Files
timetable/data/dummy/generate_dummy_data.py
Talhadeveloperr 9079ba2739 succesfull testing
2026-09-09 18:55:53 +05:00

252 lines
10 KiB
Python

"""
Generates a larger, deliberately harder dummy CSV dataset under data/dummy/
so the timetable template generator has something non-trivial to schedule
and render — more departments/batches/sections than the previous small
dataset, tighter room and instructor contention (fewer rooms/instructors
relative to demand), a second block, and mixed lecture durations (90 vs
120 minutes) so weaknesses in the scheduler surface as WARNINGs/conflicts
instead of everything trivially fitting.
Run once:
python data/dummy/generate_dummy_data.py
"""
import csv
import os
BASE_DIR = os.path.dirname(os.path.abspath(__file__))
def write_csv(relative_path, header, rows):
path = os.path.join(BASE_DIR, relative_path)
os.makedirs(os.path.dirname(path), exist_ok=True)
with open(path, "w", newline="", encoding="utf-8") as f:
writer = csv.writer(f)
writer.writerow(header)
writer.writerows(rows)
print(f"wrote {len(rows)} rows -> {path}")
def main():
# ---- blocks: two blocks now, Block A (bigger) and Block B (smaller) ----
write_csv(
"blocks/blocks.csv",
["block_id", "block_name"],
[
["A", "Block A"],
["B", "Block B"],
],
)
# ---- rooms: tight enough to keep real contention, but with enough
# total capacity that a correct solver can always find a fit (room
# capacity/student-count is not a constraint here) ----
write_csv(
"blocks/rooms.csv",
["room_id", "room_name", "block_id", "room_type", "capacity"],
[
["A-101", "Room 101", "A", "classroom", 60],
["A-102", "Room 102", "A", "classroom", 60],
["A-103", "Room 103", "A", "classroom", 50],
["A-201", "Room 201", "A", "classroom", 60],
["A-202", "Room 202", "A", "classroom", 60],
["A-L01", "Computer Lab 1", "A", "lab", 35],
["A-L02", "Computer Lab 2", "A", "lab", 35],
["A-L03", "Computer Lab 3", "A", "lab", 35],
["B-101", "Room 101 (B)", "B", "classroom", 50],
["B-102", "Room 102 (B)", "B", "classroom", 50],
["B-L01", "Electronics Lab", "B", "lab", 30],
],
)
# ---- departments / programs / batches: 3 departments, 2 programs
# each, 2 sessions (25/26) -> 12 batches total (was 6) ----
department_programs = {
"CS": ["BSCS", "BSSE"],
"EE": ["BSEE", "BSCE"],
"BBA": ["BBA", "BSAF"],
}
program_codes = {
"BSCS": "021519", "BSSE": "021520",
"BSEE": "031519", "BSCE": "031520",
"BBA": "041519", "BSAF": "041520",
}
batches_25 = []
batches_26 = []
for department, programs in department_programs.items():
for program in programs:
code = program_codes[program]
batches_25.append([f"25{code}", department, program, "25"])
batches_26.append([f"26{code}", department, program, "26"])
write_csv(
"batch_codes/25/batches.csv",
["batch_code", "department", "program", "session"],
batches_25,
)
write_csv(
"batch_codes/26/batches.csv",
["batch_code", "department", "program", "session"],
batches_26,
)
all_batches = batches_25 + batches_26
# ---- courses ----
# course_id, course_name, credit_hours, lectures_per_week,
# labs_per_week, is_general, lecture_room_id, lab_room_id,
# lecture_duration_minutes
#
# Mixed lecture durations: most courses are 90-min x N/week, but a few
# are the special "2 hours once a week" case (lecture_duration_minutes
# 120, lectures_per_week 1). A course's lab, if it has one, is always
# exactly once a week (180 minutes) — never more. Deliberately more
# courses per department than the room/instructor pool can comfortably
# absorb without contention.
courses = [
# CS department courses — lab rooms spread across A-L01/A-L02/A-L03
# so no single lab room is asked for more sessions than exist
# weekly slots for it (5 days x 2 lab-slot-pairs = 10/week).
["CS101", "Programming Fundamentals", 4, 2, 1, "no", "A-101", "A-L01", 90],
["CS201", "Data Structures", 4, 2, 1, "no", "A-102", "A-L02", 90],
["CS301", "Operating Systems", 3, 1, 1, "no", "A-101", "A-L03", 120],
["CS302", "Database Systems", 4, 2, 1, "no", "A-102", "A-L01", 90],
# EE department courses
["EE101", "Basic Electronics", 4, 2, 1, "no", "A-201", "A-L02", 90],
["EE201", "Digital Logic Design", 4, 2, 1, "no", "B-101", "B-L01", 90],
["EE301", "Signals and Systems", 3, 1, 0, "no", "B-102", "", 120],
# BBA department courses
["BBA101", "Principles of Management", 3, 2, 0, "no", "B-101", "", 90],
["BBA201", "Financial Accounting", 3, 2, 0, "no", "B-102", "", 90],
["BBA301", "Marketing Management", 3, 1, 0, "no", "A-103", "", 120],
# General (shared) courses
["GEN101", "Functional English", 3, 2, 0, "yes", "A-103", "", 90],
["GEN102", "Islamic Studies", 2, 1, 0, "yes", "A-103", "", 90],
["GEN103", "Pakistan Studies", 2, 1, 0, "yes", "A-202", "", 120],
]
write_csv(
"courses/courses.csv",
[
"course_id", "course_name", "credit_hours", "lectures_per_week",
"labs_per_week", "is_general", "lecture_room_id", "lab_room_id",
"lecture_duration_minutes",
],
courses,
)
# ---- course offerings: course + section (batches in that section) +
# instructor ----
# Deliberately tight instructor pool reused across many sections/
# courses (some instructors will end up wanted in overlapping slots),
# forcing real contention in Phase 2/Phase 3.
dept_courses = {
"CS": ["CS101", "CS201", "CS301", "CS302"],
"EE": ["EE101", "EE201", "EE301"],
"BBA": ["BBA101", "BBA201", "BBA301"],
}
# One instructor per course within a department (so no instructor ever
# teaches two different courses to the same section/batch, which would
# double their weekly slot demand against that batch's narrow shared
# availability) — still a tight pool overall since each instructor
# teaches their course across every section of the department.
dept_instructors = {
"CS": ["1001", "1002", "1003", "1009"],
"EE": ["1004", "1005", "1010"],
"BBA": ["1007", "1008", "1011"],
}
# CS302 additionally alternates between two instructors across its
# sections (instead of one instructor owning all 4) since it has the
# heaviest per-section load (2 lectures + 2 labs/week) — one person
# covering every section of it would need more slot-0 room than a
# single instructor realistically has once other batches' commitments
# are accounted for.
course_instructor_rotation = {"CS302": ["1009", "1012"]}
instructor_names_extra = {"1012": "Ms. Rabia Nasir"}
offerings = []
section_letters = ["A", "B", "C", "D"]
dept_section_counter = {"CS": 0, "EE": 0, "BBA": 0}
for batch_code, department, program, session in all_batches:
section_index = dept_section_counter[department]
section_id = section_letters[section_index % len(section_letters)]
dept_section_counter[department] += 1
instr_pool = dept_instructors[department]
# One instructor per course, consistent across every section, so
# no instructor ever teaches two different courses to the same
# batch (which would silently double their contested slot-0
# demand against that one batch). The pool is still tight overall
# since each instructor teaches their course to every section of
# the department -> real instructor contention across sections.
for i, course_id in enumerate(dept_courses[department]):
if course_id in course_instructor_rotation:
rotation = course_instructor_rotation[course_id]
instr = rotation[section_index % len(rotation)]
else:
instr = instr_pool[i % len(instr_pool)]
offerings.append([course_id, section_id, batch_code, instr])
# general courses: one offering per (course, session), section "ALL"
# grouping every batch code sharing that session
for session, batch_group in (
("25", [b[0] for b in batches_25]),
("26", [b[0] for b in batches_26]),
):
offerings.append(["GEN101", "ALL", "|".join(batch_group), "1002"])
offerings.append(["GEN102", "ALL", "|".join(batch_group), "1006"])
offerings.append(["GEN103", "ALL", "|".join(batch_group), "1006"])
write_csv(
"courses/course_offering.csv",
["course_id", "section_id", "batch_codes", "instructor_code"],
offerings,
)
# ---- instructors (4-digit employee codes) ----
# Derived from the offerings above so it can never drift: one row per
# unique (instructor, course, section) teaching assignment.
instructor_names = {
"1001": "Dr. Ahsan Raza",
"1002": "Ms. Sana Tariq",
"1003": "Mr. Bilal Khan",
"1004": "Dr. Farah Iqbal",
"1005": "Mr. Usman Ali",
"1006": "Ms. Hira Shah",
"1007": "Dr. Nadia Farooq",
"1008": "Mr. Kamran Sheikh",
"1009": "Ms. Ayesha Malik",
"1010": "Mr. Imran Qureshi",
"1011": "Dr. Saima Aziz",
**instructor_names_extra,
}
instructors = []
for course_id, section_id, batch_codes, instr_code in offerings:
instructors.append([instr_code, instructor_names[instr_code], course_id, section_id])
write_csv(
"instructor/instructors.csv",
["instructor_code", "instructor_name", "course_id", "section_id"],
instructors,
)
# ---- students (roll_no = batch_code-seq): 18 per batch (was 5) ----
students = []
for batch_code, department, program, session in all_batches:
for seq in range(1, 19):
roll_no = f"{batch_code}-{seq:03d}"
students.append([roll_no, batch_code, f"Student {roll_no}"])
write_csv(
"roll_no/students.csv",
["roll_no", "batch_code", "student_name"],
students,
)
print("\nDummy data generation complete.")
if __name__ == "__main__":
main()