TypeError：Python文件中无法理解的数据类型

Question

I am trying to read a binary file using following command:

import numpy as np
fid = open(filename, 'rb')
ax = np.fromfile(fid, dtype=np.int32, count=1)

This command is working fine, However

ay = np.fromfile(fid, dtype=np.char, count=16)

gives an TypeError: data type not understood. Any idea how can i read it as character type?

Answer 1

Your desired data type is non-existent, np.char actually is a module .

Take a look at the numpy datatypes , you could cover your byte representation using np.byte , which is a np.int8 .

Answer 2

You should use

ay = np.fromfile(fid, dtype=np.byte, count=16)

instead of

ay = np.fromfile(fid, dtype=np.char, count=16)

because numpy doesn't contain scalar type char . More about numpy data types you could see here . numpy.byte type corresponding to C char type.
If you want convert array of 16 binary digits to one int you can use following code:

aybin = np.fromfile(fid, dtype=np.char, count=16)
ay = int(("".join(str(d) for d in aybin)), 2)

Answer 3

So, after I saw your communication via comments, I update my answer as follows:

In your case you should use:

ay = np.fromfile(fid, dtype=np.byte, count=16)

However I leave the parts of my previous answer here to consider to use in other particular case.

Try numpy.loadtxt() .

More flexible way of loading data from a text file.

But if you even use numpy.fromfile() , then you have to use the arg sep , because empty (“”) separator means the file should be treated as binary .

As you can read here , in numpy manual:

sep : str

Separator between items if file is a text file. Empty (“”) separator means the file should be treated as binary. Spaces (” ”) in the separator match zero or more whitespace characters. A separator consisting only of spaces must match at least one whitespace.