Skip to content
Projects
Groups
Snippets
Help
Loading...
Help
Submit feedback
Contribute to GitLab
Sign in
Toggle navigation
gargantext
Project
Project
Details
Activity
Releases
Cycle Analytics
Repository
Repository
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Charts
Issues
0
Issues
0
List
Board
Labels
Milestones
Merge Requests
0
Merge Requests
0
CI / CD
CI / CD
Pipelines
Jobs
Schedules
Charts
Wiki
Wiki
Members
Members
Collapse sidebar
Close sidebar
Activity
Graph
Charts
Create a new issue
Jobs
Commits
Issue Boards
Open sidebar
humanities
gargantext
Commits
ea7b4d95
Commit
ea7b4d95
authored
Jan 16, 2015
by
Administrator
Browse files
Options
Browse Files
Download
Plain Diff
Merge branch 'stable' into tina
parents
b28dd82b
5c7d269d
Changes
4
Hide whitespace changes
Inline
Side-by-side
Showing
4 changed files
with
39 additions
and
35 deletions
+39
-35
functions.py
analysis/functions.py
+23
-23
views.py
gargantext_web/views.py
+8
-6
init_accounts.py
init_accounts.py
+6
-4
corpus.html
templates/corpus.html
+2
-2
No files found.
analysis/functions.py
View file @
ea7b4d95
...
...
@@ -244,31 +244,31 @@ def get_cooc(request=None, corpus_id=None, cooc_id=None, type='node_link', n=150
return
data
def
tfidf
(
corpus
,
document
,
ngram
):
'''
Compute TF-IDF (Term Frequency - Inverse Document Frequency)
See: http://en.wikipedia.org/wiki/Tf
%
E2
%80%93
idf
'''
try
:
occurences_of_ngram
=
Node_Ngram
.
objects
.
get
(
node
=
document
,
ngram
=
ngram
)
.
weight
ngrams_by_document
=
sum
([
x
.
weight
for
x
in
Node_Ngram
.
objects
.
filter
(
node
=
document
)])
term_frequency
=
occurences_of_ngram
/
ngrams_by_document
xx
=
Node
.
objects
.
filter
(
parent
=
corpus
,
type
=
NodeType
.
objects
.
get
(
name
=
"Document"
))
.
count
()
yy
=
Node_Ngram
.
objects
.
filter
(
ngram
=
ngram
)
.
count
()
# filter: ON node.parent=corpus
inverse_document_frequency
=
log
(
xx
/
yy
)
# result = tf * idf
result
=
term_frequency
*
inverse_document_frequency
except
Exception
as
error
:
print
(
error
,
ngram
)
result
=
0
return
result
#
def tfidf(corpus, document, ngram):
#
'''
#
Compute TF-IDF (Term Frequency - Inverse Document Frequency)
#
See: http://en.wikipedia.org/wiki/Tf%E2%80%93idf
#
'''
#
try:
#
occurences_of_ngram = Node_Ngram.objects.get(node=document, ngram=ngram).weight
#
ngrams_by_document = sum([ x.weight for x in Node_Ngram.objects.filter(node=document)])
#
term_frequency = occurences_of_ngram / ngrams_by_document
#
#
xx = Node.objects.filter(parent=corpus, type=NodeType.objects.get(name="Document")).count()
#
yy = Node_Ngram.objects.filter(ngram=ngram).count() # filter: ON node.parent=corpus
#
inverse_document_frequency= log(xx/yy)
#
#
# result = tf * idf
#
result = term_frequency * inverse_document_frequency
#
except Exception as error:
#
print(error, ngram)
#
result = 0
#
return result
from
analysis.tfidf
import
tfidf
def
do_tfidf
(
corpus
,
reset
=
True
):
print
(
"doing tfidf"
)
with
transaction
.
atomic
():
if
reset
==
True
:
NodeNodeNgram
.
objects
.
filter
(
nodex
=
corpus
)
.
delete
()
...
...
gargantext_web/views.py
View file @
ea7b4d95
...
...
@@ -424,10 +424,11 @@ def subcorpus(request, project_id, corpus_id, start , end ):
project
=
Node
.
objects
.
get
(
id
=
project_id
)
corpus
=
Node
.
objects
.
get
(
id
=
corpus_id
)
type_document
=
NodeType
.
objects
.
get
(
name
=
"Document"
)
# retrieving all the documents
documents
=
corpus
.
children
.
all
()
number
=
corpus
.
children
.
count
()
# documents = corpus.children.all()
documents
=
corpus
.
__class__
.
objects
.
filter
(
parent_id
=
corpus_id
,
type
=
type_document
)
number
=
len
(
documents
)
filtered_docs
=
[]
# filtering documents by range-date
...
...
@@ -492,10 +493,11 @@ def subcorpusJSON(request, project_id, corpus_id, start , end ):
project
=
Node
.
objects
.
get
(
id
=
project_id
)
corpus
=
Node
.
objects
.
get
(
id
=
corpus_id
)
type_document
=
NodeType
.
objects
.
get
(
name
=
"Document"
)
# retrieving all the documents
documents
=
corpus
.
children
.
all
()
number
=
corpus
.
children
.
count
()
# documents = corpus.children.all()
documents
=
corpus
.
__class__
.
objects
.
filter
(
parent_id
=
corpus_id
,
type
=
type_document
)
number
=
len
(
documents
)
filtered_docs
=
[]
# filtering documents by range-date
...
...
init_accounts.py
View file @
ea7b4d95
...
...
@@ -20,8 +20,10 @@ def notify_user(username, email, password):
'''
%
(
username
,
password
)
#send_mail('[Gargantext] Votre compte', message, 'alexandre.delanoe@mines-paristech.fr', [email], fail_silently=False )
send_mail
(
'[Gargantext] Votre compte'
,
message
,
'alexandre.delanoe@mines-paristech.fr'
,
[
email
],
[
'alexandre@delanoe.org'
]
)
send_mail
(
'[Gargantext] Création de votre compte'
,
message
,
'alexandre.delanoe@mines-paristech.fr'
,
[
email
],
fail_silently
=
False
)
#send_mail('[Gargantext] Votre compte', message, 'alexandre.delanoe@mines-paristech.fr', [email], ['alexandre@delanoe.org'] )
# add option for mass sending email
def
create_user
(
username
,
email
,
password
=
None
,
active
=
False
,
notify
=
True
):
...
...
@@ -58,7 +60,7 @@ def mines_account_creation(fichier=None):
accounts
=
open
(
fichier
,
"r"
)
for
line
in
accounts
.
readlines
():
username
,
email
,
password
=
line
.
split
(
','
)
#
create_user(username, email, password=password, notify=True)
delete_user
(
username
)
create_user
(
username
,
email
,
password
=
password
,
notify
=
True
)
#
delete_user(username)
fichier
.
close
()
templates/corpus.html
View file @
ea7b4d95
...
...
@@ -172,7 +172,7 @@ function dateToInt(todayTime) {
// I've clicked "Read Documents":
function updateDocuments(pagenumber,pagenav) {
pagenav = (pagenav)?pagenav:true;
pagenumber = (pagenumber)?pagenumber:1;
pr("in here pagenav:"+pagenav+" - pagenumber:"+pagenumber)
...
...
@@ -230,7 +230,7 @@ function updateDocuments(pagenumber,pagenav) {
var current_docs = {}
var BIS_dict = {}
var corpusid = window.location.href.split("corpus")[1].replace(/\//g, '')//replace all the slashes
var theurl = "
http://localhost:8000
/api/nodes/"+corpusid+"/children/duplicates?keys=title&limit=9999"
var theurl = "/api/nodes/"+corpusid+"/children/duplicates?keys=title&limit=9999"
$.ajax({
url: theurl,
success: function(data) {
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment