convlab2.dst.sumbt package

Submodules

convlab2.dst.sumbt.BeliefTrackerSlotQueryMultiSlot module

class convlab2.dst.sumbt.BeliefTrackerSlotQueryMultiSlot.BeliefTracker(args, num_labels, device='cuda')

Bases: torch.nn.modules.module.Module

forward(input_ids, input_len, labels, n_gpu=1, target_slot=None)

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.

static init_parameter(module)
initialize_slot_value_lookup(label_ids, slot_ids)
class convlab2.dst.sumbt.BeliefTrackerSlotQueryMultiSlot.BertForUtteranceEncoding(config)

Bases: transformers.modeling_bert.BertPreTrainedModel

forward(input_ids, token_type_ids, attention_mask, output_all_encoded_layers=False)

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.

class convlab2.dst.sumbt.BeliefTrackerSlotQueryMultiSlot.MultiHeadAttention(heads, d_model, dropout=0.1)

Bases: torch.nn.modules.module.Module

attention(q, k, v, d_k, mask=None, dropout=None)
forward(q, k, v, mask=None)

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.

get_scores()

Module contents