Reading erroneous data form csv file using read_csv from pandas - python

I am trying to read data from a huge csv file I have. I is showing me this error UnicodeDecodeError: 'utf-8' codec can't decode byte 0xae in position 13: invalid start byte. Is there any way to just skip through the lines that cause this exception to be thrown? From the millions of lines these are just a handful and I can't manually delete them. I tried adding error_bad_lines=False, but that did not solve the problem. I am using Python 3.6.1 that I got through Anaconda 4.4.0. I am also using a Mac if that helps. Please help me I am new to this.

Seems to me that there are some non-ascii characters in your file that cannot be decoded. Pandas accepts an encoding as an argument for read_csv (if that helps):
my_file = pd.read_csv('Path/to/file.csv', encoding = 'encoding')
The default encoding is None, which is why you might be getting those errors.Here is a link to the standard Python encodings - Try "ISO-8859-1" (aka 'latin1') or maybe 'utf8' to start.
Pandas does allow you to specify rows to skip when reading a csv, but you would need to know the index of those rows, which in your case would be very difficult.

Related

I keep getting the "UnicodeDecodeError: 'utf-8' codec can't decode byte 0x80 in position 3131: invalid start byte" error when I write simple code [duplicate]

https://github.com/affinelayer/pix2pix-tensorflow/tree/master/tools
An error occurred when compiling "process.py" on the above site.
python tools/process.py --input_dir data -- operation resize --outp
ut_dir data2/resize
data/0.jpg -> data2/resize/0.png
Traceback (most recent call last):
File "tools/process.py", line 235, in <module>
main()
File "tools/process.py", line 167, in main
src = load(src_path)
File "tools/process.py", line 113, in load
contents = open(path).read()
File"/home/user/anaconda3/envs/tensorflow_2/lib/python3.5/codecs.py", line 321, in decode
(result, consumed) = self._buffer_decode(data, self.errors, final)
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xff in position 0: invalid start byte
What is the cause of the error?
Python's version is 3.5.2.
Python tries to convert a byte-array (a bytes which it assumes to be a utf-8-encoded string) to a unicode string (str). This process of course is a decoding according to utf-8 rules. When it tries this, it encounters a byte sequence which is not allowed in utf-8-encoded strings (namely this 0xff at position 0).
Since you did not provide any code we could look at, we only could guess on the rest.
From the stack trace we can assume that the triggering action was the reading from a file (contents = open(path).read()). I propose to recode this in a fashion like this:
with open(path, 'rb') as f:
contents = f.read()
That b in the mode specifier in the open() states that the file shall be treated as binary, so contents will remain a bytes. No decoding attempt will happen this way.
Use this solution it will strip out (ignore) the characters and return the string without them. Only use this if your need is to strip them not convert them.
with open(path, encoding="utf8", errors='ignore') as f:
Using errors='ignore'
You'll just lose some characters. but if your don't care about them as they seem to be extra characters originating from a the bad formatting and programming of the clients connecting to my socket server.
Then its a easy direct solution.
reference
Use encoding format ISO-8859-1 to solve the issue.
Had an issue similar to this, Ended up using UTF-16 to decode. my code is below.
with open(path_to_file,'rb') as f:
contents = f.read()
contents = contents.rstrip("\n").decode("utf-16")
contents = contents.split("\r\n")
this would take the file contents as an import, but it would return the code in UTF format. from there it would be decoded and seperated by lines.
I've come across this thread when suffering the same error, after doing some research I can confirm, this is an error that happens when you try to decode a UTF-16 file with UTF-8.
With UTF-16 the first characther (2 bytes in UTF-16) is a Byte Order Mark (BOM), which is used as a decoding hint and doesn't appear as a character in the decoded string. This means the first byte will be either FE or FF and the second, the other.
Heavily edited after I found out the real answer
It simply means that one chose the wrong encoding to read the file.
On Mac, use file -I file.txt to find the correct encoding. On Linux, use file -i file.txt.
I had a similar issue with PNG files. and I tried the solutions above without success.
this one worked for me in python 3.8
with open(path, "rb") as f:
use only
base64.b64decode(a)
instead of
base64.b64decode(a).decode('utf-8')
This is due to the different encoding method when read the file. In python, it defaultly
encode the data with unicode. However, it may not works in various platforms.
I propose an encoding method which can help you solve this if 'utf-8' not works.
with open(path, newline='', encoding='cp1252') as csvfile:
reader = csv.reader(csvfile)
It should works if you change the encoding method here. Also, you can find other encoding method here standard-encodings , if above doesn't work for you.
Those getting similar errors while handling Pandas for data frames use the following solution.
example solution.
df = pd.read_csv("File path", encoding='cp1252')
I had this UnicodeDecodeError while trying to read a '.csv' file using pandas.read_csv(). In my case, I could not manage to overcome this issue using other encoder types. But instead of using
pd.read_csv(filename, delimiter=';')
I used:
pd.read_csv(open(filename, 'r'), delimiter=';')
which just seems working fine for me.
Note that: In open() function, use 'r' instead of 'rb'. Because 'rb' returns bytes object that causes to happen this decoder error in the first place, that is the same problem in the read_csv(). But 'r' returns str which is needed since our data is in .csv, and using the default encoding='utf-8' parameter, we can easily parse the data using read_csv() function.
if you are receiving data from a serial port, make sure you are using the right baudrate (and the other configs ) : decoding using (utf-8) but the wrong config will generate the same error
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xff in position 0: invalid start byte
to check your serial port config on linux use : stty -F /dev/ttyUSBX -a
I had a similar issue and searched all the internet for this problem
if you have this problem just copy your HTML code in a new HTML file and use the normal <meta charset="UTF-8">
and it will work....
just create a new HTML file in the same location and use a different name
Check the path of the file to be read. My code kept on giving me errors until I changed the path name to present working directory. The error was:
newchars, decodedbytes = self.decode(data, self.errors)
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xff in position 0: invalid start byte
If you are on a mac check if you for a hidden file, .DS_Store. After removing the file my program worked.
I had a similar problem.
Solved it by:
import io
with io.open(filename, 'r', encoding='utf-8') as fn:
lines = fn.readlines()
However, I had another problem. Some html files (in my case) were not utf-8, so I received a similar error. When I excluded those html files, everything worked smoothly.
So, except from fixing the code, check also the files you are reading from, maybe there is an incompatibility there indeed.
You have to use the encoding as latin1 to read this file as there are some special character in this file, use the below code snippet to read the file.
The problem here is the encoding type. When Python can't convert the data to be read, it gives an error.
You can you latin1 or other encoding values.
I say try and test to find the right one for your dataset.
I have the same issue when processing a file generated from Linux. It turns out it was related with files containing question marks..
Following code worked in my case:
df = pd.read_csv(filename,sep = '\t', encoding='cp1252')
If possible, open the file in a text editor and try to change the encoding to UTF-8. Otherwise do it programatically at the OS level.

Getting UnicodeDecodeError while reading excel in Tornado,Python

i'm using postman to send an excel file which i am reading in tornado.
Tornado code
self.request.files['1'][0]['body'].decode()
here if i send .csv than, the above code works.
if i send .xlsx file than i am stuck with this error.
UnicodeDecodeError: 'utf-8' codec can't decode byte 0x87 in position 10: invalid start byte
request.files will fetch the file but the type would be byte. so to convert byte to str i've used decode(), which works only for .csv and not for .xlsx
i tried decode('utf-8') but still no luck.
i've tried searching but didn't find any issue mentioning 0x87 problem?
The reason is that the .xlsx file has a different encoding, not utf-8. You'll need to use the original encoding to decode the file.
There's no guaranteed way of finding out the encoding of a file programmatically. I'm guessing you're making this application for general users and so you will keep encountering files with different and unexpected encodings.
A good way to deal with this is by trying to decode using multiple encodings, in case one fails. Example:
encodings = ['utf-8', 'iso-8859-1', 'windows-1251', 'windows-1252']
for encoding in encodings:
try:
decoded_file = self.request.files['1'][0]['body'].decode(encoding)
except UnicodeDecodeError:
# this will run when the current encoding fails
# just ignore the error and try the next one
pass
else:
# this will run when an encoding passes
# break the loop
# it is also a good idea to re-encode the
# decoded files to utf-8 for your purpose
decoded_file = decoded_file.encode("utf8")
break
else:
# this will run when the for loop ends
# without successfully decoding the file
# now you can return an error message
# to the user asking them to change
# the file encoding and re upload
self.write("Error: Unidentified file encoding. Re-upload with UTF-8 encoding")
return
# when the program reaches here, it means
# you have successfully decoded the file
# and you can access it from `decoded_file` variable
Here's a list of some common encodings: What is the most common encoding of each language?
I faced the same issue and this worked for me.
import io
df = pd.read_excel(io.BytesIO(self.request.files['1'][0]['body']))
try this one, following suggestions provided here:
self.request.files['1'][0]['body'].decode('iso-8859-1').encode('utf-8')

utf-8 error when opening csv file in pandas on mac

I am trying to open a csv file with Japanese characters using utf8 on my mac.
The code that I am using is as follows:
foo = pd.read_csv("filename.csv", encoding = 'utf8')
However, I have been getting the following error message.
'utf-8' codec can't decode byte 0x96 in position 0
I've tried looking around but a lot of the solutions seem to be for windows/I haven't had any success with other solutions yet.
Appreciate the help!
It seems that your file really has a non-unicode character. A correct encoding for this file strongly depends on its content, but in the most common case, 0x96 can be decoded with CP-1252. So, just try to decode it like following:
foo = pd.read_csv("filename.csv", encoding = 'cp1252')
If you don't know the original encoding of the file, you can try to detect it with third-party libs such as chardet.
I may help you a little bit more if you upload a chunk of the file to reproduce the problem.

Pandas read _excel: 'utf-8' codec can't decode byte 0xa8 in position 14: invalid start byte

Trying to read MS Excel file, version 2016. File contains several lists with data. File downloaded from DataBase and it can be opened in MS Office correctly. In example below I changed the file name.
EDIT: file contains russian and english words. Most probably used the Latin-1 encoding, but encoding='latin-1' does not help
import pandas as pd
with open('1.xlsx', 'r', encoding='utf8') as f:
data = pd.read_excel(f)
Result:
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xa8 in position 14: invalid start byte
Without encoding ='utf8'
'charmap' codec can't decode byte 0x9d in position 622: character maps to <undefined>
P.S. Task is to process 52 files, to merge data in every sheet with corresponded sheets in the 52 files. So, please no handle work advices.
The problem is that the original requester is calling read_excel with a filehandle as the first argument. As demonstrated by the last responder, the first argument should be a string containing the filename.
I ran into this same error using:
df = pd.read_excel(open("file.xlsx",'r'))
but correct is:
df = pd.read_excel("file.xlsx")
Most probably you're using Python3. In Python2 this wouldn't happen.
xlsx files are binary (actually they're an xml, but it's compressed), so you need to open them in binary mode. Use this call to open:
open('1.xlsx', 'rb')
There's no full traceback, but I imagine the UnicodeDecodeError comes from the file object, not from read_excel(). That happens because the stream of bytes can contain anything, but we don't want decoding to happen too soon; read_excel() must receive raw bytes and be able to process them.
Most probably the problem is in Russian symbols.
Charmap is default decoding method used in case no encoding is beeing noticed.
As I see if utf-8 and latin-1 do not help then try to read this file not as
pd.read_excel(f)
but
pd.read_table(f)
or even just
f.readline()
in order to check what is a symbol raise an exeception and delete this symbol/symbols.
Panda support encoding feature to read your excel
In your case you can use:
df=pd.read_excel('your_file.xlsx',encoding='utf-8')
or if you want in more of system specific without any surpise you can use:
df=pd.read_excel('your_file.xlsx',encoding='sys.getfilesystemencoding()')

UnicodeDecodeError: ('utf-8' codec) while reading a csv file [duplicate]

This question already has answers here:
UnicodeDecodeError when reading CSV file in Pandas with Python
(25 answers)
Closed 5 years ago.
what i am trying is reading a csv to make a dataframe---making changes in a column---again updating/reflecting changed value into same csv(to_csv)- again trying to read that csv to make another dataframe...there i am getting an error
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xe7 in position 7: invalid continuation byte
my code is
import pandas as pd
df = pd.read_csv("D:\ss.csv")
df.columns #o/p is Index(['CUSTOMER_MAILID', 'False', 'True'], dtype='object')
df['True'] = df['True'] + 2 #making changes to one column of type float
df.to_csv("D:\ss.csv") #updating that .csv
df1 = pd.read_csv("D:\ss.csv") #again trying to read that csv
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xe7 in position 7: invalid continuation byte
So please suggest how can i avoid the error and be able to read that csv again to a dataframe.
I know somewhere i am missing "encode = some codec type" or "decode = some type" while reading and writing to csv.
But i don't know what exactly should be changed.so need help.
Known encoding
If you know the encoding of the file you want to read in,
you can use
pd.read_csv('filename.txt', encoding='encoding')
These are the possible encodings:
https://docs.python.org/3/library/codecs.html#standard-encodings
Unknown encoding
If you do not know the encoding, you can try to use chardet, however this is not guaranteed to work. It is more a guess work.
import chardet
import pandas as pd
with open('filename.csv', 'rb') as f:
result = chardet.detect(f.read()) # or readline if the file is large
pd.read_csv('filename.csv', encoding=result['encoding'])
Is that error happening on your first read of the data, or on the second read after you write it out and read it back in again? My guess is that it's actually happening on the first read of the data, because your CSV has an encoding that isn't UTF-8.
Try opening that CSV file in Notepad++, or Excel, or LibreOffice. Does your data source have the ç (C with cedilla) character in it? If it does, then that 0xE7 byte you're seeing is probably the ç encoded in either Latin-1 or Windows-1252 (called "cp1252" in Python).
Looking at the documentation for the Pandas read_csv() function, I see it has an encoding parameter, which should be the name of the encoding you expect that CSV file to be in. So try adding encoding="cp1252" to your read_csv() call, as follows:
df = pd.read_csv(r"D:\ss.csv", encoding="cp1252")
Note that I added the character r in front of the filename, so that it will be considered a "raw string" and backslashes won't be treated specially. That way you don't get a surprise when you change the filename from ss.csv to new-ss.csv, where the string D:\new-ss.csv would be read as D, :, newline character, e, w, etc.
Anyway, try that encoding parameter on your first read_csv() call and see if it works. (It's only a guess, since I don't know your actual data. If the data file isn't private and isn't too large, try posting the data file so we can see its contents -- that would let us do better than just guessing.)
One simple solution is you can open the csv file in an editor like Sublime Text and save it with 'utf-8' encoding. Then we can easily read the file through pandas.
Above method used by importing and then detecting file type works
import chardet
import pandas as pd
import chardet
with open('filename.csv', 'rb') as f:
result = chardet.detect(f.read()) # or readline if the file is large
pd.read_csv('filename.csv', encoding=result['encoding'])
Yes you'll get this error. I have work around with this problem, by opening csv file in notepad++ and changing the encoding throught Encoding menu -> convert to UTF-8. Then saving the file. Then again running python program over it.
Other solution is using codecs module in python for encoding-decoding of files. I haven't used that.
I am new to python. Ran into this exact issue when I manually changed the extension on my excel file to .csv and tried to read it with read_csv. However, if I opened the excel file and saved as csv file instead it seemed to work.

Categories