supported: 'image', 'text', 'audio', 'molecule'.", init=True, kw_only=False), 'model_type': FieldInfo(annotation=str, required=False, default='diffusion', description='Type of the model. Supported: diffusion, diffusion_implicit, latent_diffusion, latent_diffusion_conditional, stable_diffusion, score_sde, geodiff', init=True, kw_only=False), 'prompt': FieldInfo(annotation=Union[str, Dict[str, Any]], required=False, default=None, description='Prompt for conditional generation.', init=True, kw_only=False), 'scheduler_type': FieldInfo(annotation=str, required=False, default='discrete', description='Type of the noise scheduler. Supported: ddpm, ddim, discrete, continuous', init=True, kw_only=False)}¶

classmethod __pydantic_fields_complete__()¶

Return whether the fields were successfully collected (i.e. type hints were successfully resolved).

This is a private helper, not meant to be used outside Pydantic.

Return type: bool

__pydantic_serializer__ = SchemaSerializer(serializer=PolymorphismTrampoline( PolymorphismTrampoline { class: Py( 0x0000563c9ca643d0, ), serializer: PolymorphismTrampoline( PolymorphismTrampoline { class: Py( 0x0000563c9ca643d0, ), serializer: Dataclass( DataclassSerializer { class: Py( 0x0000563c9ca643d0, ), serializer: Fields( GeneralFieldsSerializer { fields: { "algorithm_version": SerField { key: "algorithm_version", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f222f308030, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "model_type": SerField { key: "model_type", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f2160a73570, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "prompt": SerField { key: "prompt", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x0000563c55c10840, ), ), serializer: Union( UnionSerializer { choices: UnionChoices { choices: [ Str( StrSerializer, ), Dict( DictSerializer { key_serializer: Str( StrSerializer, ), value_serializer: Any( AnySerializer, ), filter: SchemaFilter { include: None, exclude: None, }, name: "dict[str, any]", }, ), ], }, name: "Union[str, dict[str, any]]", }, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "scheduler_type": SerField { key: "scheduler_type", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f216b250370, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "modality": SerField { key: "modality", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f222e9acfb0, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, }, computed_fields: Some( ComputedFields( [], ), ), mode: SimpleDict, extra_serializer: None, filter: SchemaFilter { include: None, exclude: None, }, required_fields: 5, }, ), fields: [ Py( 0x00007f222affaf10, ), Py( 0x00007f2166fb8ef0, ), Py( 0x00007f222ba93fb0, ), Py( 0x00007f214c6b59b0, ), Py( 0x00007f222ef49930, ), ], name: "DiffusersConfiguration", }, ), enabled_from_config: false, }, ), enabled_from_config: false, }, ), definitions=[])¶

__pydantic_validator__ = SchemaValidator(title="DiffusersConfiguration", validator=Dataclass( DataclassValidator { strict: false, validator: DataclassArgs( DataclassArgsValidator { fields: [ Field { kw_only: false, name: "algorithm_version", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "algorithm_version", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f222f308030, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "modality", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "modality", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f222e9acfb0, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "model_type", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "model_type", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f2160a73570, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "scheduler_type", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "scheduler_type", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f216b250370, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "prompt", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "prompt", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x0000563c55c10840, ), ), on_error: Raise, validator: Union( UnionValidator { mode: Smart, choices: [ ( Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), None, ), ( Dict( DictValidator { strict: false, key_validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), value_validator: Any( AnyValidator, ), min_length: None, max_length: None, fail_fast: false, name: "dict[str,any]", }, ), None, ), ], custom_error: None, name: "union[str,dict[str,any]]", }, ), validate_default: false, copy_default: false, name: "default[union[str,dict[str,any]]]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, ], positional_count: 5, init_only_count: None, dataclass_name: "DiffusersConfiguration", validator_name: "dataclass-args[DiffusersConfiguration]", extra_behavior: Ignore, extras_validator: None, loc_by_alias: true, validate_by_alias: None, validate_by_name: None, }, ), class: Py( 0x0000563c9ca643d0, ), generic_origin: None, fields: [ Py( 0x00007f222affaf10, ), Py( 0x00007f2166fb8ef0, ), Py( 0x00007f222ba93fb0, ), Py( 0x00007f214c6b59b0, ), Py( 0x00007f222ef49930, ), ], post_init: None, revalidate: Never, name: "DiffusersConfiguration", frozen: false, slots: true, }, ), definitions=[], cache_strings=True)¶

__repr__()¶: Return repr(self).

__signature__ = <Signature (algorithm_version: 'str' = '', modality: str = 'image', model_type: str = 'diffusion', scheduler_type: str = 'discrete', prompt: Union[str, Dict[str, Any]] = None) -> None>¶

__wrapped__¶: alias of DiffusersConfiguration

class DDPMGenerator(*args, **kwargs)[source]¶

Bases: DDPMGenerator

DDPM - Configuration to generate using unconditional denoising diffusion models.

algorithm_version: str = 'google/ddpm-cifar10-32'¶

To differentiate between different versions of an application.

There is no imposed naming convention.

model_type: str = 'diffusion'¶

scheduler_type: str = 'ddpm'¶

modality: str = 'image'¶

classmethod list_versions()[source]¶

Get possible algorithm versions.

Standard S3 and cache search adding the version used in the configuration.

Return type: Set[str]
Returns: viable values as algorithm_version for the environment.

__annotations__ = {'algorithm_application': 'ClassVar[str]', 'algorithm_name': 'ClassVar[str]', 'algorithm_type': 'ClassVar[str]', 'algorithm_version': <class 'str'>, 'domain': 'ClassVar[str]', 'modality': <class 'str'>, 'model_type': <class 'str'>, 'prompt': 'Union[str, Dict[str, Any]]', 'scheduler_type': <class 'str'>}¶

__dataclass_fields__ = {'algorithm_application': Field(name='algorithm_application',type=typing.ClassVar[str],default='DDPMGenerator',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_name': Field(name='algorithm_name',type=typing.ClassVar[str],default='DiffusersGenerationAlgorithm',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_type': Field(name='algorithm_type',type=typing.ClassVar[str],default='generation',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_version': Field(name='algorithm_version',type=<class 'str'>,default='google/ddpm-cifar10-32',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'domain': Field(name='domain',type=typing.ClassVar[str],default='vision',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'modality': Field(name='modality',type=<class 'str'>,default='image',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'model_type': Field(name='model_type',type=<class 'str'>,default='diffusion',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'prompt': Field(name='prompt',type=typing.Union[str, typing.Dict[str, typing.Any]],default=None,default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({'description': 'Prompt for conditional generation.'}),kw_only=False,_field_type=_FIELD), 'scheduler_type': Field(name='scheduler_type',type=<class 'str'>,default='ddpm',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD)}¶

__dataclass_params__ = _DataclassParams(init=True,repr=True,eq=True,order=False,unsafe_hash=False,frozen=False)¶

__doc__ = 'DDPM - Configuration to generate using unconditional denoising diffusion models.'¶

__eq__(other)¶: Return self==value.

__hash__ = None¶

__init__(*args, **kwargs)¶

__is_pydantic_dataclass__ = True¶

__match_args__ = ('algorithm_version', 'modality', 'model_type', 'scheduler_type', 'prompt')¶

__module__ = 'gt4sd.algorithms.generation.diffusion.core'¶

__parameters__ = (~T,)¶

__pydantic_complete__ = True¶

__pydantic_config__ = {}¶

__pydantic_core_schema__ = {'cls': <class 'gt4sd.algorithms.generation.diffusion.core.DDPMGenerator'>, 'config': {'title': 'DDPMGenerator'}, 'fields': ['algorithm_version', 'modality', 'model_type', 'scheduler_type', 'prompt'], 'frozen': False, 'post_init': False, 'ref': 'types.DDPMGenerator:94818325954368', 'schema': {'collect_init_only': False, 'computed_fields': [], 'dataclass_name': 'DDPMGenerator', 'fields': [{'type': 'dataclass-field', 'name': 'algorithm_version', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'google/ddpm-cifar10-32'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'modality', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'image'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'model_type', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'diffusion'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'scheduler_type', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'ddpm'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'prompt', 'schema': {'type': 'default', 'schema': {'type': 'union', 'choices': [{'type': 'str'}, {'type': 'dict', 'keys_schema': {'type': 'str'}, 'values_schema': {'type': 'any'}}]}, 'default': None}, 'kw_only': False, 'init': True, 'metadata': {'pydantic_js_updates': {'description': 'Prompt for conditional generation.'}}}], 'type': 'dataclass-args'}, 'slots': True, 'type': 'dataclass'}¶

__pydantic_decorators__ = DecoratorInfos(validators={}, field_validators={}, root_validators={}, field_serializers={}, model_serializers={}, model_validators={}, computed_fields={})¶

__pydantic_fields__ = {'algorithm_version': FieldInfo(annotation=str, required=False, default='google/ddpm-cifar10-32', init=True, kw_only=False), 'modality': FieldInfo(annotation=str, required=False, default='image', init=True, kw_only=False), 'model_type': FieldInfo(annotation=str, required=False, default='diffusion', init=True, kw_only=False), 'prompt': FieldInfo(annotation=Union[str, Dict[str, Any]], required=False, default=None, description='Prompt for conditional generation.', init=True, kw_only=False), 'scheduler_type': FieldInfo(annotation=str, required=False, default='ddpm', init=True, kw_only=False)}¶

classmethod __pydantic_fields_complete__()¶

Return whether the fields were successfully collected (i.e. type hints were successfully resolved).

This is a private helper, not meant to be used outside Pydantic.

Return type: bool

__pydantic_serializer__ = SchemaSerializer(serializer=PolymorphismTrampoline( PolymorphismTrampoline { class: Py( 0x0000563c9ca2ff40, ), serializer: PolymorphismTrampoline( PolymorphismTrampoline { class: Py( 0x0000563c9ca2ff40, ), serializer: Dataclass( DataclassSerializer { class: Py( 0x0000563c9ca2ff40, ), serializer: Fields( GeneralFieldsSerializer { fields: { "modality": SerField { key: "modality", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f222e9acfb0, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "model_type": SerField { key: "model_type", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f2160a73570, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "algorithm_version": SerField { key: "algorithm_version", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f214c69f0f0, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "scheduler_type": SerField { key: "scheduler_type", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f215f60ddf0, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "prompt": SerField { key: "prompt", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x0000563c55c10840, ), ), serializer: Union( UnionSerializer { choices: UnionChoices { choices: [ Str( StrSerializer, ), Dict( DictSerializer { key_serializer: Str( StrSerializer, ), value_serializer: Any( AnySerializer, ), filter: SchemaFilter { include: None, exclude: None, }, name: "dict[str, any]", }, ), ], }, name: "Union[str, dict[str, any]]", }, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, }, computed_fields: Some( ComputedFields( [], ), ), mode: SimpleDict, extra_serializer: None, filter: SchemaFilter { include: None, exclude: None, }, required_fields: 5, }, ), fields: [ Py( 0x00007f222affaf10, ), Py( 0x00007f2166fb8ef0, ), Py( 0x00007f222ba93fb0, ), Py( 0x00007f214c6b59b0, ), Py( 0x00007f222ef49930, ), ], name: "DDPMGenerator", }, ), enabled_from_config: false, }, ), enabled_from_config: false, }, ), definitions=[])¶

__pydantic_validator__ = SchemaValidator(title="DDPMGenerator", validator=Dataclass( DataclassValidator { strict: false, validator: DataclassArgs( DataclassArgsValidator { fields: [ Field { kw_only: false, name: "algorithm_version", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "algorithm_version", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f214c69f0f0, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "modality", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "modality", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f222e9acfb0, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "model_type", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "model_type", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f2160a73570, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "scheduler_type", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "scheduler_type", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f215f60ddf0, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "prompt", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "prompt", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x0000563c55c10840, ), ), on_error: Raise, validator: Union( UnionValidator { mode: Smart, choices: [ ( Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), None, ), ( Dict( DictValidator { strict: false, key_validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), value_validator: Any( AnyValidator, ), min_length: None, max_length: None, fail_fast: false, name: "dict[str,any]", }, ), None, ), ], custom_error: None, name: "union[str,dict[str,any]]", }, ), validate_default: false, copy_default: false, name: "default[union[str,dict[str,any]]]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, ], positional_count: 5, init_only_count: None, dataclass_name: "DDPMGenerator", validator_name: "dataclass-args[DDPMGenerator]", extra_behavior: Ignore, extras_validator: None, loc_by_alias: true, validate_by_alias: None, validate_by_name: None, }, ), class: Py( 0x0000563c9ca2ff40, ), generic_origin: None, fields: [ Py( 0x00007f222affaf10, ), Py( 0x00007f2166fb8ef0, ), Py( 0x00007f222ba93fb0, ), Py( 0x00007f214c6b59b0, ), Py( 0x00007f222ef49930, ), ], post_init: None, revalidate: Never, name: "DDPMGenerator", frozen: false, slots: true, }, ), definitions=[], cache_strings=True)¶

__repr__()¶: Return repr(self).

__signature__ = <Signature (*args: Any, algorithm_version: str = 'google/ddpm-cifar10-32', modality: str = 'image', model_type: str = 'diffusion', scheduler_type: str = 'ddpm', prompt: Union[str, Dict[str, Any]] = None) -> None>¶

__wrapped__¶: alias of DDPMGenerator

algorithm_application: ClassVar[str] = 'DDPMGenerator'¶

Unique name for the application that is the use of this configuration together with a specific algorithm.

Will be set when registering to ApplicationsRegistry, but can be given by direct registration (See register_algorithm_application)

algorithm_name: ClassVar[str] = 'DiffusersGenerationAlgorithm'¶

Name of the algorithm to use with this configuration.

Will be set when registering to ApplicationsRegistry

class DDIMGenerator(*args, **kwargs)[source]¶

Bases: DDIMGenerator

DDIM - Configuration to generate using a denoising diffusion implicit model.

algorithm_version: str = 'dboshardy/ddim-butterflies-128'¶

To differentiate between different versions of an application.

There is no imposed naming convention.

model_type: str = 'diffusion_implicit'¶

scheduler_type: str = 'ddim'¶

modality: str = 'image'¶

classmethod list_versions()[source]¶

Get possible algorithm versions.

Standard S3 and cache search adding the version used in the configuration.

Return type: Set[str]
Returns: viable values as algorithm_version for the environment.

__annotations__ = {'algorithm_application': 'ClassVar[str]', 'algorithm_name': 'ClassVar[str]', 'algorithm_type': 'ClassVar[str]', 'algorithm_version': <class 'str'>, 'domain': 'ClassVar[str]', 'modality': <class 'str'>, 'model_type': <class 'str'>, 'prompt': 'Union[str, Dict[str, Any]]', 'scheduler_type': <class 'str'>}¶

__dataclass_fields__ = {'algorithm_application': Field(name='algorithm_application',type=typing.ClassVar[str],default='DDIMGenerator',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_name': Field(name='algorithm_name',type=typing.ClassVar[str],default='DiffusersGenerationAlgorithm',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_type': Field(name='algorithm_type',type=typing.ClassVar[str],default='generation',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_version': Field(name='algorithm_version',type=<class 'str'>,default='dboshardy/ddim-butterflies-128',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'domain': Field(name='domain',type=typing.ClassVar[str],default='vision',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'modality': Field(name='modality',type=<class 'str'>,default='image',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'model_type': Field(name='model_type',type=<class 'str'>,default='diffusion_implicit',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'prompt': Field(name='prompt',type=typing.Union[str, typing.Dict[str, typing.Any]],default=None,default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({'description': 'Prompt for conditional generation.'}),kw_only=False,_field_type=_FIELD), 'scheduler_type': Field(name='scheduler_type',type=<class 'str'>,default='ddim',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD)}¶

__dataclass_params__ = _DataclassParams(init=True,repr=True,eq=True,order=False,unsafe_hash=False,frozen=False)¶

__doc__ = 'DDIM - Configuration to generate using a denoising diffusion implicit model.'¶

__eq__(other)¶: Return self==value.

__hash__ = None¶

__init__(*args, **kwargs)¶

__is_pydantic_dataclass__ = True¶

__match_args__ = ('algorithm_version', 'modality', 'model_type', 'scheduler_type', 'prompt')¶

__module__ = 'gt4sd.algorithms.generation.diffusion.core'¶

__parameters__ = (~T,)¶

__pydantic_complete__ = True¶

__pydantic_config__ = {}¶

__pydantic_core_schema__ = {'cls': <class 'gt4sd.algorithms.generation.diffusion.core.DDIMGenerator'>, 'config': {'title': 'DDIMGenerator'}, 'fields': ['algorithm_version', 'modality', 'model_type', 'scheduler_type', 'prompt'], 'frozen': False, 'post_init': False, 'ref': 'types.DDIMGenerator:94818326247568', 'schema': {'collect_init_only': False, 'computed_fields': [], 'dataclass_name': 'DDIMGenerator', 'fields': [{'type': 'dataclass-field', 'name': 'algorithm_version', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'dboshardy/ddim-butterflies-128'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'modality', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'image'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'model_type', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'diffusion_implicit'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'scheduler_type', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'ddim'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'prompt', 'schema': {'type': 'default', 'schema': {'type': 'union', 'choices': [{'type': 'str'}, {'type': 'dict', 'keys_schema': {'type': 'str'}, 'values_schema': {'type': 'any'}}]}, 'default': None}, 'kw_only': False, 'init': True, 'metadata': {'pydantic_js_updates': {'description': 'Prompt for conditional generation.'}}}], 'type': 'dataclass-args'}, 'slots': True, 'type': 'dataclass'}¶

__pydantic_decorators__ = DecoratorInfos(validators={}, field_validators={}, root_validators={}, field_serializers={}, model_serializers={}, model_validators={}, computed_fields={})¶

__pydantic_fields__ = {'algorithm_version': FieldInfo(annotation=str, required=False, default='dboshardy/ddim-butterflies-128', init=True, kw_only=False), 'modality': FieldInfo(annotation=str, required=False, default='image', init=True, kw_only=False), 'model_type': FieldInfo(annotation=str, required=False, default='diffusion_implicit', init=True, kw_only=False), 'prompt': FieldInfo(annotation=Union[str, Dict[str, Any]], required=False, default=None, description='Prompt for conditional generation.', init=True, kw_only=False), 'scheduler_type': FieldInfo(annotation=str, required=False, default='ddim', init=True, kw_only=False)}¶

classmethod __pydantic_fields_complete__()¶

Return whether the fields were successfully collected (i.e. type hints were successfully resolved).

This is a private helper, not meant to be used outside Pydantic.

Return type: bool

__pydantic_serializer__ = SchemaSerializer(serializer=PolymorphismTrampoline( PolymorphismTrampoline { class: Py( 0x0000563c9ca77890, ), serializer: PolymorphismTrampoline( PolymorphismTrampoline { class: Py( 0x0000563c9ca77890, ), serializer: Dataclass( DataclassSerializer { class: Py( 0x0000563c9ca77890, ), serializer: Fields( GeneralFieldsSerializer { fields: { "scheduler_type": SerField { key: "scheduler_type", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f215f5e2670, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "model_type": SerField { key: "model_type", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f214c69edd0, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "algorithm_version": SerField { key: "algorithm_version", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f214c69ece0, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "modality": SerField { key: "modality", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f222e9acfb0, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "prompt": SerField { key: "prompt", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x0000563c55c10840, ), ), serializer: Union( UnionSerializer { choices: UnionChoices { choices: [ Str( StrSerializer, ), Dict( DictSerializer { key_serializer: Str( StrSerializer, ), value_serializer: Any( AnySerializer, ), filter: SchemaFilter { include: None, exclude: None, }, name: "dict[str, any]", }, ), ], }, name: "Union[str, dict[str, any]]", }, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, }, computed_fields: Some( ComputedFields( [], ), ), mode: SimpleDict, extra_serializer: None, filter: SchemaFilter { include: None, exclude: None, }, required_fields: 5, }, ), fields: [ Py( 0x00007f222affaf10, ), Py( 0x00007f2166fb8ef0, ), Py( 0x00007f222ba93fb0, ), Py( 0x00007f214c6b59b0, ), Py( 0x00007f222ef49930, ), ], name: "DDIMGenerator", }, ), enabled_from_config: false, }, ), enabled_from_config: false, }, ), definitions=[])¶

__pydantic_validator__ = SchemaValidator(title="DDIMGenerator", validator=Dataclass( DataclassValidator { strict: false, validator: DataclassArgs( DataclassArgsValidator { fields: [ Field { kw_only: false, name: "algorithm_version", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "algorithm_version", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f214c69ece0, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "modality", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "modality", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f222e9acfb0, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "model_type", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "model_type", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f214c69edd0, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "scheduler_type", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "scheduler_type", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f215f5e2670, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "prompt", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "prompt", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x0000563c55c10840, ), ), on_error: Raise, validator: Union( UnionValidator { mode: Smart, choices: [ ( Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), None, ), ( Dict( DictValidator { strict: false, key_validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), value_validator: Any( AnyValidator, ), min_length: None, max_length: None, fail_fast: false, name: "dict[str,any]", }, ), None, ), ], custom_error: None, name: "union[str,dict[str,any]]", }, ), validate_default: false, copy_default: false, name: "default[union[str,dict[str,any]]]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, ], positional_count: 5, init_only_count: None, dataclass_name: "DDIMGenerator", validator_name: "dataclass-args[DDIMGenerator]", extra_behavior: Ignore, extras_validator: None, loc_by_alias: true, validate_by_alias: None, validate_by_name: None, }, ), class: Py( 0x0000563c9ca77890, ), generic_origin: None, fields: [ Py( 0x00007f222affaf10, ), Py( 0x00007f2166fb8ef0, ), Py( 0x00007f222ba93fb0, ), Py( 0x00007f214c6b59b0, ), Py( 0x00007f222ef49930, ), ], post_init: None, revalidate: Never, name: "DDIMGenerator", frozen: false, slots: true, }, ), definitions=[], cache_strings=True)¶

__repr__()¶: Return repr(self).

__signature__ = <Signature (*args: Any, algorithm_version: str = 'dboshardy/ddim-butterflies-128', modality: str = 'image', model_type: str = 'diffusion_implicit', scheduler_type: str = 'ddim', prompt: Union[str, Dict[str, Any]] = None) -> None>¶

__wrapped__¶: alias of DDIMGenerator

algorithm_application: ClassVar[str] = 'DDIMGenerator'¶

Unique name for the application that is the use of this configuration together with a specific algorithm.

Will be set when registering to ApplicationsRegistry, but can be given by direct registration (See register_algorithm_application)

algorithm_name: ClassVar[str] = 'DiffusersGenerationAlgorithm'¶

Name of the algorithm to use with this configuration.

Will be set when registering to ApplicationsRegistry

class LDMGenerator(*args, **kwargs)[source]¶

Bases: LDMGenerator

Unconditional Latent Diffusion Model - Configuration to generate using a latent diffusion model.

algorithm_version: str = 'CompVis/ldm-celebahq-256'¶

To differentiate between different versions of an application.

There is no imposed naming convention.

model_type: str = 'latent_diffusion'¶

scheduler_type: str = 'discrete'¶

modality: str = 'image'¶

classmethod list_versions()[source]¶

Get possible algorithm versions.

Standard S3 and cache search adding the version used in the configuration.

Return type: Set[str]
Returns: viable values as algorithm_version for the environment.

__annotations__ = {'algorithm_application': 'ClassVar[str]', 'algorithm_name': 'ClassVar[str]', 'algorithm_type': 'ClassVar[str]', 'algorithm_version': <class 'str'>, 'domain': 'ClassVar[str]', 'modality': <class 'str'>, 'model_type': <class 'str'>, 'prompt': 'Union[str, Dict[str, Any]]', 'scheduler_type': <class 'str'>}¶

__dataclass_fields__ = {'algorithm_application': Field(name='algorithm_application',type=typing.ClassVar[str],default='LDMGenerator',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_name': Field(name='algorithm_name',type=typing.ClassVar[str],default='DiffusersGenerationAlgorithm',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_type': Field(name='algorithm_type',type=typing.ClassVar[str],default='generation',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_version': Field(name='algorithm_version',type=<class 'str'>,default='CompVis/ldm-celebahq-256',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'domain': Field(name='domain',type=typing.ClassVar[str],default='vision',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'modality': Field(name='modality',type=<class 'str'>,default='image',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'model_type': Field(name='model_type',type=<class 'str'>,default='latent_diffusion',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'prompt': Field(name='prompt',type=typing.Union[str, typing.Dict[str, typing.Any]],default=None,default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({'description': 'Prompt for conditional generation.'}),kw_only=False,_field_type=_FIELD), 'scheduler_type': Field(name='scheduler_type',type=<class 'str'>,default='discrete',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD)}¶

__dataclass_params__ = _DataclassParams(init=True,repr=True,eq=True,order=False,unsafe_hash=False,frozen=False)¶

__doc__ = 'Unconditional Latent Diffusion Model - Configuration to generate using a latent diffusion model.'¶

__eq__(other)¶: Return self==value.

__hash__ = None¶

__init__(*args, **kwargs)¶

__is_pydantic_dataclass__ = True¶

__match_args__ = ('algorithm_version', 'modality', 'model_type', 'scheduler_type', 'prompt')¶

__module__ = 'gt4sd.algorithms.generation.diffusion.core'¶

__parameters__ = (~T,)¶

__pydantic_complete__ = True¶

__pydantic_config__ = {}¶

__pydantic_core_schema__ = {'cls': <class 'gt4sd.algorithms.generation.diffusion.core.LDMGenerator'>, 'config': {'title': 'LDMGenerator'}, 'fields': ['algorithm_version', 'modality', 'model_type', 'scheduler_type', 'prompt'], 'frozen': False, 'post_init': False, 'ref': 'types.LDMGenerator:94818325813200', 'schema': {'collect_init_only': False, 'computed_fields': [], 'dataclass_name': 'LDMGenerator', 'fields': [{'type': 'dataclass-field', 'name': 'algorithm_version', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'CompVis/ldm-celebahq-256'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'modality', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'image'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'model_type', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'latent_diffusion'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'scheduler_type', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'discrete'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'prompt', 'schema': {'type': 'default', 'schema': {'type': 'union', 'choices': [{'type': 'str'}, {'type': 'dict', 'keys_schema': {'type': 'str'}, 'values_schema': {'type': 'any'}}]}, 'default': None}, 'kw_only': False, 'init': True, 'metadata': {'pydantic_js_updates': {'description': 'Prompt for conditional generation.'}}}], 'type': 'dataclass-args'}, 'slots': True, 'type': 'dataclass'}¶

__pydantic_decorators__ = DecoratorInfos(validators={}, field_validators={}, root_validators={}, field_serializers={}, model_serializers={}, model_validators={}, computed_fields={})¶

__pydantic_fields__ = {'algorithm_version': FieldInfo(annotation=str, required=False, default='CompVis/ldm-celebahq-256', init=True, kw_only=False), 'modality': FieldInfo(annotation=str, required=False, default='image', init=True, kw_only=False), 'model_type': FieldInfo(annotation=str, required=False, default='latent_diffusion', init=True, kw_only=False), 'prompt': FieldInfo(annotation=Union[str, Dict[str, Any]], required=False, default=None, description='Prompt for conditional generation.', init=True, kw_only=False), 'scheduler_type': FieldInfo(annotation=str, required=False, default='discrete', init=True, kw_only=False)}¶

classmethod __pydantic_fields_complete__()¶

Return whether the fields were successfully collected (i.e. type hints were successfully resolved).

This is a private helper, not meant to be used outside Pydantic.

Return type: bool

__pydantic_serializer__ = SchemaSerializer(serializer=PolymorphismTrampoline( PolymorphismTrampoline { class: Py( 0x0000563c9ca0d7d0, ), serializer: PolymorphismTrampoline( PolymorphismTrampoline { class: Py( 0x0000563c9ca0d7d0, ), serializer: Dataclass( DataclassSerializer { class: Py( 0x0000563c9ca0d7d0, ), serializer: Fields( GeneralFieldsSerializer { fields: { "model_type": SerField { key: "model_type", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f215f60a010, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "prompt": SerField { key: "prompt", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x0000563c55c10840, ), ), serializer: Union( UnionSerializer { choices: UnionChoices { choices: [ Str( StrSerializer, ), Dict( DictSerializer { key_serializer: Str( StrSerializer, ), value_serializer: Any( AnySerializer, ), filter: SchemaFilter { include: None, exclude: None, }, name: "dict[str, any]", }, ), ], }, name: "Union[str, dict[str, any]]", }, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "scheduler_type": SerField { key: "scheduler_type", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f216b250370, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "modality": SerField { key: "modality", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f222e9acfb0, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "algorithm_version": SerField { key: "algorithm_version", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f214c69f280, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, }, computed_fields: Some( ComputedFields( [], ), ), mode: SimpleDict, extra_serializer: None, filter: SchemaFilter { include: None, exclude: None, }, required_fields: 5, }, ), fields: [ Py( 0x00007f222affaf10, ), Py( 0x00007f2166fb8ef0, ), Py( 0x00007f222ba93fb0, ), Py( 0x00007f214c6b59b0, ), Py( 0x00007f222ef49930, ), ], name: "LDMGenerator", }, ), enabled_from_config: false, }, ), enabled_from_config: false, }, ), definitions=[])¶

__pydantic_validator__ = SchemaValidator(title="LDMGenerator", validator=Dataclass( DataclassValidator { strict: false, validator: DataclassArgs( DataclassArgsValidator { fields: [ Field { kw_only: false, name: "algorithm_version", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "algorithm_version", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f214c69f280, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "modality", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "modality", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f222e9acfb0, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "model_type", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "model_type", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f215f60a010, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "scheduler_type", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "scheduler_type", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f216b250370, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "prompt", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "prompt", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x0000563c55c10840, ), ), on_error: Raise, validator: Union( UnionValidator { mode: Smart, choices: [ ( Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), None, ), ( Dict( DictValidator { strict: false, key_validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), value_validator: Any( AnyValidator, ), min_length: None, max_length: None, fail_fast: false, name: "dict[str,any]", }, ), None, ), ], custom_error: None, name: "union[str,dict[str,any]]", }, ), validate_default: false, copy_default: false, name: "default[union[str,dict[str,any]]]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, ], positional_count: 5, init_only_count: None, dataclass_name: "LDMGenerator", validator_name: "dataclass-args[LDMGenerator]", extra_behavior: Ignore, extras_validator: None, loc_by_alias: true, validate_by_alias: None, validate_by_name: None, }, ), class: Py( 0x0000563c9ca0d7d0, ), generic_origin: None, fields: [ Py( 0x00007f222affaf10, ), Py( 0x00007f2166fb8ef0, ), Py( 0x00007f222ba93fb0, ), Py( 0x00007f214c6b59b0, ), Py( 0x00007f222ef49930, ), ], post_init: None, revalidate: Never, name: "LDMGenerator", frozen: false, slots: true, }, ), definitions=[], cache_strings=True)¶

__repr__()¶: Return repr(self).

__signature__ = <Signature (*args: Any, algorithm_version: str = 'CompVis/ldm-celebahq-256', modality: str = 'image', model_type: str = 'latent_diffusion', scheduler_type: str = 'discrete', prompt: Union[str, Dict[str, Any]] = None) -> None>¶

__wrapped__¶: alias of LDMGenerator

algorithm_application: ClassVar[str] = 'LDMGenerator'¶

Unique name for the application that is the use of this configuration together with a specific algorithm.

Will be set when registering to ApplicationsRegistry, but can be given by direct registration (See register_algorithm_application)

algorithm_name: ClassVar[str] = 'DiffusersGenerationAlgorithm'¶

Name of the algorithm to use with this configuration.

Will be set when registering to ApplicationsRegistry

class ScoreSdeGenerator(*args, **kwargs)[source]¶

Bases: ScoreSdeGenerator

Score SDE Generative Model - Configuration to generate using a score-based diffusion generative model.

algorithm_version: str = 'google/ncsnpp-celebahq-256'¶

To differentiate between different versions of an application.

There is no imposed naming convention.

model_type: str = 'score_sde'¶

scheduler_type: str = 'continuous'¶

modality: str = 'image'¶

classmethod list_versions()[source]¶

Get possible algorithm versions.

Standard S3 and cache search adding the version used in the configuration.

Return type: Set[str]
Returns: viable values as algorithm_version for the environment.

__annotations__ = {'algorithm_application': 'ClassVar[str]', 'algorithm_name': 'ClassVar[str]', 'algorithm_type': 'ClassVar[str]', 'algorithm_version': <class 'str'>, 'domain': 'ClassVar[str]', 'modality': <class 'str'>, 'model_type': <class 'str'>, 'prompt': 'Union[str, Dict[str, Any]]', 'scheduler_type': <class 'str'>}¶

__dataclass_fields__ = {'algorithm_application': Field(name='algorithm_application',type=typing.ClassVar[str],default='ScoreSdeGenerator',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_name': Field(name='algorithm_name',type=typing.ClassVar[str],default='DiffusersGenerationAlgorithm',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_type': Field(name='algorithm_type',type=typing.ClassVar[str],default='generation',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_version': Field(name='algorithm_version',type=<class 'str'>,default='google/ncsnpp-celebahq-256',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'domain': Field(name='domain',type=typing.ClassVar[str],default='vision',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'modality': Field(name='modality',type=<class 'str'>,default='image',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'model_type': Field(name='model_type',type=<class 'str'>,default='score_sde',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'prompt': Field(name='prompt',type=typing.Union[str, typing.Dict[str, typing.Any]],default=None,default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({'description': 'Prompt for conditional generation.'}),kw_only=False,_field_type=_FIELD), 'scheduler_type': Field(name='scheduler_type',type=<class 'str'>,default='continuous',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD)}¶

__dataclass_params__ = _DataclassParams(init=True,repr=True,eq=True,order=False,unsafe_hash=False,frozen=False)¶

__doc__ = 'Score SDE Generative Model - Configuration to generate using a score-based diffusion generative model.'¶

__eq__(other)¶: Return self==value.

__hash__ = None¶

__init__(*args, **kwargs)¶

__is_pydantic_dataclass__ = True¶

__match_args__ = ('algorithm_version', 'modality', 'model_type', 'scheduler_type', 'prompt')¶

__module__ = 'gt4sd.algorithms.generation.diffusion.core'¶

__parameters__ = (~T,)¶

__pydantic_complete__ = True¶

__pydantic_config__ = {}¶

__pydantic_core_schema__ = {'cls': <class 'gt4sd.algorithms.generation.diffusion.core.ScoreSdeGenerator'>, 'config': {'title': 'ScoreSdeGenerator'}, 'fields': ['algorithm_version', 'modality', 'model_type', 'scheduler_type', 'prompt'], 'frozen': False, 'post_init': False, 'ref': 'types.ScoreSdeGenerator:94818325823488', 'schema': {'collect_init_only': False, 'computed_fields': [], 'dataclass_name': 'ScoreSdeGenerator', 'fields': [{'type': 'dataclass-field', 'name': 'algorithm_version', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'google/ncsnpp-celebahq-256'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'modality', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'image'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'model_type', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'score_sde'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'scheduler_type', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'continuous'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'prompt', 'schema': {'type': 'default', 'schema': {'type': 'union', 'choices': [{'type': 'str'}, {'type': 'dict', 'keys_schema': {'type': 'str'}, 'values_schema': {'type': 'any'}}]}, 'default': None}, 'kw_only': False, 'init': True, 'metadata': {'pydantic_js_updates': {'description': 'Prompt for conditional generation.'}}}], 'type': 'dataclass-args'}, 'slots': True, 'type': 'dataclass'}¶

__pydantic_decorators__ = DecoratorInfos(validators={}, field_validators={}, root_validators={}, field_serializers={}, model_serializers={}, model_validators={}, computed_fields={})¶

__pydantic_fields__ = {'algorithm_version': FieldInfo(annotation=str, required=False, default='google/ncsnpp-celebahq-256', init=True, kw_only=False), 'modality': FieldInfo(annotation=str, required=False, default='image', init=True, kw_only=False), 'model_type': FieldInfo(annotation=str, required=False, default='score_sde', init=True, kw_only=False), 'prompt': FieldInfo(annotation=Union[str, Dict[str, Any]], required=False, default=None, description='Prompt for conditional generation.', init=True, kw_only=False), 'scheduler_type': FieldInfo(annotation=str, required=False, default='continuous', init=True, kw_only=False)}¶

classmethod __pydantic_fields_complete__()¶

Return whether the fields were successfully collected (i.e. type hints were successfully resolved).

This is a private helper, not meant to be used outside Pydantic.

Return type: bool

__pydantic_serializer__ = SchemaSerializer(serializer=PolymorphismTrampoline( PolymorphismTrampoline { class: Py( 0x0000563c9ca10000, ), serializer: PolymorphismTrampoline( PolymorphismTrampoline { class: Py( 0x0000563c9ca10000, ), serializer: Dataclass( DataclassSerializer { class: Py( 0x0000563c9ca10000, ), serializer: Fields( GeneralFieldsSerializer { fields: { "algorithm_version": SerField { key: "algorithm_version", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f214c69f370, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "modality": SerField { key: "modality", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f222e9acfb0, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "scheduler_type": SerField { key: "scheduler_type", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f222cf16830, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "model_type": SerField { key: "model_type", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f214c6b5a70, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "prompt": SerField { key: "prompt", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x0000563c55c10840, ), ), serializer: Union( UnionSerializer { choices: UnionChoices { choices: [ Str( StrSerializer, ), Dict( DictSerializer { key_serializer: Str( StrSerializer, ), value_serializer: Any( AnySerializer, ), filter: SchemaFilter { include: None, exclude: None, }, name: "dict[str, any]", }, ), ], }, name: "Union[str, dict[str, any]]", }, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, }, computed_fields: Some( ComputedFields( [], ), ), mode: SimpleDict, extra_serializer: None, filter: SchemaFilter { include: None, exclude: None, }, required_fields: 5, }, ), fields: [ Py( 0x00007f222affaf10, ), Py( 0x00007f2166fb8ef0, ), Py( 0x00007f222ba93fb0, ), Py( 0x00007f214c6b59b0, ), Py( 0x00007f222ef49930, ), ], name: "ScoreSdeGenerator", }, ), enabled_from_config: false, }, ), enabled_from_config: false, }, ), definitions=[])¶

__pydantic_validator__ = SchemaValidator(title="ScoreSdeGenerator", validator=Dataclass( DataclassValidator { strict: false, validator: DataclassArgs( DataclassArgsValidator { fields: [ Field { kw_only: false, name: "algorithm_version", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "algorithm_version", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f214c69f370, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "modality", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "modality", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f222e9acfb0, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "model_type", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "model_type", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f214c6b5a70, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "scheduler_type", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "scheduler_type", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f222cf16830, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "prompt", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "prompt", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x0000563c55c10840, ), ), on_error: Raise, validator: Union( UnionValidator { mode: Smart, choices: [ ( Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), None, ), ( Dict( DictValidator { strict: false, key_validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), value_validator: Any( AnyValidator, ), min_length: None, max_length: None, fail_fast: false, name: "dict[str,any]", }, ), None, ), ], custom_error: None, name: "union[str,dict[str,any]]", }, ), validate_default: false, copy_default: false, name: "default[union[str,dict[str,any]]]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, ], positional_count: 5, init_only_count: None, dataclass_name: "ScoreSdeGenerator", validator_name: "dataclass-args[ScoreSdeGenerator]", extra_behavior: Ignore, extras_validator: None, loc_by_alias: true, validate_by_alias: None, validate_by_name: None, }, ), class: Py( 0x0000563c9ca10000, ), generic_origin: None, fields: [ Py( 0x00007f222affaf10, ), Py( 0x00007f2166fb8ef0, ), Py( 0x00007f222ba93fb0, ), Py( 0x00007f214c6b59b0, ), Py( 0x00007f222ef49930, ), ], post_init: None, revalidate: Never, name: "ScoreSdeGenerator", frozen: false, slots: true, }, ), definitions=[], cache_strings=True)¶

__repr__()¶: Return repr(self).

__signature__ = <Signature (*args: Any, algorithm_version: str = 'google/ncsnpp-celebahq-256', modality: str = 'image', model_type: str = 'score_sde', scheduler_type: str = 'continuous', prompt: Union[str, Dict[str, Any]] = None) -> None>¶

__wrapped__¶: alias of ScoreSdeGenerator

algorithm_application: ClassVar[str] = 'ScoreSdeGenerator'¶

Unique name for the application that is the use of this configuration together with a specific algorithm.

Will be set when registering to ApplicationsRegistry, but can be given by direct registration (See register_algorithm_application)

algorithm_name: ClassVar[str] = 'DiffusersGenerationAlgorithm'¶

Name of the algorithm to use with this configuration.

Will be set when registering to ApplicationsRegistry

class LDMTextToImageGenerator(*args, **kwargs)[source]¶

Bases: LDMTextToImageGenerator

Conditional Latent Diffusion Model - Configuration for conditional text2image generation using a latent diffusion model.

algorithm_version: str = 'CompVis/ldm-text2im-large-256'¶

To differentiate between different versions of an application.

There is no imposed naming convention.

model_type: str = 'latent_diffusion_conditional'¶

scheduler_type: str = 'discrete'¶

modality: str = 'token2image'¶

classmethod list_versions()[source]¶

Get possible algorithm versions.

Standard S3 and cache search adding the version used in the configuration.

Return type: Set[str]
Returns: viable values as algorithm_version for the environment.

__annotations__ = {'algorithm_application': 'ClassVar[str]', 'algorithm_name': 'ClassVar[str]', 'algorithm_type': 'ClassVar[str]', 'algorithm_version': <class 'str'>, 'domain': 'ClassVar[str]', 'modality': <class 'str'>, 'model_type': <class 'str'>, 'prompt': 'Union[str, Dict[str, Any]]', 'scheduler_type': <class 'str'>}¶

__dataclass_fields__ = {'algorithm_application': Field(name='algorithm_application',type=typing.ClassVar[str],default='LDMTextToImageGenerator',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_name': Field(name='algorithm_name',type=typing.ClassVar[str],default='DiffusersGenerationAlgorithm',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_type': Field(name='algorithm_type',type=typing.ClassVar[str],default='generation',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_version': Field(name='algorithm_version',type=<class 'str'>,default='CompVis/ldm-text2im-large-256',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'domain': Field(name='domain',type=typing.ClassVar[str],default='vision',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'modality': Field(name='modality',type=<class 'str'>,default='token2image',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'model_type': Field(name='model_type',type=<class 'str'>,default='latent_diffusion_conditional',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'prompt': Field(name='prompt',type=typing.Union[str, typing.Dict[str, typing.Any]],default=None,default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({'description': 'Prompt for conditional generation.'}),kw_only=False,_field_type=_FIELD), 'scheduler_type': Field(name='scheduler_type',type=<class 'str'>,default='discrete',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD)}¶

__dataclass_params__ = _DataclassParams(init=True,repr=True,eq=True,order=False,unsafe_hash=False,frozen=False)¶

__doc__ = 'Conditional Latent Diffusion Model - Configuration for conditional text2image generation using a latent diffusion model.'¶

__eq__(other)¶: Return self==value.

__hash__ = None¶

__init__(*args, **kwargs)¶

__is_pydantic_dataclass__ = True¶

__match_args__ = ('algorithm_version', 'modality', 'model_type', 'scheduler_type', 'prompt')¶

__module__ = 'gt4sd.algorithms.generation.diffusion.core'¶

__parameters__ = (~T,)¶

__pydantic_complete__ = True¶

__pydantic_config__ = {}¶

__pydantic_core_schema__ = {'cls': <class 'gt4sd.algorithms.generation.diffusion.core.LDMTextToImageGenerator'>, 'config': {'title': 'LDMTextToImageGenerator'}, 'fields': ['algorithm_version', 'modality', 'model_type', 'scheduler_type', 'prompt'], 'frozen': False, 'post_init': False, 'ref': 'types.LDMTextToImageGenerator:94818326240304', 'schema': {'collect_init_only': False, 'computed_fields': [], 'dataclass_name': 'LDMTextToImageGenerator', 'fields': [{'type': 'dataclass-field', 'name': 'algorithm_version', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'CompVis/ldm-text2im-large-256'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'modality', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'token2image'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'model_type', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'latent_diffusion_conditional'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'scheduler_type', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'discrete'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'prompt', 'schema': {'type': 'default', 'schema': {'type': 'union', 'choices': [{'type': 'str'}, {'type': 'dict', 'keys_schema': {'type': 'str'}, 'values_schema': {'type': 'any'}}]}, 'default': None}, 'kw_only': False, 'init': True, 'metadata': {'pydantic_js_updates': {'description': 'Prompt for conditional generation.'}}}], 'type': 'dataclass-args'}, 'slots': True, 'type': 'dataclass'}¶

__pydantic_decorators__ = DecoratorInfos(validators={}, field_validators={}, root_validators={}, field_serializers={}, model_serializers={}, model_validators={}, computed_fields={})¶

__pydantic_fields__ = {'algorithm_version': FieldInfo(annotation=str, required=False, default='CompVis/ldm-text2im-large-256', init=True, kw_only=False), 'modality': FieldInfo(annotation=str, required=False, default='token2image', init=True, kw_only=False), 'model_type': FieldInfo(annotation=str, required=False, default='latent_diffusion_conditional', init=True, kw_only=False), 'prompt': FieldInfo(annotation=Union[str, Dict[str, Any]], required=False, default=None, description='Prompt for conditional generation.', init=True, kw_only=False), 'scheduler_type': FieldInfo(annotation=str, required=False, default='discrete', init=True, kw_only=False)}¶

classmethod __pydantic_fields_complete__()¶

Return whether the fields were successfully collected (i.e. type hints were successfully resolved).

This is a private helper, not meant to be used outside Pydantic.

Return type: bool

__pydantic_serializer__ = SchemaSerializer(serializer=PolymorphismTrampoline( PolymorphismTrampoline { class: Py( 0x0000563c9ca75c30, ), serializer: PolymorphismTrampoline( PolymorphismTrampoline { class: Py( 0x0000563c9ca75c30, ), serializer: Dataclass( DataclassSerializer { class: Py( 0x0000563c9ca75c30, ), serializer: Fields( GeneralFieldsSerializer { fields: { "scheduler_type": SerField { key: "scheduler_type", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f216b250370, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "algorithm_version": SerField { key: "algorithm_version", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f214c69f410, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "modality": SerField { key: "modality", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f214c6b5ab0, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "prompt": SerField { key: "prompt", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x0000563c55c10840, ), ), serializer: Union( UnionSerializer { choices: UnionChoices { choices: [ Str( StrSerializer, ), Dict( DictSerializer { key_serializer: Str( StrSerializer, ), value_serializer: Any( AnySerializer, ), filter: SchemaFilter { include: None, exclude: None, }, name: "dict[str, any]", }, ), ], }, name: "Union[str, dict[str, any]]", }, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "model_type": SerField { key: "model_type", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f214c69f460, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, }, computed_fields: Some( ComputedFields( [], ), ), mode: SimpleDict, extra_serializer: None, filter: SchemaFilter { include: None, exclude: None, }, required_fields: 5, }, ), fields: [ Py( 0x00007f222affaf10, ), Py( 0x00007f2166fb8ef0, ), Py( 0x00007f222ba93fb0, ), Py( 0x00007f214c6b59b0, ), Py( 0x00007f222ef49930, ), ], name: "LDMTextToImageGenerator", }, ), enabled_from_config: false, }, ), enabled_from_config: false, }, ), definitions=[])¶

__pydantic_validator__ = SchemaValidator(title="LDMTextToImageGenerator", validator=Dataclass( DataclassValidator { strict: false, validator: DataclassArgs( DataclassArgsValidator { fields: [ Field { kw_only: false, name: "algorithm_version", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "algorithm_version", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f214c69f410, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "modality", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "modality", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f214c6b5ab0, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "model_type", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "model_type", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f214c69f460, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "scheduler_type", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "scheduler_type", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f216b250370, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "prompt", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "prompt", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x0000563c55c10840, ), ), on_error: Raise, validator: Union( UnionValidator { mode: Smart, choices: [ ( Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), None, ), ( Dict( DictValidator { strict: false, key_validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), value_validator: Any( AnyValidator, ), min_length: None, max_length: None, fail_fast: false, name: "dict[str,any]", }, ), None, ), ], custom_error: None, name: "union[str,dict[str,any]]", }, ), validate_default: false, copy_default: false, name: "default[union[str,dict[str,any]]]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, ], positional_count: 5, init_only_count: None, dataclass_name: "LDMTextToImageGenerator", validator_name: "dataclass-args[LDMTextToImageGenerator]", extra_behavior: Ignore, extras_validator: None, loc_by_alias: true, validate_by_alias: None, validate_by_name: None, }, ), class: Py( 0x0000563c9ca75c30, ), generic_origin: None, fields: [ Py( 0x00007f222affaf10, ), Py( 0x00007f2166fb8ef0, ), Py( 0x00007f222ba93fb0, ), Py( 0x00007f214c6b59b0, ), Py( 0x00007f222ef49930, ), ], post_init: None, revalidate: Never, name: "LDMTextToImageGenerator", frozen: false, slots: true, }, ), definitions=[], cache_strings=True)¶

__repr__()¶: Return repr(self).

__signature__ = <Signature (*args: Any, algorithm_version: str = 'CompVis/ldm-text2im-large-256', modality: str = 'token2image', model_type: str = 'latent_diffusion_conditional', scheduler_type: str = 'discrete', prompt: Union[str, Dict[str, Any]] = None) -> None>¶

__wrapped__¶: alias of LDMTextToImageGenerator

algorithm_application: ClassVar[str] = 'LDMTextToImageGenerator'¶

Unique name for the application that is the use of this configuration together with a specific algorithm.

Will be set when registering to ApplicationsRegistry, but can be given by direct registration (See register_algorithm_application)

algorithm_name: ClassVar[str] = 'DiffusersGenerationAlgorithm'¶

Name of the algorithm to use with this configuration.

Will be set when registering to ApplicationsRegistry

class StableDiffusionGenerator(*args, **kwargs)[source]¶

Bases: StableDiffusionGenerator

Stable Diffusion Model - Configuration for conditional text2image generation using a stable diffusion model. You have to provide authentication credentials to use this model.

algorithm_version: str = 'CompVis/stable-diffusion-v1-4'¶

To differentiate between different versions of an application.

There is no imposed naming convention.

model_type: str = 'stable_diffusion'¶

scheduler_type: str = 'discrete'¶

modality: str = 'token2image'¶

classmethod list_versions()[source]¶

Get possible algorithm versions.

Standard S3 and cache search adding the version used in the configuration.

Return type: Set[str]
Returns: viable values as algorithm_version for the environment.

__annotations__ = {'algorithm_application': 'ClassVar[str]', 'algorithm_name': 'ClassVar[str]', 'algorithm_type': 'ClassVar[str]', 'algorithm_version': <class 'str'>, 'domain': 'ClassVar[str]', 'modality': <class 'str'>, 'model_type': <class 'str'>, 'prompt': 'Union[str, Dict[str, Any]]', 'scheduler_type': <class 'str'>}¶

__dataclass_fields__ = {'algorithm_application': Field(name='algorithm_application',type=typing.ClassVar[str],default='StableDiffusionGenerator',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_name': Field(name='algorithm_name',type=typing.ClassVar[str],default='DiffusersGenerationAlgorithm',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_type': Field(name='algorithm_type',type=typing.ClassVar[str],default='generation',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_version': Field(name='algorithm_version',type=<class 'str'>,default='CompVis/stable-diffusion-v1-4',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'domain': Field(name='domain',type=typing.ClassVar[str],default='vision',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'modality': Field(name='modality',type=<class 'str'>,default='token2image',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'model_type': Field(name='model_type',type=<class 'str'>,default='stable_diffusion',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'prompt': Field(name='prompt',type=typing.Union[str, typing.Dict[str, typing.Any]],default=None,default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({'description': 'Prompt for conditional generation.'}),kw_only=False,_field_type=_FIELD), 'scheduler_type': Field(name='scheduler_type',type=<class 'str'>,default='discrete',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD)}¶

__dataclass_params__ = _DataclassParams(init=True,repr=True,eq=True,order=False,unsafe_hash=False,frozen=False)¶

__doc__ = 'Stable Diffusion Model - Configuration for conditional text2image generation using a stable diffusion model.\n You have to provide authentication credentials to use this model.\n '¶

__eq__(other)¶: Return self==value.

__hash__ = None¶

__init__(*args, **kwargs)¶

__is_pydantic_dataclass__ = True¶

__match_args__ = ('algorithm_version', 'modality', 'model_type', 'scheduler_type', 'prompt')¶

__module__ = 'gt4sd.algorithms.generation.diffusion.core'¶

__parameters__ = (~T,)¶

__pydantic_complete__ = True¶

__pydantic_config__ = {}¶

__pydantic_core_schema__ = {'cls': <class 'gt4sd.algorithms.generation.diffusion.core.StableDiffusionGenerator'>, 'config': {'title': 'StableDiffusionGenerator'}, 'fields': ['algorithm_version', 'modality', 'model_type', 'scheduler_type', 'prompt'], 'frozen': False, 'post_init': False, 'ref': 'types.StableDiffusionGenerator:94818326243024', 'schema': {'collect_init_only': False, 'computed_fields': [], 'dataclass_name': 'StableDiffusionGenerator', 'fields': [{'type': 'dataclass-field', 'name': 'algorithm_version', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'CompVis/stable-diffusion-v1-4'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'modality', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'token2image'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'model_type', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'stable_diffusion'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'scheduler_type', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'discrete'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'prompt', 'schema': {'type': 'default', 'schema': {'type': 'union', 'choices': [{'type': 'str'}, {'type': 'dict', 'keys_schema': {'type': 'str'}, 'values_schema': {'type': 'any'}}]}, 'default': None}, 'kw_only': False, 'init': True, 'metadata': {'pydantic_js_updates': {'description': 'Prompt for conditional generation.'}}}], 'type': 'dataclass-args'}, 'slots': True, 'type': 'dataclass'}¶

__pydantic_decorators__ = DecoratorInfos(validators={}, field_validators={}, root_validators={}, field_serializers={}, model_serializers={}, model_validators={}, computed_fields={})¶

__pydantic_fields__ = {'algorithm_version': FieldInfo(annotation=str, required=False, default='CompVis/stable-diffusion-v1-4', init=True, kw_only=False), 'modality': FieldInfo(annotation=str, required=False, default='token2image', init=True, kw_only=False), 'model_type': FieldInfo(annotation=str, required=False, default='stable_diffusion', init=True, kw_only=False), 'prompt': FieldInfo(annotation=Union[str, Dict[str, Any]], required=False, default=None, description='Prompt for conditional generation.', init=True, kw_only=False), 'scheduler_type': FieldInfo(annotation=str, required=False, default='discrete', init=True, kw_only=False)}¶

classmethod __pydantic_fields_complete__()¶

Return whether the fields were successfully collected (i.e. type hints were successfully resolved).

This is a private helper, not meant to be used outside Pydantic.

Return type: bool

__pydantic_serializer__ = SchemaSerializer(serializer=PolymorphismTrampoline( PolymorphismTrampoline { class: Py( 0x0000563c9ca766d0, ), serializer: PolymorphismTrampoline( PolymorphismTrampoline { class: Py( 0x0000563c9ca766d0, ), serializer: Dataclass( DataclassSerializer { class: Py( 0x0000563c9ca766d0, ), serializer: Fields( GeneralFieldsSerializer { fields: { "modality": SerField { key: "modality", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f214c6b5ab0, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "model_type": SerField { key: "model_type", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f215f60a060, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "prompt": SerField { key: "prompt", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x0000563c55c10840, ), ), serializer: Union( UnionSerializer { choices: UnionChoices { choices: [ Str( StrSerializer, ), Dict( DictSerializer { key_serializer: Str( StrSerializer, ), value_serializer: Any( AnySerializer, ), filter: SchemaFilter { include: None, exclude: None, }, name: "dict[str, any]", }, ), ], }, name: "Union[str, dict[str, any]]", }, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "scheduler_type": SerField { key: "scheduler_type", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f216b250370, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "algorithm_version": SerField { key: "algorithm_version", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f214c69f4b0, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, }, computed_fields: Some( ComputedFields( [], ), ), mode: SimpleDict, extra_serializer: None, filter: SchemaFilter { include: None, exclude: None, }, required_fields: 5, }, ), fields: [ Py( 0x00007f222affaf10, ), Py( 0x00007f2166fb8ef0, ), Py( 0x00007f222ba93fb0, ), Py( 0x00007f214c6b59b0, ), Py( 0x00007f222ef49930, ), ], name: "StableDiffusionGenerator", }, ), enabled_from_config: false, }, ), enabled_from_config: false, }, ), definitions=[])¶

__pydantic_validator__ = SchemaValidator(title="StableDiffusionGenerator", validator=Dataclass( DataclassValidator { strict: false, validator: DataclassArgs( DataclassArgsValidator { fields: [ Field { kw_only: false, name: "algorithm_version", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "algorithm_version", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f214c69f4b0, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "modality", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "modality", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f214c6b5ab0, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "model_type", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "model_type", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f215f60a060, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "scheduler_type", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "scheduler_type", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f216b250370, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "prompt", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "prompt", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x0000563c55c10840, ), ), on_error: Raise, validator: Union( UnionValidator { mode: Smart, choices: [ ( Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), None, ), ( Dict( DictValidator { strict: false, key_validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), value_validator: Any( AnyValidator, ), min_length: None, max_length: None, fail_fast: false, name: "dict[str,any]", }, ), None, ), ], custom_error: None, name: "union[str,dict[str,any]]", }, ), validate_default: false, copy_default: false, name: "default[union[str,dict[str,any]]]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, ], positional_count: 5, init_only_count: None, dataclass_name: "StableDiffusionGenerator", validator_name: "dataclass-args[StableDiffusionGenerator]", extra_behavior: Ignore, extras_validator: None, loc_by_alias: true, validate_by_alias: None, validate_by_name: None, }, ), class: Py( 0x0000563c9ca766d0, ), generic_origin: None, fields: [ Py( 0x00007f222affaf10, ), Py( 0x00007f2166fb8ef0, ), Py( 0x00007f222ba93fb0, ), Py( 0x00007f214c6b59b0, ), Py( 0x00007f222ef49930, ), ], post_init: None, revalidate: Never, name: "StableDiffusionGenerator", frozen: false, slots: true, }, ), definitions=[], cache_strings=True)¶

__repr__()¶: Return repr(self).

__signature__ = <Signature (*args: Any, algorithm_version: str = 'CompVis/stable-diffusion-v1-4', modality: str = 'token2image', model_type: str = 'stable_diffusion', scheduler_type: str = 'discrete', prompt: Union[str, Dict[str, Any]] = None) -> None>¶

__wrapped__¶: alias of StableDiffusionGenerator

algorithm_application: ClassVar[str] = 'StableDiffusionGenerator'¶

Unique name for the application that is the use of this configuration together with a specific algorithm.

Will be set when registering to ApplicationsRegistry, but can be given by direct registration (See register_algorithm_application)

algorithm_name: ClassVar[str] = 'DiffusersGenerationAlgorithm'¶

Name of the algorithm to use with this configuration.

Will be set when registering to ApplicationsRegistry

class GeoDiffGenerator(*args, **kwargs)[source]¶

Bases: GeoDiffGenerator

GeoDiff Diffusion Model - Configuration for conditional 3D molecule structure generation given 2D information using a GeoDiff diffusion model.

algorithm_version: str = 'fusing/gfn-molecule-gen-drugs'¶

To differentiate between different versions of an application.

There is no imposed naming convention.

model_type: str = 'geodiff'¶

scheduler_type: str = 'ddpm'¶

modality: str = 'molecule'¶

classmethod list_versions()[source]¶

Get possible algorithm versions.

Standard S3 and cache search adding the version used in the configuration.

Return type: Set[str]
Returns: viable values as algorithm_version for the environment.

get_target_description()[source]¶

Get description of the target for generation.

Return type: Optional[Dict[str, str], None]
Returns: target description, returns None in case no target is used.

__annotations__ = {'algorithm_application': 'ClassVar[str]', 'algorithm_name': 'ClassVar[str]', 'algorithm_type': 'ClassVar[str]', 'algorithm_version': <class 'str'>, 'domain': 'ClassVar[str]', 'modality': <class 'str'>, 'model_type': <class 'str'>, 'prompt': 'Union[str, Dict[str, Any]]', 'scheduler_type': <class 'str'>}¶

__dataclass_fields__ = {'algorithm_application': Field(name='algorithm_application',type=typing.ClassVar[str],default='GeoDiffGenerator',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_name': Field(name='algorithm_name',type=typing.ClassVar[str],default='DiffusersGenerationAlgorithm',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_type': Field(name='algorithm_type',type=typing.ClassVar[str],default='generation',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'algorithm_version': Field(name='algorithm_version',type=<class 'str'>,default='fusing/gfn-molecule-gen-drugs',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'domain': Field(name='domain',type=typing.ClassVar[str],default='vision',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=<dataclasses._MISSING_TYPE object>,_field_type=_FIELD_CLASSVAR), 'modality': Field(name='modality',type=<class 'str'>,default='molecule',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'model_type': Field(name='model_type',type=<class 'str'>,default='geodiff',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD), 'prompt': Field(name='prompt',type=typing.Union[str, typing.Dict[str, typing.Any]],default=None,default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({'description': 'Prompt for conditional generation.'}),kw_only=False,_field_type=_FIELD), 'scheduler_type': Field(name='scheduler_type',type=<class 'str'>,default='ddpm',default_factory=<dataclasses._MISSING_TYPE object>,init=True,repr=True,hash=None,compare=True,metadata=mappingproxy({}),kw_only=False,_field_type=_FIELD)}¶

__dataclass_params__ = _DataclassParams(init=True,repr=True,eq=True,order=False,unsafe_hash=False,frozen=False)¶

__doc__ = 'GeoDiff Diffusion Model - Configuration for conditional 3D molecule structure generation given 2D information using a GeoDiff diffusion model.'¶

__eq__(other)¶: Return self==value.

__hash__ = None¶

__init__(*args, **kwargs)¶

__is_pydantic_dataclass__ = True¶

__match_args__ = ('algorithm_version', 'modality', 'model_type', 'scheduler_type', 'prompt')¶

__module__ = 'gt4sd.algorithms.generation.diffusion.core'¶

__parameters__ = (~T,)¶

__pydantic_complete__ = True¶

__pydantic_config__ = {}¶

__pydantic_core_schema__ = {'cls': <class 'gt4sd.algorithms.generation.diffusion.core.GeoDiffGenerator'>, 'config': {'title': 'GeoDiffGenerator'}, 'fields': ['algorithm_version', 'modality', 'model_type', 'scheduler_type', 'prompt'], 'frozen': False, 'post_init': False, 'ref': 'types.GeoDiffGenerator:94818326256432', 'schema': {'collect_init_only': False, 'computed_fields': [], 'dataclass_name': 'GeoDiffGenerator', 'fields': [{'type': 'dataclass-field', 'name': 'algorithm_version', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'fusing/gfn-molecule-gen-drugs'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'modality', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'molecule'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'model_type', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'geodiff'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'scheduler_type', 'schema': {'type': 'default', 'schema': {'type': 'str'}, 'default': 'ddpm'}, 'kw_only': False, 'init': True, 'metadata': {}}, {'type': 'dataclass-field', 'name': 'prompt', 'schema': {'type': 'default', 'schema': {'type': 'union', 'choices': [{'type': 'str'}, {'type': 'dict', 'keys_schema': {'type': 'str'}, 'values_schema': {'type': 'any'}}]}, 'default': None}, 'kw_only': False, 'init': True, 'metadata': {'pydantic_js_updates': {'description': 'Prompt for conditional generation.'}}}], 'type': 'dataclass-args'}, 'slots': True, 'type': 'dataclass'}¶

__pydantic_decorators__ = DecoratorInfos(validators={}, field_validators={}, root_validators={}, field_serializers={}, model_serializers={}, model_validators={}, computed_fields={})¶

__pydantic_fields__ = {'algorithm_version': FieldInfo(annotation=str, required=False, default='fusing/gfn-molecule-gen-drugs', init=True, kw_only=False), 'modality': FieldInfo(annotation=str, required=False, default='molecule', init=True, kw_only=False), 'model_type': FieldInfo(annotation=str, required=False, default='geodiff', init=True, kw_only=False), 'prompt': FieldInfo(annotation=Union[str, Dict[str, Any]], required=False, default=None, description='Prompt for conditional generation.', init=True, kw_only=False), 'scheduler_type': FieldInfo(annotation=str, required=False, default='ddpm', init=True, kw_only=False)}¶

classmethod __pydantic_fields_complete__()¶

Return whether the fields were successfully collected (i.e. type hints were successfully resolved).

This is a private helper, not meant to be used outside Pydantic.

Return type: bool

__pydantic_serializer__ = SchemaSerializer(serializer=PolymorphismTrampoline( PolymorphismTrampoline { class: Py( 0x0000563c9ca79b30, ), serializer: PolymorphismTrampoline( PolymorphismTrampoline { class: Py( 0x0000563c9ca79b30, ), serializer: Dataclass( DataclassSerializer { class: Py( 0x0000563c9ca79b30, ), serializer: Fields( GeneralFieldsSerializer { fields: { "scheduler_type": SerField { key: "scheduler_type", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f215f60ddf0, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "prompt": SerField { key: "prompt", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x0000563c55c10840, ), ), serializer: Union( UnionSerializer { choices: UnionChoices { choices: [ Str( StrSerializer, ), Dict( DictSerializer { key_serializer: Str( StrSerializer, ), value_serializer: Any( AnySerializer, ), filter: SchemaFilter { include: None, exclude: None, }, name: "dict[str, any]", }, ), ], }, name: "Union[str, dict[str, any]]", }, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "algorithm_version": SerField { key: "algorithm_version", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f214c69f5a0, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "model_type": SerField { key: "model_type", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f214c6b5b30, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, "modality": SerField { key: "modality", alias: None, serializer: Some( WithDefault( WithDefaultSerializer { default: Default( Py( 0x00007f222aff4230, ), ), serializer: Str( StrSerializer, ), }, ), ), required: true, serialize_by_alias: None, serialization_exclude_if: None, }, }, computed_fields: Some( ComputedFields( [], ), ), mode: SimpleDict, extra_serializer: None, filter: SchemaFilter { include: None, exclude: None, }, required_fields: 5, }, ), fields: [ Py( 0x00007f222affaf10, ), Py( 0x00007f2166fb8ef0, ), Py( 0x00007f222ba93fb0, ), Py( 0x00007f214c6b59b0, ), Py( 0x00007f222ef49930, ), ], name: "GeoDiffGenerator", }, ), enabled_from_config: false, }, ), enabled_from_config: false, }, ), definitions=[])¶

__pydantic_validator__ = SchemaValidator(title="GeoDiffGenerator", validator=Dataclass( DataclassValidator { strict: false, validator: DataclassArgs( DataclassArgsValidator { fields: [ Field { kw_only: false, name: "algorithm_version", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "algorithm_version", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f214c69f5a0, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "modality", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "modality", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f222aff4230, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "model_type", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "model_type", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f214c6b5b30, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "scheduler_type", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "scheduler_type", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x00007f215f60ddf0, ), ), on_error: Raise, validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), validate_default: false, copy_default: false, name: "default[str]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, Field { kw_only: false, name: "prompt", init: true, init_only: false, lookup_path_collection: LookupPathCollection { by_name: LookupPath { first_item: PathItemString( "prompt", ), rest: [], }, by_alias: [], }, validator: WithDefault( WithDefaultValidator { default: Default( Py( 0x0000563c55c10840, ), ), on_error: Raise, validator: Union( UnionValidator { mode: Smart, choices: [ ( Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), None, ), ( Dict( DictValidator { strict: false, key_validator: Str( StrValidator { strict: false, coerce_numbers_to_str: false, }, ), value_validator: Any( AnyValidator, ), min_length: None, max_length: None, fail_fast: false, name: "dict[str,any]", }, ), None, ), ], custom_error: None, name: "union[str,dict[str,any]]", }, ), validate_default: false, copy_default: false, name: "default[union[str,dict[str,any]]]", undefined: Py( 0x00007f222cf13af0, ), }, ), frozen: false, }, ], positional_count: 5, init_only_count: None, dataclass_name: "GeoDiffGenerator", validator_name: "dataclass-args[GeoDiffGenerator]", extra_behavior: Ignore, extras_validator: None, loc_by_alias: true, validate_by_alias: None, validate_by_name: None, }, ), class: Py( 0x0000563c9ca79b30, ), generic_origin: None, fields: [ Py( 0x00007f222affaf10, ), Py( 0x00007f2166fb8ef0, ), Py( 0x00007f222ba93fb0, ), Py( 0x00007f214c6b59b0, ), Py( 0x00007f222ef49930, ), ], post_init: None, revalidate: Never, name: "GeoDiffGenerator", frozen: false, slots: true, }, ), definitions=[], cache_strings=True)¶

__repr__()¶: Return repr(self).

__signature__ = <Signature (*args: Any, algorithm_version: str = 'fusing/gfn-molecule-gen-drugs', modality: str = 'molecule', model_type: str = 'geodiff', scheduler_type: str = 'ddpm', prompt: Union[str, Dict[str, Any]] = None) -> None>¶

__wrapped__¶: alias of GeoDiffGenerator

algorithm_application: ClassVar[str] = 'GeoDiffGenerator'¶

Unique name for the application that is the use of this configuration together with a specific algorithm.

Will be set when registering to ApplicationsRegistry, but can be given by direct registration (See register_algorithm_application)

algorithm_name: ClassVar[str] = 'DiffusersGenerationAlgorithm'¶

Name of the algorithm to use with this configuration.

Will be set when registering to ApplicationsRegistry