devops/encoach_backend
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add regular ielts modules to custom level.

Browse Source
This commit is contained in:
Cristiano Ferreira
2024-08-15 13:58:07 +01:00
parent beccf8b501
commit d68617f33b
2 changed files with 573 additions and 299 deletions
Download Patch File Download Diff File Expand all files Collapse all files

366
helper/exercises.py
View File

@@ -15,19 +15,19 @@ from helper.speech_to_text_helper import has_x_words
nltk.download('words')
def gen_reading_passage_1(topic, req_exercises, difficulty):
def gen_reading_passage_1(topic, difficulty, req_exercises, number_of_exercises_q=queue.Queue(), start_id=1):
if (len(req_exercises) == 0):
req_exercises = random.sample(READING_EXERCISE_TYPES, 2)
number_of_exercises_q = divide_number_into_parts(TOTAL_READING_PASSAGE_1_EXERCISES, len(req_exercises))
if (number_of_exercises_q.empty()):
number_of_exercises_q = divide_number_into_parts(TOTAL_READING_PASSAGE_1_EXERCISES, len(req_exercises))
passage = generate_reading_passage_1_text(topic)
if passage == "":
return gen_reading_passage_1(topic, req_exercises, difficulty)
start_id = 1
return gen_reading_passage_1(topic, difficulty, req_exercises, number_of_exercises_q, start_id)
exercises = generate_reading_exercises(passage["text"], req_exercises, number_of_exercises_q, start_id, difficulty)
if contains_empty_dict(exercises):
return gen_reading_passage_1(topic, req_exercises, difficulty)
return gen_reading_passage_1(topic, difficulty, req_exercises, number_of_exercises_q, start_id)
return {
"exercises": exercises,
"text": {
@@ -38,19 +38,19 @@ def gen_reading_passage_1(topic, req_exercises, difficulty):
}
def gen_reading_passage_2(topic, req_exercises, difficulty):
def gen_reading_passage_2(topic, difficulty, req_exercises, number_of_exercises_q=queue.Queue(), start_id=14):
if (len(req_exercises) == 0):
req_exercises = random.sample(READING_EXERCISE_TYPES, 2)
number_of_exercises_q = divide_number_into_parts(TOTAL_READING_PASSAGE_2_EXERCISES, len(req_exercises))
if (number_of_exercises_q.empty()):
number_of_exercises_q = divide_number_into_parts(TOTAL_READING_PASSAGE_2_EXERCISES, len(req_exercises))
passage = generate_reading_passage_2_text(topic)
if passage == "":
return gen_reading_passage_2(topic, req_exercises, difficulty)
start_id = 14
return gen_reading_passage_2(topic, difficulty, req_exercises, number_of_exercises_q, start_id)
exercises = generate_reading_exercises(passage["text"], req_exercises, number_of_exercises_q, start_id, difficulty)
if contains_empty_dict(exercises):
return gen_reading_passage_2(topic, req_exercises, difficulty)
return gen_reading_passage_2(topic, difficulty, req_exercises, number_of_exercises_q, start_id)
return {
"exercises": exercises,
"text": {
@@ -61,19 +61,19 @@ def gen_reading_passage_2(topic, req_exercises, difficulty):
}
def gen_reading_passage_3(topic, req_exercises, difficulty):
def gen_reading_passage_3(topic, difficulty, req_exercises, number_of_exercises_q=queue.Queue(), start_id=27):
if (len(req_exercises) == 0):
req_exercises = random.sample(READING_EXERCISE_TYPES, 2)
number_of_exercises_q = divide_number_into_parts(TOTAL_READING_PASSAGE_3_EXERCISES, len(req_exercises))
if (number_of_exercises_q.empty()):
number_of_exercises_q = divide_number_into_parts(TOTAL_READING_PASSAGE_3_EXERCISES, len(req_exercises))
passage = generate_reading_passage_3_text(topic)
if passage == "":
return gen_reading_passage_3(topic, req_exercises, difficulty)
start_id = 27
return gen_reading_passage_3(topic, difficulty, req_exercises, number_of_exercises_q, start_id)
exercises = generate_reading_exercises(passage["text"], req_exercises, number_of_exercises_q, start_id, difficulty)
if contains_empty_dict(exercises):
return gen_reading_passage_3(topic, req_exercises, difficulty)
return gen_reading_passage_3(topic, difficulty, req_exercises, number_of_exercises_q, start_id)
return {
"exercises": exercises,
"text": {
@@ -865,7 +865,8 @@ def gen_idea_match_exercise(text: str, quantity: int, start_id):
{
"role": "user",
"content": (
'From the text extract ' + str(quantity) + ' ideas, theories, opinions and who they are from. The text: ' + str(text))
'From the text extract ' + str(
quantity) + ' ideas, theories, opinions and who they are from. The text: ' + str(text))
}
]
@@ -882,6 +883,7 @@ def gen_idea_match_exercise(text: str, quantity: int, start_id):
"type": "matchSentences"
}
def build_options(ideas):
options = []
letters = iter(string.ascii_uppercase)
@@ -892,6 +894,7 @@ def build_options(ideas):
})
return options
def build_sentences(ideas, start_id):
sentences = []
letters = iter(string.ascii_uppercase)
@@ -906,6 +909,7 @@ def build_sentences(ideas, start_id):
sentence["id"] = i
return sentences
def assign_letters_to_paragraphs(paragraphs):
result = []
letters = iter(string.ascii_uppercase)
@@ -1272,7 +1276,8 @@ def replace_exercise_if_exists(all_exams, current_exercise, current_exam, seen_k
current_exercise["options"])
for exercise in exercise_dict.get("exercises", [])[0]["questions"]
):
return replace_exercise_if_exists(all_exams, generate_single_mc_level_question(), current_exam, seen_keys)
return replace_exercise_if_exists(all_exams, generate_single_mc_level_question(), current_exam,
seen_keys)
return current_exercise, seen_keys
@@ -1302,7 +1307,8 @@ def replace_blank_space_exercise_if_exists_utas(all_exams, current_exercise, cur
key = (current_exercise['prompt'], tuple(sorted(option['text'] for option in current_exercise['options'])))
# Check if the key is in the set
if key in seen_keys:
return replace_exercise_if_exists_utas(all_exams, generate_single_mc_blank_space_level_question(), current_exam, seen_keys)
return replace_exercise_if_exists_utas(all_exams, generate_single_mc_blank_space_level_question(), current_exam,
seen_keys)
else:
seen_keys.add(key)
@@ -1313,7 +1319,8 @@ def replace_blank_space_exercise_if_exists_utas(all_exams, current_exercise, cur
current_exercise["options"])
for exercise in exam.get("questions", [])
):
return replace_exercise_if_exists_utas(all_exams, generate_single_mc_blank_space_level_question(), current_exam,
return replace_exercise_if_exists_utas(all_exams, generate_single_mc_blank_space_level_question(),
current_exam,
seen_keys)
return current_exercise, seen_keys
@@ -1323,7 +1330,8 @@ def replace_underlined_exercise_if_exists_utas(all_exams, current_exercise, curr
key = (current_exercise['prompt'], tuple(sorted(option['text'] for option in current_exercise['options'])))
# Check if the key is in the set
if key in seen_keys:
return replace_exercise_if_exists_utas(all_exams, generate_single_mc_underlined_level_question(), current_exam, seen_keys)
return replace_exercise_if_exists_utas(all_exams, generate_single_mc_underlined_level_question(), current_exam,
seen_keys)
else:
seen_keys.add(key)
@@ -1334,7 +1342,8 @@ def replace_underlined_exercise_if_exists_utas(all_exams, current_exercise, curr
current_exercise["options"])
for exercise in exam.get("questions", [])
):
return replace_exercise_if_exists_utas(all_exams, generate_single_mc_underlined_level_question(), current_exam,
return replace_exercise_if_exists_utas(all_exams, generate_single_mc_underlined_level_question(),
current_exam,
seen_keys)
return current_exercise, seen_keys
@@ -1376,8 +1385,8 @@ def generate_single_mc_blank_space_level_question():
},
{
"role": "user",
"content": ('Generate 1 multiple choice blank space question of 4 options for an english level exam, it can be easy, '
'intermediate or advanced.')
"content": ('Generate 1 multiple choice blank space question of 4 options for an english level exam, '
'it can be easy, intermediate or advanced.')
}
]
@@ -1401,8 +1410,8 @@ def generate_single_mc_underlined_level_question():
},
{
"role": "user",
"content": ('Generate 1 multiple choice blank space question of 4 options for an english level exam, it can be easy, '
'intermediate or advanced.')
"content": ('Generate 1 multiple choice blank space question of 4 options for an english level exam, '
'it can be easy, intermediate or advanced.')
},
{
@@ -1469,9 +1478,9 @@ def gen_multiple_choice_blank_space_utas(quantity: int, start_id: int, all_exams
if all_exams is not None:
seen_keys = set()
for i in range(len(question["questions"])):
question["questions"][i], seen_keys = replace_blank_space_exercise_if_exists_utas(all_exams, question["questions"][i],
question,
seen_keys)
question["questions"][i], seen_keys = (
replace_blank_space_exercise_if_exists_utas(all_exams, question["questions"][i], question,
seen_keys))
response = fix_exercise_ids(question, start_id)
response["questions"] = randomize_mc_options_order(response["questions"])
return response
@@ -1546,11 +1555,9 @@ def gen_multiple_choice_underlined_utas(quantity: int, start_id: int, all_exams=
if all_exams is not None:
seen_keys = set()
for i in range(len(question["questions"])):
question["questions"][i], seen_keys = replace_underlined_exercise_if_exists_utas(all_exams,
question["questions"][
i],
question,
seen_keys)
question["questions"][i], seen_keys = (
replace_underlined_exercise_if_exists_utas(all_exams, question["questions"][i], question,
seen_keys))
response = fix_exercise_ids(question, start_id)
response["questions"] = randomize_mc_options_order(response["questions"])
return response
@@ -1765,7 +1772,8 @@ def generate_level_mc(start_id: int, quantity: int, all_questions=None):
if all_questions is not None:
seen_keys = set()
for i in range(len(question["questions"])):
question["questions"][i], seen_keys = replace_exercise_if_exists_utas(all_questions, question["questions"][i],
question["questions"][i], seen_keys = replace_exercise_if_exists_utas(all_questions,
question["questions"][i],
question,
seen_keys)
response = fix_exercise_ids(question, start_id)
@@ -1791,3 +1799,293 @@ def randomize_mc_options_order(questions):
question['solution'] = option['id']
return questions
def gen_writing_task_1(topic, difficulty):
messages = [
{
"role": "system",
"content": ('You are a helpful assistant designed to output JSON on this format: '
'{"prompt": "prompt content"}')
},
{
"role": "user",
"content": ('Craft a prompt for an IELTS Writing Task 1 General Training exercise that instructs the '
'student to compose a letter. The prompt should present a specific scenario or situation, '
'based on the topic of "' + topic + '", requiring the student to provide information, '
'advice, or instructions within the letter. '
'Make sure that the generated prompt is '
'of ' + difficulty + 'difficulty and does not contain '
'forbidden subjects in muslim '
'countries.')
},
{
"role": "user",
"content": 'The prompt should end with "In the letter you should" followed by 3 bullet points of what '
'the answer should include.'
}
]
token_count = count_total_tokens(messages)
response = make_openai_call(GPT_3_5_TURBO, messages, token_count, "prompt",
GEN_QUESTION_TEMPERATURE)
return {
"question": add_newline_before_hyphen(response["prompt"].strip()),
"difficulty": difficulty,
"topic": topic
}
def add_newline_before_hyphen(s):
return s.replace(" -", "\n-")
def gen_writing_task_2(topic, difficulty):
messages = [
{
"role": "system",
"content": ('You are a helpful assistant designed to output JSON on this format: '
'{"prompt": "prompt content"}')
},
{
"role": "user",
"content": (
'Craft a comprehensive question of ' + difficulty + 'difficulty like the ones for IELTS Writing '
'Task 2 General Training that directs the '
'candidate'
'to delve into an in-depth analysis of '
'contrasting perspectives on the topic '
'of "' + topic + '". The candidate should be '
'asked to discuss the '
'strengths and weaknesses of '
'both viewpoints.')
},
{
"role": "user",
"content": 'The question should lead to an answer with either "theories", "complicated information" or '
'be "very descriptive" on the topic.'
}
]
token_count = count_total_tokens(messages)
response = make_openai_call(GPT_4_O, messages, token_count, "prompt", GEN_QUESTION_TEMPERATURE)
return {
"question": response["prompt"].strip(),
"difficulty": difficulty,
"topic": topic
}
def gen_speaking_part_1(first_topic: str, second_topic: str, difficulty):
json_format = {
"first_topic": "topic 1",
"second_topic": "topic 2",
"questions": [
"Introductory question about the first topic, starting the topic with 'Let's talk about x' and then the "
"question.",
"Follow up question about the first topic",
"Follow up question about the first topic",
"Question about second topic",
"Follow up question about the second topic",
]
}
messages = [
{
"role": "system",
"content": (
'You are a helpful assistant designed to output JSON on this format: ' + str(json_format))
},
{
"role": "user",
"content": (
'Craft 5 simple and single questions of easy difficulty for IELTS Speaking Part 1 '
'that encourages candidates to delve deeply into '
'personal experiences, preferences, or insights on the topic '
'of "' + first_topic + '" and the topic of "' + second_topic + '". '
'Make sure that the generated '
'question'
'does not contain forbidden '
'subjects in'
'muslim countries.')
},
{
"role": "user",
"content": 'The questions should lead to the usage of 4 verb tenses (present perfect, present, '
'past and future).'
},
{
"role": "user",
"content": 'They must be 1 single question each and not be double-barreled questions.'
}
]
token_count = count_total_tokens(messages)
response = make_openai_call(GPT_4_O, messages, token_count, ["first_topic"],
GEN_QUESTION_TEMPERATURE)
response["type"] = 1
response["difficulty"] = difficulty
return response
def gen_speaking_part_2(topic: str, difficulty):
json_format = {
"topic": "topic",
"question": "question",
"prompts": [
"prompt_1",
"prompt_2",
"prompt_3"
],
"suffix": "And explain why..."
}
messages = [
{
"role": "system",
"content": 'You are a helpful assistant designed to output JSON on this format: ' + str(json_format)
},
{
"role": "user",
"content": (
'Create a question of medium difficulty for IELTS Speaking Part 2 '
'that encourages candidates to narrate a '
'personal experience or story related to the topic '
'of "' + topic + '". Include 3 prompts that '
'guide the candidate to describe '
'specific aspects of the experience, '
'such as details about the situation, '
'their actions, and the reasons it left a '
'lasting impression. Make sure that the '
'generated question does not contain '
'forbidden subjects in muslim countries.')
},
{
"role": "user",
"content": 'The prompts must not be questions. Also include a suffix like the ones in the IELTS exams '
'that start with "And explain why".'
}
]
token_count = count_total_tokens(messages)
response = make_openai_call(GPT_4_O, messages, token_count, GEN_FIELDS, GEN_QUESTION_TEMPERATURE)
response["type"] = 2
response["difficulty"] = difficulty
response["topic"] = topic
return response
def gen_speaking_part_3(topic: str, difficulty):
json_format = {
"topic": "topic",
"questions": [
"Introductory question about the topic.",
"Follow up question about the topic",
"Follow up question about the topic",
"Follow up question about the topic",
"Follow up question about the topic"
]
}
messages = [
{
"role": "system",
"content": (
'You are a helpful assistant designed to output JSON on this format: ' + str(json_format))
},
{
"role": "user",
"content": (
'Formulate a set of 5 single questions of hard difficulty for IELTS Speaking Part 3 that encourage candidates to engage in a '
'meaningful discussion on the topic of "' + topic + '". Provide inquiries, ensuring '
'they explore various aspects, perspectives, and implications related to the topic.'
'Make sure that the generated question does not contain forbidden subjects in muslim countries.')
},
{
"role": "user",
"content": 'They must be 1 single question each and not be double-barreled questions.'
}
]
token_count = count_total_tokens(messages)
response = make_openai_call(GPT_4_O, messages, token_count, GEN_FIELDS, GEN_QUESTION_TEMPERATURE)
# Remove the numbers from the questions only if the string starts with a number
response["questions"] = [re.sub(r"^\d+\.\s*", "", question) if re.match(r"^\d+\.", question) else question for
question in response["questions"]]
response["type"] = 3
response["difficulty"] = difficulty
response["topic"] = topic
return response
def gen_listening_section_1(topic, difficulty, req_exercises, number_of_exercises_q=queue.Queue(), start_id=1):
if (len(req_exercises) == 0):
req_exercises = random.sample(LISTENING_1_EXERCISE_TYPES, 1)
if (number_of_exercises_q.empty()):
number_of_exercises_q = divide_number_into_parts(TOTAL_LISTENING_SECTION_1_EXERCISES, len(req_exercises))
processed_conversation = generate_listening_1_conversation(topic)
exercises = generate_listening_conversation_exercises(parse_conversation(processed_conversation),
req_exercises,
number_of_exercises_q,
start_id, difficulty)
return {
"exercises": exercises,
"text": processed_conversation,
"difficulty": difficulty
}
def gen_listening_section_2(topic, difficulty, req_exercises, number_of_exercises_q=queue.Queue(), start_id=11):
if (len(req_exercises) == 0):
req_exercises = random.sample(LISTENING_2_EXERCISE_TYPES, 2)
if (number_of_exercises_q.empty()):
number_of_exercises_q = divide_number_into_parts(TOTAL_LISTENING_SECTION_2_EXERCISES, len(req_exercises))
monologue = generate_listening_2_monologue(topic)
exercises = generate_listening_monologue_exercises(str(monologue), req_exercises, number_of_exercises_q,
start_id, difficulty)
return {
"exercises": exercises,
"text": monologue,
"difficulty": difficulty
}
def gen_listening_section_3(topic, difficulty, req_exercises, number_of_exercises_q=queue.Queue(), start_id=21):
if (len(req_exercises) == 0):
req_exercises = random.sample(LISTENING_3_EXERCISE_TYPES, 1)
if (number_of_exercises_q.empty()):
number_of_exercises_q = divide_number_into_parts(TOTAL_LISTENING_SECTION_3_EXERCISES, len(req_exercises))
processed_conversation = generate_listening_3_conversation(topic)
exercises = generate_listening_conversation_exercises(parse_conversation(processed_conversation), req_exercises,
number_of_exercises_q,
start_id, difficulty)
return {
"exercises": exercises,
"text": processed_conversation,
"difficulty": difficulty
}
def gen_listening_section_4(topic, difficulty, req_exercises, number_of_exercises_q=queue.Queue(), start_id=31):
if (len(req_exercises) == 0):
req_exercises = random.sample(LISTENING_EXERCISE_TYPES, 2)
if (number_of_exercises_q.empty()):
number_of_exercises_q = divide_number_into_parts(TOTAL_LISTENING_SECTION_4_EXERCISES, len(req_exercises))
monologue = generate_listening_4_monologue(topic)
exercises = generate_listening_monologue_exercises(str(monologue), req_exercises, number_of_exercises_q,
start_id, difficulty)
return {
"exercises": exercises,
"text": monologue,
"difficulty": difficulty
}