kiwi.systems.decoders.nuqe

Module Contents

Classes

NuQETargetDecoder	Base class for all neural network modules.
NuQESourceDecoder	Base class for all neural network modules.
NuQEDecoder	Neural Quality Estimation (NuQE) model for word level quality estimation.

class kiwi.systems.decoders.nuqe.NuQETargetDecoder(input_dim, config)

Bases: torch.nn.Module

Base class for all neural network modules.

Your models should also subclass this class.

Modules can also contain other Modules, allowing to nest them in a tree structure. You can assign the submodules as regular attributes:

import torch.nn as nn
import torch.nn.functional as F

class Model(nn.Module):
    def __init__(self):
        super(Model, self).__init__()
        self.conv1 = nn.Conv2d(1, 20, 5)
        self.conv2 = nn.Conv2d(20, 20, 5)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        return F.relu(self.conv2(x))

Submodules assigned in this way will be registered, and will have their parameters converted too when you call to(), etc.

class Config

Bases: kiwi.utils.io.BaseConfig

Base class for all pydantic configs. Used to configure base behaviour of configs.

hidden_sizes :conlist(int, min_items=4, max_items=4) = [400, 200, 100, 50]

dropout :confloat(ge=0.0, le=1.0) = 0.4

forward(self, features, batch_inputs)

size(self)

class kiwi.systems.decoders.nuqe.NuQESourceDecoder(input_dim, config)

Bases: torch.nn.Module

Base class for all neural network modules.

Your models should also subclass this class.

Modules can also contain other Modules, allowing to nest them in a tree structure. You can assign the submodules as regular attributes:

import torch.nn as nn
import torch.nn.functional as F

class Model(nn.Module):
    def __init__(self):
        super(Model, self).__init__()
        self.conv1 = nn.Conv2d(1, 20, 5)
        self.conv2 = nn.Conv2d(20, 20, 5)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        return F.relu(self.conv2(x))

Submodules assigned in this way will be registered, and will have their parameters converted too when you call to(), etc.

class Config

Bases: kiwi.utils.io.BaseConfig

Base class for all pydantic configs. Used to configure base behaviour of configs.

hidden_sizes :conlist(int, min_items=4, max_items=4) = [400, 200, 100, 50]

dropout :confloat(ge=0.0, le=1.0) = 0.4

forward(self, features, batch_inputs)

size(self)

class kiwi.systems.decoders.nuqe.NuQEDecoder(inputs_dims, config: Config)

Bases: kiwi.systems._meta_module.MetaModule

Neural Quality Estimation (NuQE) model for word level quality estimation.

class Config

Bases: kiwi.utils.io.BaseConfig

Base class for all pydantic configs. Used to configure base behaviour of configs.

target :NuQETargetDecoder.Config

source :NuQESourceDecoder.Config

forward(self, features, batch_inputs)

size(self, field=None)

kiwi.systems.decoders.linear kiwi.systems.encoders