ff1c23eaec
import_examgoal.py required /tmp/pyqeval helper modules that were never committed, so it failed on any fresh machine. The two upstream model files now live in scripts/examgoal_model/ (relative import fixed) and the default pickle path is scripts/examgoal_db.pkl.
76 lines
No EOL
3.2 KiB
Python
76 lines
No EOL
3.2 KiB
Python
"""
|
|
This file has Question class
|
|
"""
|
|
|
|
#from sentence_transformers import SentenceTransformer
|
|
|
|
class Question:
|
|
"""Loads data from database into inherited class variables"""
|
|
def __repr__(self)->str:
|
|
template = f"""
|
|
QuestionId: {self.question_id}
|
|
Exam: {self.exam}
|
|
Year: {self.year}
|
|
Subject: {self.subject}
|
|
Chapter: {self.chapter}
|
|
"""
|
|
return template
|
|
def __init__(self, question_json: dict,embedding_model:str="all-MiniLM-L6-v2") -> None:
|
|
"""Initialization of Question
|
|
:param:
|
|
question_json: json part of question (this is kind of hardcoded configured to examgoal database
|
|
and the structure of this system)
|
|
embedding_model: A vail SentenceTransformer embedding model name . default to all-MiniLM-L6-v2
|
|
"""
|
|
#model = SentenceTransformer(embedding_model)
|
|
|
|
self.question_id = question_json.get("question_id", "")
|
|
self.examGroup = question_json.get("examGroup", "")
|
|
self.exam = question_json.get("exam", "")
|
|
self.subject = question_json.get("subject", "")
|
|
self.chpaterGroup = question_json.get("chapterGroup", "")
|
|
self.chapter = question_json.get("chapter", "")
|
|
self.year = question_json.get("year", 0)
|
|
self.paperTitle = question_json.get("paperTitle", "")
|
|
self.difficulty = question_json.get("difficulty", "")
|
|
self.topic = question_json.get("topic", "")
|
|
self.type = question_json.get("type", "")
|
|
self.examDate = question_json.get("examDate", None)
|
|
self.answer = question_json.get("answer",None)
|
|
question_en = question_json.get("question", {}).get("en", {})
|
|
self.question = question_en.get("content", "")
|
|
self.options = question_en.get("options", {})
|
|
self.correct_options = question_en.get("correct_options", [])
|
|
self.explanation = question_en.get("explanation", "")
|
|
|
|
self.isOutOfSyllabus = question_json.get("isOutOfSyllabus", False)
|
|
self.isBonus = question_json.get("isBonus", False)
|
|
|
|
|
|
self.isImgQuestion = self.check_image_in_text(self.question)
|
|
self.isImgExplanation = self.check_image_in_text(self.explanation)
|
|
self.isImgOption = self.check_image_in_options()
|
|
|
|
#self.embedding = model.encode(self.question) #disbaled for v003 testing version | this is a computation
|
|
#heavy param will take 8 mins to calc the embeddings | latest update moving embeddings to its seprate cache!!!
|
|
|
|
def check_image_in_text(self,text:str)->bool:
|
|
"""Check if text consists of image (inform of html)"""
|
|
text_s = text.split("<img")
|
|
if len(text_s) != 1:
|
|
return True
|
|
else:
|
|
return False
|
|
|
|
def check_image_in_options(self)->bool:
|
|
"""Check if question consists of image (inform of html)"""
|
|
options_json = self.options
|
|
option_bool_list = []
|
|
for option in options_json:
|
|
option_content = option["content"]
|
|
option_content_s = option_content.split("<img")
|
|
if len(option_content_s) != 1:
|
|
option_bool_list.append(True)
|
|
else:
|
|
option_bool_list.append(False)
|
|
return option_bool_list |