medcat.utils.meta_cat.models

Module Contents

Classes

LSTM	Base class for all neural network modules.
BertForMetaAnnotation	Base class for all neural network modules.

Attributes

logger

medcat.utils.meta_cat.models.logger

class medcat.utils.meta_cat.models.LSTM(embeddings, 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().__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.

Note

As per the example above, an __init__() call to the parent class must be made before assignment on the child.

Variables:

training (bool) – Boolean represents whether this module is in training or evaluation mode.

Parameters:

• embeddings (Optional[torch.Tensor]) –

• config (medcat.meta_cat.ConfigMetaCAT) –

__init__(embeddings, config)

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

Parameters:

• embeddings (Optional[torch.Tensor]) –

• config (medcat.meta_cat.ConfigMetaCAT) –

Return type:

None

forward(input_ids, center_positions, attention_mask=None, ignore_cpos=False)

Parameters:

• input_ids (torch.LongTensor) –

• center_positions (torch.Tensor) –

• attention_mask (Optional[torch.FloatTensor]) –

• ignore_cpos (bool) –

Return type:

torch.Tensor

class medcat.utils.meta_cat.models.BertForMetaAnnotation(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().__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.

Note

As per the example above, an __init__() call to the parent class must be made before assignment on the child.

Variables:

training (bool) – Boolean represents whether this module is in training or evaluation mode.

_keys_to_ignore_on_load_unexpected: List[str] = ['pooler']

__init__(config)

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

forward(input_ids=None, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, inputs_embeds=None, labels=None, center_positions=[], ignore_cpos=None, output_attentions=None, output_hidden_states=None, return_dict=None)

labels (torch.LongTensor of shape (batch_size, sequence_length), optional):

Labels for computing the token classification loss. Indices should be in [0, ..., config.num_labels - 1].

Parameters:

• input_ids (Optional[torch.LongTensor]) – The input IDs. Defaults to None.

• attention_mask (Optional[torch.FloatTensor]) – The attention mask. Defaults to None.

• token_type_ids (Optional[torch.LongTensor]) – Type IDs of the tokens. Defaults to None.

• position_ids (Optional[torch.LongTensor]) – Position IDs. Defaults to None.

• head_mask (Optional[torch.FloatTensor]) – Head mask. Defaults to None.

• inputs_embeds (Optional[torch.FloatTensor]) – Input embeddings. Defaults to None.

• labels (Optional[torch.LongTensor]) – Labels. Defaults to None.

• center_positions (Optional[Any]) – Cennter positions. Defaults to None.

• output_attentions (Optional[bool]) – Output attentions. Defaults to None.

• ignore_cpos (Optional[bool]) – If center positions are to be ignored.

• output_hidden_states (Optional[bool]) – Output hidden states. Defaults to None.

• return_dict (Optional[bool]) – Whether to return a dict. Defaults to None.

Returns:

TokenClassifierOutput – The token classifier output.