How to create an empty matrix with numpy in python ?

Examples of how to create an empty matrix with numpy in python:

Create an empty matrix using the numpy function empty()

To create for example an empty matrix of 10 columns and 0 row, a solution is to use the numpy function empty() function:

import numpy as np

A = np.empty((0,10))

Then

print(A)

gives

[]

and if we check the matrix dimensions using shape:

print(A.shape)

we get:

(0,10)

Note: by default the matrix type is float64:

 print(A.dtype)

returns

float64

Fill an empty matrix with a for loop

It is then possible for example to create an empty matrix and to update it using the function numpy.concatenate

import numpy as np

A = np.empty((0,10))

for i in range(14):

    new_row = np.random.randint(100, size=(1, 10))

    A = np.concatenate((A,new_row))

print(A)

returns

[[35. 70. 33. 58. 18. 78. 14. 44. 17.  5.]
 [42. 27.  8. 79. 19.  8. 88.  4. 40. 84.]
 [51. 96.  2. 66. 46. 10. 78. 98. 12. 85.]
 [17. 36. 11.  6. 84. 81. 99. 20. 12. 14.]
 [72. 14. 47.  8. 18. 22. 77. 99. 30. 99.]
 [44. 11.  1. 70. 26. 20. 74. 61. 95. 60.]
 [60. 20. 80. 32. 71. 22. 61. 31. 41. 27.]
 [35. 96. 65. 80. 85. 40. 27.  3. 44.  9.]
 [11. 83. 22. 95. 43. 96. 95. 26. 58. 35.]
 [25. 95. 80. 20. 52. 32. 36. 92. 42. 69.]
 [37. 49. 31. 51. 39. 72.  8. 61. 91. 26.]
 [74. 66. 24. 17.  4. 75. 67. 61. 62. 70.]
 [21. 71. 59. 51. 85. 60. 36. 39. 17. 82.]
 [ 7. 25. 27. 57. 99.  3.  5. 32. 64. 97.]]

References

• NumPy array creation: empty() function
• numpy.concatenate