Comments by Member 13647869 (Top 50 by date)

i have put that space, or do i have to remove that?

i changed it to this:
word[0]=datetime.strptime(word[0],'%Y-%m-%d %H:%M:%S')

and got this:

ValueError: unconverted data remains:

sorry, it seems that I have put a wrong sample of the text file I changed it

i tried it i got this error:

ValueError: time data '2009-04-02 22:50:06 ' does not match format '%y-%m-%d %H:%M:%S'


plus, i need to index the element of each list, because each list has a different date string in the first element

ok i got this <class 'str'>. Any idea how I can change the sentiment value to an integer and the date to timestamp?

The first two columns that I am plotting are numerical. Do you guys mean that I have to change the date from 2009-06-11 to another format? The second column is 1 and 0 so its already in a numerical form

sorry but i only took one line from StackOverflow, so that does not mean i reply to the StackOverflow message

I am sorry, I am very confused about this topic

but how can combine the lines? do I use .join?

Let me try to explain further. Originaly the file looks like this:
T 2009-06-25 21:15:29
U http://twitter.com/mujiang
W No Post Title

T 2009-06-26 16:20:35
U http://twitter.com/mujiang
W No Post Title

T 2009-06-26 21:04:20
U http://twitter.com/mujiang
W No Post Title

T 2009-06-27 17:50:59
U http://twitter.com/mujiang
W No Post Title

T 2009-06-27 17:56:09
U http://twitter.com/mujiang
W No Post Title

T 2009-06-28 00:00:10
U http://twitter.com/mujiang
W No Post Title



And so on. But because I need to do some sentiment analysis on this file, and every three lines are linked, I need to group them. so every three lines become one line

As far as I understand the relationship between the location and keyword is an OR relationship. the solution is to go to the module that allows you to stream and change the relationship to AND.Just letting this out there

I tried that, but it gives me the tweets from the location. I need the tweets with the keyword 'Cat' from the users in the location

I knew that there are answers on google. ok

I know that. I have been googling for the past two days, but every time i get a method it doesn't work, its ok though thanks

Not to be rude, but if you have been doing such a project you would understand what I am talking about

with all my due respect, as far as I know, this website works this way: I have an issue I post my question, whoever WANTS to, helps...you don't have to do anything if you don't want to

search it trust me its interesting

they don't, but i used that format so i can somewhat see the tweets and if they are correct

I am receiving tweets from Twitter using their API

what is returned in json.loads is the live twitter stream of a keyword that I am searching for, i can see the outcome as tweets, and some of which are being repeated and i do not want that to happen

This part:
def on_data(self, data):
all_data = json.loads(data)
tweet = all_data["text"]
sentiment_value, confidence = s.sentiment(tweet)
tweets= open("tweets.txt","a",encoding="utf-8")
tweets.write(tweet)
tweets.write('\n\n\n')
tweets.close()

is getting the tweets, saving them in a txt file and printing them, so the loop should be before that?

It has happened before with the same exact error. when I restarted my computer the next day, I tried to download the nltk package and it worked... so

Yh the thing is I don't understand what it's telling me. there are not iterations in the code as far as I can tell. Could you please explain whats going on?

ok....

I think my encoding was not correct, or my python version was causing this

Thank you for your feedback MacCutchan, but it turns out that i had another problem that was causing this problem!

Thank you, Peter, for your help!!!!!!

THANK YOU VERY MUCH. I AM VERY GRATEFUL!

I was not sure if the code was correct, let me try and see

https://www.dropbox.com/sh/qrrzdnt6aayphlh/AADQVUpymD75qReuaIGL7MUGa?dl=0

Perfect! While I share the negative.txt file, could you please advice me how to change my code to accommodate the encoding of utf-8? I am using python 3.6.4

Data:
https://www.dropbox.com/sh/qrrzdnt6aayphlh/AADQVUpymD75qReuaIGL7MUGa?dl=0


Code:

https://www.dropbox.com/sh/4qmj43g3ub2of9g/AAB-gLeE2EkpckG8-qbXqpVoa?dl=0

ok, just give me a few minutes! i will create one for the full code, and one for the data

EXACTLY! That file, should I create it? or because its pickled the code would do that?

What if i post the file, as in part of it?

That was my original assumption, which is why in the code its reading the file, r (short_pos = open("positive.txt","r").read()), but then the ascii gave me that error.
I am very confused. As every person I ask gives me a different answer.

ok, so my problem is, that line is not doing what its supposed to be doing

the files were given to me, as in I do not know what encoding that is used. I changed the positive just to see if the difference, as in compare my assumption with the original one. I wanted to see what would happen if I encode positve.txt with utf-8 would it work? so I left negative.txt the same

Ok hold on, let me check if we are on the same page. This file, 'pickled_algos/documents.pickle', should be created by me? As in I go to my documents, create a new file, and name it save_documents? Or does the code create it? From the examples I have seen, the code creates it. Sorry for all the questions and confusion, as i said this is all very new to me

It's not just text, as I said it contains text AND numbers. What i am trying to do is use this file to create classifiers for a sentiment analysis

Ok, please bare with me, I am new to all of this =z so as you said I have to use the encode parameter, but the thing is I thought I should encode it using utf-8, but it didn't work:

i wrote this:
short_pos = open("positive.txt", "r",encoding='utf-8').read()
short_neg = open("negative.txt","r",encoding='utf-8').read()

and got this error:
short_neg = open("negative.txt","r",encoding='utf-8').read()
File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/codecs.py", line 321, in decode
(result, consumed) = self._buffer_decode(data, self.errors, final)
UnicodeDecodeError: 'utf-8' codec can't decode byte 0x97 in position 3118: invalid start byte

That's the problem, the file isn't opening

The file contains text and numbers, so if I read it as binary data, it wouldn't work, would it? As in, i cant use the split lines function and so on

I am not quite sure what your question means. Could you please explain?

so if I create the file save-documents in my documents folder, that should do it?