keras_deep_lstm2: keras deep lstm 2

In systats/textlearnR:

Description

Word embedding + (bidirectional) long short-term memory + Deep dense layer

Usage

keras_deep_lstm2(input_dim, embed_dim = 128, seq_len = 50,
  lstm_dim = 32, lstm_drop = 0.2, bidirectional = F,
  hidden_dims = c(32, 32, 32), output_dim = 2,
  output_fun = "softmax")

Arguments

input_dim	Number of unique vocabluary/tokens
embed_dim	Number of word vectors
seq_len	Length of the input sequences
lstm_dim	Number of lstm neurons (default 32)
lstm_drop	default is 2
bidirectional	default is F
hidden_dims	Number of neurons per layer as vector of integers
output_dim	Number of neurons of the output layer
output_fun	Output activation function

Details

Taken from https://www.kaggle.com/gidutz/text2score-keras-rnn-word-embedding

Value

keras model

systats/textlearnR documentation built on May 6, 2019, 8:31 p.m.