dlk.core package

Subpackages

• dlk.core.callbacks package
  • Submodules
  • dlk.core.callbacks.checkpoint module
  • dlk.core.callbacks.early_stop module
  • dlk.core.callbacks.lr_monitor module
  • dlk.core.callbacks.weight_average module
  • Module contents
• dlk.core.imodels package
  • Submodules
  • dlk.core.imodels.basic module
  • dlk.core.imodels.distill module
  • Module contents
• dlk.core.initmethods package
  • Submodules
  • dlk.core.initmethods.default module
  • dlk.core.initmethods.range_norm module
  • dlk.core.initmethods.range_uniform module
  • Module contents
• dlk.core.layers package
  • Subpackages
    • dlk.core.layers.decoders package
      • Submodules
      • dlk.core.layers.decoders.identity module
      • dlk.core.layers.decoders.linear module
      • dlk.core.layers.decoders.linear_crf module
      • Module contents
    • dlk.core.layers.embeddings package
      • Submodules
      • dlk.core.layers.embeddings.combine_word_char_cnn module
      • dlk.core.layers.embeddings.identity module
      • dlk.core.layers.embeddings.pretrained_transformers module
      • dlk.core.layers.embeddings.random module
      • dlk.core.layers.embeddings.static module
      • dlk.core.layers.embeddings.static_char_cnn module
      • Module contents
    • dlk.core.layers.encoders package
      • Submodules
      • dlk.core.layers.encoders.identity module
      • dlk.core.layers.encoders.linear module
      • dlk.core.layers.encoders.lstm module
      • Module contents
  • Module contents
• dlk.core.losses package
  • Submodules
  • dlk.core.losses.bce module
  • dlk.core.losses.cross_entropy module
  • dlk.core.losses.identity module
  • dlk.core.losses.mse module
  • dlk.core.losses.multi_loss module
  • Module contents
• dlk.core.models package
  • Submodules
  • dlk.core.models.basic module
  • Module contents
• dlk.core.modules package
  • Submodules
  • dlk.core.modules.bert module
  • dlk.core.modules.conv1d module
  • dlk.core.modules.crf module
  • dlk.core.modules.distil_bert module
  • dlk.core.modules.linear module
  • dlk.core.modules.logits_gather module
  • dlk.core.modules.lstm module
  • dlk.core.modules.roberta module
  • Module contents
• dlk.core.optimizers package
  • Submodules
  • dlk.core.optimizers.adamw module
  • dlk.core.optimizers.sgd module
  • Module contents
• dlk.core.schedulers package
  • Submodules
  • dlk.core.schedulers.constant module
  • dlk.core.schedulers.constant_warmup module
  • dlk.core.schedulers.cosine_warmup module
  • dlk.core.schedulers.linear_warmup module
  • dlk.core.schedulers.multi_group_schedule module
  • dlk.core.schedulers.rec_decay module
  • Module contents

Submodules

dlk.core.base_module module

class dlk.core.base_module.BaseModel

Bases: torch.nn.modules.module.Module, dlk.core.base_module.ModuleOutputRenameMixin, dlk.core.base_module.IModuleIO, dlk.core.base_module.IModuleStep

All pytorch models should inheritance this class

forward(inputs: Dict[str, torch.Tensor]) → Dict[str, torch.Tensor]

all models should apply this method

Parameters

inputs – one mini-batch inputs

Returns

one mini-batch outputs

training: bool

class dlk.core.base_module.BaseModule(config: dlk.core.base_module.BaseModuleConfig)

Bases: torch.nn.modules.module.Module, dlk.core.base_module.ModuleOutputRenameMixin, dlk.core.base_module.IModuleIO, dlk.core.base_module.IModuleStep

All pytorch modules should inheritance this class

check_keys_are_provided(provide: Set[str]) → None

check this module required key are provided

Returns

pass or not

forward(inputs: Dict[str, torch.Tensor]) → Dict[str, torch.Tensor]

all module should apply this method

Parameters

inputs – one mini-batch inputs

Returns

one mini-batch outputs

init_weight(method)

init the weight of submodules by ‘method’

Parameters

method – init method

Returns

None

provide_keys() → Set

return all keys of the dict of the module returned

Returns

all keys

training: bool

class dlk.core.base_module.BaseModuleConfig(config: Dict)

Bases: dlk.utils.config.BaseConfig

docstring for BaseLayerConfig

class dlk.core.base_module.IModuleIO

Bases: object

interface for check the modules input and output

abstract check_keys_are_provided(provide: List[str]) → bool

check this module required key are provided

Returns

pass or not

check_module_chain(module_list: List[dlk.core.base_module.BaseModule]) → bool

check the interfaces of the list of modules are alignd or not.

Parameters

module_list – a series modules

Returns

pass or not

Raises

ValueError – the check is not passed

abstract provide_keys() → List[str]

return all keys of the dict of the module returned

Returns

all keys

class dlk.core.base_module.IModuleStep

Bases: object

docstring for ModuleStepMixin

abstract predict_step(inputs: Dict[str, torch.Tensor]) → Dict[str, torch.Tensor]

do predict for one batch

Parameters

inputs – one mini-batch inputs

Returns

the predicts outputs

test_step(inputs: Dict[str, torch.Tensor]) → Dict[str, torch.Tensor]

do test for one batch

Parameters

inputs – one mini-batch inputs

Returns

one mini-batch outputs

abstract training_step(inputs: Dict[str, torch.Tensor]) → Dict[str, torch.Tensor]

do training for one batch

Parameters

inputs – one mini-batch inputs

Returns

one mini-batch outputs

abstract validation_step(inputs: Dict[str, torch.Tensor]) → Dict[str, torch.Tensor]

do validataion for one batch

Parameters

inputs – one mini-batch inputs

Returns

one mini-batch outputs

class dlk.core.base_module.ModuleOutputRenameMixin

Bases: object

Just rename the output key name by config to adapt the input field of downstream module.

dict_rename(input: Dict, output_map: Dict[str, str]) → Dict

rename the key of input(dict) by output_map(name map)

Parameters

• input – will rename input

• output_map – name map

Returns

renamed input

get_input_name(name: str) → str

use config._input_map map the name to real name

Parameters

name – input_name

Returns

real_name

get_output_name(name: str) → str

use config._output_map map the name to real name

Parameters

name – output_name

Returns

real_name

get_real_name(name: str, name_map: Dict[str, str]) → str

use the name_map to map the input name to real name

Parameters

• name – input_name

• name_map – name map

Returns

real_name

set_rename(input: Set, output_map: Dict[str, str]) → Set

rename all the name in input by output_map

Parameters

• input – a set of names

• output_map – name map

Returns

renamed input

class dlk.core.base_module.SimpleModule(config: dlk.core.base_module.BaseModuleConfig)

Bases: dlk.core.base_module.BaseModule

docstring for SimpleModule, SimpleModule, all train/predict/test/validation step call the forward

forward(inputs: Dict[str, torch.Tensor]) → Dict[str, torch.Tensor]

in simple module, all step fit to this method

Parameters

inputs – one mini-batch inputs

Returns

one mini-batch outputs

predict_step(inputs: Dict[str, torch.Tensor]) → Dict[str, torch.Tensor]

do predict for one batch

Parameters

inputs – one mini-batch inputs

Returns

one mini-batch outputs

test_step(inputs: Dict[str, torch.Tensor]) → Dict[str, torch.Tensor]

do test for one batch

Parameters

inputs – one mini-batch inputs

Returns

one mini-batch outputs

training: bool

training_step(inputs: Dict[str, torch.Tensor]) → Dict[str, torch.Tensor]

do train for one batch

Parameters

inputs – one mini-batch inputs

Returns

one mini-batch outputs

validation_step(inputs: Dict[str, torch.Tensor]) → Dict[str, torch.Tensor]

do validation for one batch

Parameters

inputs – one mini-batch inputs

Returns

one mini-batch outputs

Module contents