Understand the images of various matrix operations used in Keras (Tensorflow) with examples

Introduction

When you read the Keras code, you will come across various matrix operations. It is very difficult to read without knowing the contents of these operations. This time, I will share the code that I read, because I confirmed the matrix operations that often appear in Keras based on actual examples.

environment

Even though Keras, the environment I confirmed this time is an old version of the era when it was a part of tensorflow. However, I don't think matrix operations are much different than they are now.

Precautions when operating the sample

The source code of this article assumes the following imports. Please copy and paste it if you want it to work.

import numpy as np
import warnings
import os
warnings.simplefilter('ignore')

import tensorflow as tf
from tensorflow.keras.models import Sequential, Model
from tensorflow.keras.layers import Dense, Input, Dropout, BatchNormalization

Confirmation method

The confirmation method is as follows, assuming that the processing in Keras can be imagined.

  1. Create a Keras model to perform operations

  2. Confirm the calculation contents by giving input to the predict method of the model and outputting the result.

let's try it

concat Concatenate the list of tensors along the specified axis.

Source

    input1 = Input(shape=(2,))
    input2 = Input(shape=(2,))
    output = tf.concat(axis=1, values=[input1, input2])
    model = Model(inputs=[input1, input2], outputs=[output])
    print(model.summary())
    print(model.predict(x=[np.array([[10, 20], [20, 30]]), np.array([[20, 5], [30, 2]])]))

result

[[10. 20. 20.  5.]
 [20. 30. 30.  2.]]

stack Stack the list of rank R tensors into rank R + 1 tensors. The big difference from concat is that it is not connected as it is, but it is connected after adding one axis, and it is possible to retrieve the information before connection even after connection.

Source

    # stack
    input1 = Input(shape=(2,))
    input2 = Input(shape=(2,))
    output = tf.stack(axis=1, values=[input1, input2])
    model = Model(inputs=[input1, input2], outputs=[output])
    print(model.summary())
    print(model.predict(x=[np.array([[10, 20], [20, 30]]), np.array([[20, 5], [30, 2]])]))

result

[[[10. 20.]
  [20.  5.]]
 [[20. 30.]
  [30.  2.]]]

expand_dims Add a dimension of size 1 with the subscript "axis".

Source

    input1 = Input(shape=(1,))
    output = tf.expand_dims(input1, axis=1)
    model = Model(inputs=[input1], outputs=[output])
    print(model.summary())
    print(model.predict(x=[np.array([[10], [20], [30]])]))

result

[[[10.]]
 [[20.]]
 [[30.]]]

squeeze The opposite of expand_dims, excluding one dimension with the subscript "axis" from the tensor.

Source

    input1 = Input(shape=(1,1,))
    output = tf.squeeze(input1, axis=1)
    model = Model(inputs=[input1], outputs=[output])
    print(model.summary())
    print(model.predict(x=[np.array([[[10]], [[20]], [[30]]])]))

result

[[10.]
 [20.]
 [30.]]

reduce_max Calculate the maximum value of an element across the dimensions of the tensor.

Source

The following is a 3x2 matrix calculated in 0, 1 and 2 dimensions. It is interesting that in 0 dimension, the result is broadcast as it is, in 1 dimension, the result is exactly the same, and in 2 dimension, the order is reduced.

    input1 = Input(shape=(1,2))
    output = tf.reduce_max(input1, axis=0)
    model = Model(inputs=[input1], outputs=[output])
    print(model.summary())
    print(model.predict(x=[np.array([[[10, 20]], [[20, 5]], [[30, 4]]])]))

    input1 = Input(shape=(1,2))
    output = tf.reduce_max(input1, axis=1)
    model = Model(inputs=[input1], outputs=[output])
    print(model.summary())
    print(model.predict(x=[np.array([[[10, 20]], [[20, 5]], [[30, 4]]])]))

    input1 = Input(shape=(1,2))
    output = tf.reduce_max(input1, axis=2)
    model = Model(inputs=[input1], outputs=[output])
    print(model.summary())
    print(model.predict(x=[np.array([[[10, 20]], [[20, 5]], [[30, 4]]])]))

result

[[30. 20.]
 [30. 20.]
 [30. 20.]]


[[10. 20.]
 [20.  5.]
 [30.  4.]]

[[20.]
 [20.]
 [30.]]

reduce_sum Calculate the sum of the elements of the entire dimension of the tensor. Similarly, I tried to calculate the source in 0 dimension, 1 dimension, and 2 dimensions for a 3x2 matrix. The idea is the same as reduce_max.

Source

    input1 = Input(shape=(1,2))
    output = tf.reduce_sum(input1, axis=0)
    model = Model(inputs=[input1], outputs=[output])
    print(model.summary())
    print(model.predict(x=[np.array([[[10, 20]], [[20, 5]], [[30, 4]]])]))

    input1 = Input(shape=(1,2))
    output = tf.reduce_sum(input1, axis=1)
    model = Model(inputs=[input1], outputs=[output])
    print(model.summary())
    print(model.predict(x=[np.array([[[10, 20]], [[20, 5]], [[30, 4]]])]))

    input1 = Input(shape=(1,2))
    output = tf.reduce_sum(input1, axis=2)
    model = Model(inputs=[input1], outputs=[output])
    print(model.summary())
    print(model.predict(x=[np.array([[[10, 20]], [[20, 5]], [[30, 4]]])]))

result

[[60. 29.]
 [60. 29.]
 [60. 29.]]

[[10. 20.]
 [20.  5.]
 [30.  4.]]

[[30.]
 [25.]
 [34.]]

matmul Abbreviation for "Multiplies matrix", multiplying matrix a by matrix b to generate a * b. It is often used for fully connected layers.

Source

The following is an example in which a 1x1 matrix is generated as a result of calculating the 1x2 and 2x1 matrices.

    input1 = Input(shape=(2,))
    input2 = Input(shape=(2,1))
    output = tf.matmul(input1, input2)
    model = Model(inputs=[input1, input2], outputs=[output])
    print(model.summary())
    print(model.predict(x=[np.array([[10, 20]]), np.array([[[1], [2]]])]))

result

[[[50.]]]

slice For the tensor, specify the start position and extraction size, and extract a part of the tensor.

Source

    input1 = Input(shape=(4,))
    input1_reshape = tf.reshape(input1, [4])
    input2 = Input(shape=(1), dtype=tf.int32)
    input2_reshape = tf.reshape(input2, [1])
    input3 = Input(shape=(1), dtype=tf.int32)
    input3_reshape = tf.reshape(input3, [1])
    output = tf.slice(input1_reshape, input2_reshape, input3_reshape)
    model = Model(inputs=[input1, input2, input3], outputs=[output])
    print(model.summary())
    print(model.predict(x=[np.array([[1, 8, 3, 4]]), np.array([[1]]), np.array([[1]])]))

result

[8.]

gather Specify the index for the tensor and get the element of the index.

Source

For the list [1,8,3,4], a matrix with the 0th and 3rd items taken out and the 1st and 2nd items taken out is returned. If the input shape is not fixed, an error will occur, so reshape is forcibly performed.

    input1 = Input(shape=(4,))
    input1_reshape = tf.reshape(input1, [4])
    input2 = Input(shape=(2,2), dtype=tf.int32)
    output = tf.gather(input1_reshape, input2)
    model = Model(inputs=[input1, input2], outputs=[output])
    print(model.summary())
    print(model.predict(x=[np.array([[1, 8, 3, 4]]), np.array([[[0, 3],[1, 2]]])]))

result

[[[1. 4.]
  [8. 3.]]]

in conclusion

――This time, only some operations are posted, but I hope that your reading comprehension will have improved a little. ――Even if you give it to the input of the Keras model and let it calculate, it was not easy due to an error such as the number of dimensions of the matrix does not match, but it was quite difficult, but it was very useful for understanding the matrix. --In particular, it is important to match the value of the input shape argument with the shape of the matrix given to predcit. ――This sample is also the minimum code for creating and predicting Models that take multiple inputs in Keras, so I think it will be useful when creating complex models in the future.

reference

Recommended Posts

Understand the images of various matrix operations used in Keras (Tensorflow) with examples
Summary of various operations in Tensorflow
[Python3] Understand the basics of file operations
Shuffle hundreds of thousands of images evenly with tensorflow.
Fix the argument of the function used in map
Numerical representation of days of the week in various languages
Investigate the relationship between TensorFlow and Keras in transition
Used from the introduction of Node.js in WSL environment
Identify the name from the flower image with keras (tensorflow)
Record of the first machine learning challenge with Keras
A collection of Excel operations often used in Python
Understand the metropolitan hasting method (one of the methods in Markov chain Monte Carlo method) with implementation
Let's introduce the library currently used by engineers with about 3 years of experience in Django
Try scraping the data of COVID-19 in Tokyo with Python
[TensorFlow 2 / Keras] How to run learning with CTC Loss in Keras
I tried to find the average of the sequence with TensorFlow
Reinforcement learning in the shortest time with Keras with OpenAI Gym
Introducing the potential of Plotly scatter plots with practical examples
The story of displaying images with OpenCV or PIL (only)
[Homology] Count the number of holes in data with Python
Find the eigenvalues of a real symmetric matrix in Python
Get the index of each element of the confusion matrix in Python
The story of downgrading the version of tensorflow in the demo of Mask R-CNN.