Incompatible types in benchmarks.word_tokenization

[bact]

Description

MyPy reports a bunch of typing issues in pythainlp/benchmarks/word_tokenization.py

Expected results

• All functions have explicit type hinting information
• No typing incompatible issues

Current results

ref_sample in these two lines for examples, are seen as str and should not have shape attribute.

pythainlp/pythainlp/benchmarks/word_tokenization.py

Lines 164 to 165 in 9a1274b

	c_pos_pred = c_pos_pred[c_pos_pred < ref_sample.shape[0]]
	c_neg_pred = c_neg_pred[c_neg_pred < ref_sample.shape[0]]

But it looks like from _binary_representation function, it may has a type of ND array.

However, the _binary_representation type hints and docstring said they are str:

pythainlp/pythainlp/benchmarks/word_tokenization.py

Lines 208 to 221 in 9a1274b

	def _binary_representation(txt: str, verbose: bool = False):
	"""
	Transform text into {0, 1} sequence.
	where (1) indicates that the corresponding character is the beginning of
	a word. For example, ผม|ไม่|ชอบ|กิน|ผัก -> 10100...
	:param str txt: input text that we want to transform
	:param bool verbose: for debugging purposes
	:return: {0, 1} sequence
	:rtype: str
	"""
	chars = np.array(list(txt))

So there're confusions here to be fixed.

Steps to reproduce

Use MyPy to check the code

PyThaiNLP version

5

Python version

any

Operating system and version

any

More info

No response

Possible solution

No response

Files

No response

[wannaphong]

@bact Is it fixed? #1289

[bact]

Fixed by #1232, #1270, #1289