Python Extract ZIP Files: A Step Guide – Python Tutorial
In this tutorial, we will introduce you how to extract files from a zip file using python. We will use python zipfile class to implement it.
A Full List of Part-of-Speech of Word in Chinese Jieba Tool – NLP Tutorial
In chinese jieba tool, we have not found a list about chinese part-of-speech used in it. In this tutorial, we will introduce you how to get this list.
Fix Python yield AttributeError: ‘generator’ object has no attribute ‘next’ – Python Tutorial
When we are using python yield statement, we may get AttributeError: ‘generator’ object has no attribute ‘next’. In this tutorial, we will introduce how to fix this problem.
Understand librosa.load() is Between -1.0 and 1.0 – Librosa Tutorial
When we use librosa.load() to read an audio file, we may get a numpy ndarray, the value of it is -1.0 and 1.0. In this tutorial, we will introduce you the reason.
Python Remove Silence in WAV Using Librosa – Librosa Tutorial
In this tutorial, we will introduce how to use python librosa to remove silence in a wav file, which is very useful if you plan to process wav files.
Fix AttributeError: module ‘librosa’ has no attribute ‘output’ – Librosa Tutorial
In this tutorial, we will use an example to show you how to fix AttributeError: module ‘librosa’ has no attribute ‘output’ error.
Ultimate Guide to IndRNN: Handling Longer Sequence Than LSTM – LSTM Tutorial
IndRNN is proposed in paper: Independently Recurrent Neural Network (IndRNN): Building A Longer and Deeper RNN. In this tutorial, we will use some examples to introduce this model.
Get Forward and Backward LSTM Output of Bi-LSTM in TensorFlow – TensorFlow Tutorial
![We usually use Bi-LSTM or Bi-GRU to model sequence. How to get its output? In this tutorial, we will discuss this topic. Bi-LSTM or Bi-GRU A Bi-LSTM or Bi-GRU looks like below: As to Bi-LSTM, it contains a forward LSTM and backword LSTM. How to create Bi-LSTM in TensorFlow? In tensroflow, we can use tf.nn.bidirectional_dynamic_rnn() to create a Bi-LSTM network. Here is an example: import tensorflow as tf import numpy as np #5*4*10 inputs = tf.Variable(tf.truncated_normal([3, 5, 100], stddev=0.1), name="inputs") hidden_size = 10 outputs, state = tf.nn.bidirectional_dynamic_rnn( cell_fw=tf.nn.rnn_cell.LSTMCell(hidden_size, forget_bias=1.0), cell_bw=tf.nn.rnn_cell.LSTMCell(hidden_size, forget_bias=1.0), inputs=inputs, sequence_length= None, dtype=tf.float32, scope='bilstm_doc_word' ) You should notice the inputs tensor of Bi-LSTM is [batch_size, max_seq_length, dim]. In this example, it is [3, 5, 100]. How to get the output of Bi-LSTM? There are three types of outputs in Bi-LSTM. Type 1: concatenating the output of forward and backward LSTM. outputs_merge = tf.concat(outputs, 2) Then you will get a output with shape [batch_size, max_seq_length, 2*hidden_dim], here hidden_dim is the hidden state size in lstm. As to example above, you will find outputs_merge is [3, 5, 2*10] Type 2: Get the mean output of the forward and backward LSTM. You can find how to do in this tutorial: Average the Output of RNN/GRU/LSTM/BiLSTM for Variable Length Sequences – Deep Learning Tutorial Type 3: Get the last hidden output of the forward and backward LSTM. Here is an example: last_forward = outputs[0][:,-1,:] # 3*5*10 last_backward = outputs[1][:,0,:] init = tf.global_variables_initializer() init_local = tf.local_variables_initializer() with tf.Session() as sess: sess.run([init, init_local]) np.set_printoptions(precision=4, suppress=True) f =sess.run(last_forward) b = sess.run(last_backward) m = sess.run(outputs_merge) print("last forward shape=", f.shape) print(f) print("last backward shape=", b.shape) print(b) print("concatenate forward and backward") print(m) Run this code, you will see:](https://www.tutorialexample.com/wp-content/uploads/2021/11/Get-Forward-and-Backward-LSTM-Output-of-Bi-LSTM-in-TensorFlow-TensorFlow-Tutorial-200x140.png)
We usually use Bi-LSTM or Bi-GRU to model sequence. How to get its output? In this tutorial, we will discuss this topic.
Fix TypeError: Object of type ‘ndarray’ is not JSON serializable – Python Tutorial
In this tutorial, we will use an example to show you how to fix TypeError: Object of type ‘ndarray’ is not JSON serializable in python.
An Improvement to Densely Connected Bidirectional LSTM – LSTM Tutorial
In this tutorial, we will introduce Densely Connected Bidirectional LSTM (DC-BiLSTM) and give an improvement method for it.
