data.text_tokenizer package

Submodules

data.text_tokenizer.base_tokenizer module

class data.text_tokenizer.base_tokenizer.BaseTokenizer(opts, *args, **kwargs)

Bases: Module

__init__(opts, *args, **kwargs)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

classmethod add_arguments(parser: ArgumentParser) → ArgumentParser

get_vocab_size()

get_eot_token()

get_sot_token()

get_encodings()

forward(input_sentence: Any, *args, **kwargs) → Any

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

data.text_tokenizer.clip_tokenizer module

class data.text_tokenizer.clip_tokenizer.ClipTokenizer(opts, *args, **kwargs)

Bases: BaseTokenizer

__init__(opts, *args, **kwargs)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

classmethod add_arguments(parser: ArgumentParser) → ArgumentParser

get_vocab_size()

get_encodings()

get_eot_token()

get_sot_token()

forward(input_sentence: str, *args, **kwargs) → Tensor

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

Module contents

data.text_tokenizer.arguments_tokenizer(parser: ArgumentParser) → ArgumentParser

data.text_tokenizer.build_tokenizer(opts, *args, **kwargs) → BaseTokenizer

Helper function to build the text tokenizer from command-line arguments.

Parameters:

opts – Command-line arguments

Returns:

Image projection head module.