Skip to content

several bug fixes to twsi_evaluation.py script (see comments near changes) #2

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
mpelevina wants to merge 1 commit into
base: master
Choose a base branch
from
Open

several bug fixes to twsi_evaluation.py script (see comments near changes) #2

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
34 changes: 21 additions & 13 deletions twsi_evaluation.py
100644 → 100755
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -97,7 +97,8 @@ def load_assigned_senses():
"""
def load_twsi_senses():
print "Loading TWSI sense inventory..."
substitutions = read_csv(INVENTORY_FILE, '/\t+/', encoding='utf8', header=None)
# otherwise a warning was shown, that c engine cannot be used because c engine cannot work with pattern as separators (or smth like this)
substitutions = read_csv(INVENTORY_FILE, '/\t+/', encoding='utf8', header=None, engine="python")

@alexanderpanchenko alexanderpanchenko Jan 21, 2016

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

why python engine?

for i,s in substitutions.iterrows():
# create new TwsiSubst for the given word
word, t_id, subs = s[0].split('\t')
Expand All @@ -123,7 +124,7 @@ def load_sense_inventory(filename):
mapping_f = "data/Mapping_"+split(INVENTORY_FILE)[1]+"_"+split(filename)[1]
print "Mapping saved to "+mapping_f
m_f = open(mapping_f, 'w')
inventory = read_csv(filename, '/\t+/', encoding='utf8', header=None)
inventory = read_csv(filename, '/\t+/', encoding='utf8', header=None, engine="python")

@alexanderpanchenko alexanderpanchenko Jan 21, 2016

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

why python engine?

for r,inv in inventory.iterrows():
word, ident, terms = inv[0].split('\t')
if word in twsi_subst:
Expand All @@ -138,14 +139,16 @@ def load_sense_inventory(filename):
el = el.strip()
# split element into word and score
el_split = el.rsplit(VALSEP, 1)
word = el_split[0]
# this command was overwriting the 'word' variable
# semantic(word) = element of the inventory, semantic(word2) = element of the sense cluster representing the 'word'
word2 = el_split[0]
if len(el_split) > 1 and not re.match('\D+', el_split[1]):
if word in word_vec:
word_vec[word] += float(el_split[1])
if word2 in word_vec:
word_vec[word2] += float(el_split[1])
else:
word_vec[word] = float(el_split[1])
word_vec[word2] = float(el_split[1])
else:
word_vec[word] = 1.0
word_vec[word2] = 1.0

# matching terms to TWSI sense ids
scores = dict()
Expand All @@ -161,9 +164,11 @@ def load_sense_inventory(filename):

# assignment
assigned_id = get_max_score(scores)
sense_mappings[ident] = assigned_id
# assignment 'sense_mappings[ident] = ...' assumed ident's are unique over the whole inventory
sense_mappings[word+ident] = assigned_id
if d:
print "SCORES: "+str(scores)
# now this output works correctly (previously outputted the last element of the sense cluster)
print "ASSIGNED ID: "+word+" "+ident+"\t"+str(assigned_id)
print "\nLoading done\n"

Expand All @@ -177,7 +182,7 @@ def evaluate_predicted_labels(filename):
retrieved = 0
itemcount = 0
checked = set()
predictions = read_csv(filename, '/\t+/', encoding='utf8', header=0)
predictions = read_csv(filename, '/\t+/', encoding='utf8', header=0, engine="python")
for i,p in predictions.iterrows():
pred = p[0].split('\t')
key = str(pred[0])+str(pred[1])
Expand All @@ -186,16 +191,19 @@ def evaluate_predicted_labels(filename):
itemcount += 1
if oracle == "":
print "Sentence "+pred[0]+": Key '"+oracle+"' without sense assignment\n"
oracle_p = -1
# typo? oracle_p mentioned nowhere else
# also keep 'oracle' a string
oracle = '-1'
if key not in checked:
checked.add(key)
if oracle in sense_mappings and gold == sense_mappings[oracle]:
# adaptation of get operation to sense_mapping according with the change in load_sense_inventory function
if pred[1] + oracle in sense_mappings and gold == sense_mappings[pred[1] + oracle]:
correct += 1
if int(oracle) > -1:
retrieved += 1
if d:
if oracle in sense_mappings:
print "Sentence: "+key+"\tPrediction: "+oracle+"\tGold: "+key+"\tPredicted_TWSI_sense: "+str(sense_mappings[oracle])+"\tMatch:"+str(gold == sense_mappings[oracle])
if pred[1] + oracle in sense_mappings:
print "Sentence: "+key+"\tPrediction: "+oracle+"\tGold: "+key+"\tPredicted_TWSI_sense: "+str(sense_mappings[pred[1] + oracle])+"\tMatch:"+str(gold == sense_mappings[pred[1] + oracle])
else:
print "Sentence: "+key+"\tPrediction: "+oracle+"\tGold: "+key+"\tPredicted_TWSI_sense: "+"none"+"\tMatch: False"

Expand Down