Jaggu's world Comments

skin cancer reconstructive surgery: Redmine installation is easy now !!!

2009-12-22T09:21:52Z

Hi first time here and stumbling upon sites for getting some interesting info, this site seems informative. I would like to visit here very often.

shilpi: Historic paper on the origin and development of Indian Language Technologies

2009-12-16T11:37:57Z

hi jagannadh , this is shilpi ... i have been workin under sinha sir. n we met when u came there for prsg meet wid ur cdac team. are u the same jagannadh from cdac trivendrum?

jaganadhg: Finding bigrams with NLTK.

2009-11-25T05:37:41Z

Hello Muhammad
change pd = ProbDistI() to pd = ProbDistI
and ProbDistI has no function called inc . That is the problem caused in your program. See the help for ProbDistI .

Mohamed Lichouri: Finding bigrams with NLTK.

2009-11-24T15:20:57Z

Hello Jaganadh G;
i need your helo please, i have write this code:
# -*- coding: cp1252 -*-

# import FreqDist class
from nltk.probability import FreqDist

a=(open('ml_bigram.txt').read())

# create frequency distribution object
fd = FreqDist()

# for each token in the relevant text, increment its counter
for word in a.split():
fd.inc(word)

# Get a list of words sorted by frequency
for word in fd.keys():
print word, fd[word]

# save the result

result = open("freq_word.txt", "w")
result.write(str(fd))
result.close()

#**************************
this code work great with frequency distribution but when i use the same code with probability distribution i don't get anything

# -*- coding: cp1252 -*-

# import ProbDist class
from nltk.probability import ProbDistI

a=(open('ml_bigram.txt').read())

# create probability distribution object
pd = ProbDistI()

# for each token in the relevant text, increment its counter
for word in a.split():
pd.inc(word)

# Get a list of words sorted by probability
for word in pd.keys():
print word, pd[word]

# save the result

result = open("prob_word.txt", "w")
result.write(str(pd))
result.close()

best regards,
Mohamed Lichouri

Jaganadh G: Finding bigrams with NLTK.

2009-11-24T08:49:13Z

Happy to know that the code helped you.
Infact removed the 'codecs' from the code ? I included it for reading Unicode text.

Jaganadh G: Historic paper on the origin and development of Indian Language Technologies

2009-11-24T08:46:17Z

If you have access to IEEE site you can download it from there.
Please try .

Mohamed Lichouri: Finding bigrams with NLTK.

2009-11-21T23:51:57Z

Thank you for posting this code;
i had really a big problem with bigrams function.
i have change it a little and it work very well.
Best regards;
#!/usr/bin/env python
import sys
from nltk import bigrams

def gen_ML_Bigram(text):
texfbig = open(text).read()
tokens = texfbig.split()
ml_bigram = bigrams(tokens)
out = open("ml_bigram.txt",'w')
for ml in ml_bigram:
out.write(" ".join(ml))
out.write("n")

inp='wordtokens.txt'
gen_ML_Bigram(inp)

dilawar: Historic paper on the origin and development of Indian Language Technologies

2009-11-21T07:25:35Z

Yaar paper bhej dena meri email id par...

ചെലക്കാണ്ട് പോടാ: New book in 'Head First' series with python

2009-11-18T16:04:31Z

how is it?

Syam: Some thoughts on Tweeting

2009-11-14T14:08:43Z

"But I found that twitter has an impotent role in my professional area"
Impotent or Important?
And 'teeting' vs 'tweeting'