mgaughan/mw-lifecycle-analysis
1
0
Fork 0
Code

scaffolding out some work TODO on getting the olmo categories to be sentence-level

Browse Source
This commit is contained in:
mgaughan 2025-09-02 12:48:11 -05:00
parent 53775c51db
commit 2d396ceb26
1 changed files with 8 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
p2/quest/python_scripts/info_labeling.py
View File

@ -19,7 +19,7 @@ tokenizer = AutoTokenizer.from_pretrained("allenai/OLMo-2-1124-13B")
#TODO: text_preprocessing per https://arxiv.org/pdf/1902.07093
priming = "For the **GIVEN COMMENT**, please categorize it into one of the defined [[CATEGORIES]]. Each [[CATEGORY]] is described in the TYPOLOGY for reference.Your task is to match the **GIVEN COMMENT** to the **[[CATEGORY]]** that most accurately describes the content of the comment. Only provide the category as your output. Do not provide any text beyond the category name."
priming = "For the **GIVEN SENTENCE**, please categorize it into one of the defined [[CATEGORIES]]. Each [[CATEGORY]] is described in the TYPOLOGY for reference.Your task is to match the **GIVEN SENTENCE** to the **[[CATEGORY]]** that most accurately describes the content of the comment. Only provide the category as your output. Do not provide any text beyond the category name."
#the typology descriptions are taken straight from https://arxiv.org/pdf/1902.07093
typology = """
TYPOLOGY:
@ -58,7 +58,7 @@ TYPOLOGY:
"""
#instructions="Only respond with the GIVEN COMMENT's [[CATEGORY]] classification. Do not provide any more information."
instructions="The comment's category is: "
instructions="The sentence's category is: "
with open("/home/nws8519/git/mw-lifecycle-analysis/p2/quest/072325_biberplus_labels.csv", mode='r', newline='') as file:
reader = csv.reader(file)
@ -93,8 +93,10 @@ with open("/home/nws8519/git/mw-lifecycle-analysis/p2/quest/072325_biberplus_lab
comment_text = re.sub(url_pattern, 'URL', comment_text)
# 4. if possible, replace @ with SCREEN_NAME
comment_text = re.sub(r'(^|\s)@\w+', 'SCREEN_NAME', comment_text)
# TODO: 5. split into an array of sentences
#print(comment_text)
#TODO: do this for each sentence in the comment
text_dict['cleaned_comment_text'] = comment_text
#build out prompt construction; more specificity in data provided
given_data = f"**GIVEN COMMENT: \n ' Type -{text_dict['comment_type']} \n Text -{text_dict['cleaned_comment_text']}**'\n"
@ -106,7 +108,6 @@ with open("/home/nws8519/git/mw-lifecycle-analysis/p2/quest/072325_biberplus_lab
#deterministic sampling and getting the response back
response = olmo.generate(**inputs, max_new_tokens=256, do_sample=False)
response_txt = tokenizer.batch_decode(response, skip_special_tokens=True)[0]
#print("this is the response:::: ----------------------------")
#print(response_txt)
#getting the resulting codes
@ -118,6 +119,9 @@ with open("/home/nws8519/git/mw-lifecycle-analysis/p2/quest/072325_biberplus_lab
else:
following_text = "NO CATEGORY"
#print(following_text)
# TODO: collate olmo categories back together into an ordered list
# TODO: add the list of sentence-level olmo categories into dictionary
text_dict['olmo_category'] = following_text
'''
for item in result.strip(";").split(";"):
@ -130,6 +134,6 @@ with open("/home/nws8519/git/mw-lifecycle-analysis/p2/quest/072325_biberplus_lab
array_of_categorizations.append(text_dict)
#CSV everything
df = pd.DataFrame(array_of_categorizations)
df.to_csv('072525_olmo_messages_categorized.csv', index=False)
#df.to_csv('072525_olmo_messages_categorized.csv', index=False)