How to convert numpy to float32?

Question

I have a data I want to split and convert it to float32 but it is showing the real number as a string

   data = open('Path dataset')
        for line in data:
        train = np.array([np.float32(x) for x in line.split(",")[:]])

And the error that showing me is:

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-83-53e8671c416d> in <module>
      1 for line in data_coba:
----> 2     train = np.array([np.float32(x) for x in line.split(",")[:]])          
      3 #print(train_test_coba)

<ipython-input-83-53e8671c416d> in <listcomp>(.0)
      1 for line in data_coba:
----> 2     train = np.array([np.float32(x) for x in line.split(",")[:]])           
      3 #print(train_test_coba)

ValueError: could not convert string to float: 'ï»¿50.89482266'

What is wrong with this?

@abhilb yes correct but my data is real numbers it does not contain any of string ï»¿ — AbuOmair
– AbuOmair, Commented Dec 3, 2019 at 6:02
change the encoding while opening the file to utf-8-sig , ie. open('Path dataset', encoding='utf-8-sig') — Shijith
– Shijith, Commented Dec 3, 2019 at 6:05
You are hit by BOM. See en.wikipedia.org/wiki/Byte_order_mark — tstanisl
– tstanisl, Commented Dec 3, 2019 at 8:38

Arijit Gupta · Accepted Answer · 2019-12-03 06:02:13Z

1

It seems that your dataset contains characters other than just comma-separated numbers. So the error is possibly occurring when it is trying to convert these non-numeric characters to float32. I suggest you check your dataset again and maybe try splitting it more.

answered Dec 3, 2019 at 6:02

Arijit Gupta

732 silver badges8 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

Arijit Gupta Over a year ago

Maybe you can separately check what for x in line.split(",")[:]] gives as output and move further from there.

davidbilla · Accepted Answer · 2019-12-03 06:33:31Z

0

You have to use encoding='utf-8' in the open function

data = open('Path dataset',encoding='utf-8')

answered Dec 3, 2019 at 6:33

davidbilla

2,2321 gold badge22 silver badges28 bronze badges

2 Comments

AbuOmair Over a year ago

when I use this condition, I get this error ValueError: could not convert string to float: '\ufeff51.45440668'

davidbilla Over a year ago

What does the 'ï»¿' signify in the value? If it is ok to ignore those unwanted characters, you can replace these characters. In: np.float32(re.sub(r'[^a-zA-Z0-9_\.]', '', 'ï»¿50.89482266')) Out: 50.89482

Collectives™ on Stack Overflow

How to convert numpy to float32?

2 Answers 2

1 Comment

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related