gets the HMM state posteriors More...
Inheritance diagram for nabu.neuralnetworks.decoders.alignment_decoder.AlignmentDecoder:
Public Member Functions | |
| def | __call__ (self, inputs, input_seq_length) |
| decode a batch of data More... | |
| def | write (self, outputs, directory, names) |
| write the output of the decoder to disk More... | |
| def | update_evaluation_loss (self, loss, outputs, references, reference_seq_length) |
| update the evaluation loss More... | |
 Public Member Functions inherited from nabu.neuralnetworks.decoders.decoder.Decoder | |
| def | __init__ (self, conf, model) |
| Decoder constructor. More... | |
| def | __call__ (self, inputs, input_seq_length) |
| decode a batch of data More... | |
| def | write (self, outputs, directory, names) |
| write the output of the decoder to disk More... | |
| def | update_evaluation_loss (self, loss, outputs, references, reference_seq_length) |
| update the evaluation loss More... | |
Additional Inherited Members | |
| Public Attributes inherited from nabu.neuralnetworks.decoders.decoder.Decoder | |
| conf | |
| model | |
Detailed Description
gets the HMM state posteriors
Member Function Documentation
§ __call__()
| def nabu.neuralnetworks.decoders.alignment_decoder.AlignmentDecoder.__call__ | ( | self, | |
| inputs, | |||
| input_seq_length | |||
| ) |
decode a batch of data
- Parameters
-
inputs the inputs as a dictionary of [batch_size x time x ...] tensors input_seq_length the input sequence lengths as a dictionary of [batch_size] vectors
- Returns
- the decoded sequences as a dictionary of outputs
§ update_evaluation_loss()
| def nabu.neuralnetworks.decoders.alignment_decoder.AlignmentDecoder.update_evaluation_loss | ( | self, | |
| loss, | |||
| outputs, | |||
| references, | |||
| reference_seq_length | |||
| ) |
update the evaluation loss
- Parameters
-
loss the current evaluation loss outputs the outputs of the decoder as a dictionary references the references as a dictionary reference_seq_length the sequence lengths of the references
- Returns
- an op to update the evalution loss
§ write()
| def nabu.neuralnetworks.decoders.alignment_decoder.AlignmentDecoder.write | ( | self, | |
| outputs, | |||
| directory, | |||
| names | |||
| ) |
write the output of the decoder to disk
- Parameters
-
outputs the outputs of the decoder directory the directory where the results should be written names the names of the utterances in outputs
The documentation for this class was generated from the following file:
