Geek Answers Handbook

API Reference for RNN and Seq2Seq models in tensor flow

Where can I find API links that indicate available functions in the RNN and Seq2Seq models.

On the github page, it was mentioned that rnn and seq2seq were ported to tf.nn

+4
source share
3 answers

[NOTE: this answer has been updated for r1.0 ... but explains legacy_seq2seqinstead tensorflow/tensorflow/contrib/seq2seq/]

The good news is that the seq2seq models presented in the tensor flow are quite complex, including attachments, buckets, an attention mechanism, multi-purpose one-to-many models, etc.

, Python , "" RNN "API" seq2seq, ... , .

, , , , ... , , -level API- Python

seq2seq RNN r1.0:

models/tutorials/rnn/translate/translate.py

... main(), train(), decode(), , ...

models/tutorials/rnn/translate/seq2seq_model.py

... class Seq2SeqModel() RNN- , , ... , , .

tensorflow/tensorflow/contrib/legacy_seq2seq/python/ops/seq2seq.py

... seq2seq . . model_with_buckets(), embedding_attention_seq2seq(), embedding_attention_decoder(), attention_decoder(), sequence_loss() .. one2many_rnn_seq2seq /, basic_rnn_seq2seq. , , .

tensorflow/tensorflow/contrib/rnn/python/ops/core_rnn.py

... RNN, static_rnn(), , , :

def simple_rnn(cell, inputs, dtype, score):
    with variable_scope.variable_scope(scope or "simple_RNN") as varscope1:
            if varscope1.caching_device is None:
                varscope1.set_caching_device(lambda op: op.device)

        batch_size = array_ops.shape(inputs[0])[0]
        outputs = []
        state = cell.zero_state(batch_size, dtype)            

        for time, input_t in enumerate(inputs):
           if time > 0:      
             variable_scope.get_variable_scope().reuse_variables()


           (output, state) = cell(input_t, state)

           outputs.append(output)

        return outputs, state
+7

RNN docs for the current / main version of TensorFlow: https://www.tensorflow.org/versions/master/api_docs/python/nn.html#recurrent-neural-networks

RNN docs for a specific version of TensorFlow: https://www.tensorflow.org/versions/r0.10/api_docs/python/nn.html#recurrent-neural-networks

For the curious, here are a few notes on why RNN documents were not available initially: API documents do not list RNNs

0
source

More articles:


All Articles