added scripts

Author: exponentian

scripts/script1.py

@@ -0,0 +1,227 @@
+import csv
+import numpy as np
+
+
+PATH = './'
+
+MAX_PER_PROJECT = 2
+MIN_PROPOSAL_VALUE = 6
+MIN_SATISFACTION_VALUE = 5
+
+
+def read_data_and_clean():
+    students = []
+    projects = []
+    preferences = {}
+    rankings = {}
+    count_map = {}
+
+    with open(PATH + 'data1/raw.csv', 'r') as file:
+        reader = csv.reader(file)
+        for i, row in enumerate(reader):
+            if i == 0:
+                projects = row[2:]
+                rankings = {project: [] for project in projects}
+                count_map = {project: {str(i): 0 for i in range(1, 100)} for project in projects}
+            else:
+                student = f'{row[0]} ({row[1]})'
+                if student:
+                    if student in students:
+                        print('Found duplicate student:', student)
+                    else:
+                        students.append(student)
+                        preferences[student] = {}
+                        for j, col in enumerate(row[2:]):
+                            if not col:
+                                col = str(len(row[2:]))
+                            project = projects[j]
+                            preferences[student][project] = col
+                            rankings[project].append((student, col))
+                            count_map[project][str(col)] += 1
+
+    c = 0
+    temp = {project: '' for project in projects}
+    for project, dict in count_map.items():
+        s = dict['1'] + dict['2']
+        temp[project] = s
+        c += 1
+
+    for ranking, items in rankings.items():
+        items.sort(key=lambda x: int(x[1]))
+
+    ranking_map = {project: {} for project in projects}
+    for project, items in rankings.items():
+        maxx = -1
+        rank = 1
+        for item in items:
+            student = item[0]
+            pref = item[1]
+
+            if maxx == -1:
+                maxx = int(pref)
+            elif int(pref) > maxx:
+                maxx = int(pref)
+                rank += 1
+
+            ranking_map[project][student] = str(rank)
+
+    max_per_projects = {}
+    exceptions = {'E': 3, 'F': 1, 'H': 4, 'I': 6, 'L': 1, 'M': 1}
+    for p in projects:
+        if p in list(exceptions.keys()):
+            max_per_projects[p] = exceptions[p]
+        else:
+            max_per_projects[p] = MAX_PER_PROJECT
+
+    return students, projects, max_per_projects, preferences, rankings, ranking_map
+
+def get_averages_of_preferences(projects, preferences):
+    totals = [0] * len(projects)
+    counts = [0] * len(projects)
+    averages = [0] * len(projects)
+
+    for st, prefs in preferences.items():
+        i = 0
+        for project_name, pref in prefs.items():
+            counts[i] += 1
+            totals[i] += int(pref)
+            i += 1
+
+    for i in range(len(totals)):
+        averages[i] = totals[i] / counts[i]
+
+    indexes = sorted(range(len(averages)), key=lambda i: averages[i], reverse=True)
+    
+    return indexes
+
+
+def calculate_averages_of_proposals(projects, allocations, proposals):
+    averages_dict = {project: 0.0 for project in projects}
+
+    for project, students in allocations.items():
+        total = 0
+        for student in students:
+            total += int(proposals[student])
+
+        averages_dict[project] = total / len(students)
+
+    averages = [avg for p, avg in averages_dict.items()]
+    indexes = sorted(range(len(averages)), key=lambda i: averages[i])
+    
+    averages_out = {}
+    for i in range(len(averages)):
+        averages_out[ projects[i] ] = averages[i]
+
+    return averages_out, indexes
+
+
+def matching_first_round(students, projects, max_per_projects, preferences, rankings, ranking_map):
+    indexes = get_averages_of_preferences(projects, preferences)
+
+    allocations = {project: [] for project in projects}
+    proposals = {student: '' for student in students}
+    ranking_allocations = {student: '' for student in students}
+    for i in indexes:
+        project = projects[i]
+        proejct_rankings = rankings[project]
+
+        for j, item in enumerate(proejct_rankings):
+            student = item[0]
+            pref = item[1]
+            if not proposals[student] and (pref in ['1','2','3','4','5']):
+                allocations[project].append(student)
+                proposals[student] = pref
+                ranking_allocations[student] = ranking_map[project][student]
+
+                allocations[project].sort(key=lambda s: int(preferences[s][project]))
+
+                if len(allocations[project]) > max_per_projects[project]:
+                    rejected_student = allocations[project].pop()
+                    proposals[rejected_student] = ''
+                    ranking_allocations[rejected_student] = ''
+
+    unassigned_students = []
+    for sid, pref in proposals.items():
+        if not pref:
+            unassigned_students.append(sid)
+
+    return allocations, proposals, ranking_allocations, unassigned_students
+
+
+def checking(projects, max_per_projects, preferences, allocations, proposals, ranking_allocations, unassigned_students):
+    students = []
+
+    for student in students:
+        this_group = ''
+        this_proposal = proposals[student]
+        this_prefs = dict(sorted(preferences[student].items(), key=lambda a: int(a[1])))
+
+        for project, members in allocations.items():
+            if student in members:
+                this_group = project
+                break
+
+        possible_projects = []
+        i = 0
+        for pro, pref in this_prefs.items():
+            if i < 3:
+                possible_projects.append(pro)
+            i += 1
+
+        for pro in possible_projects:
+            assigned_memnbers = allocations[pro]
+            for member in assigned_memnbers:
+                their_prefs = dict(sorted(preferences[member].items(), key=lambda a: int(a[1])))
+                if member not in students:
+                    print(member, their_prefs)
+
+
+def write_csv(allocations, proposals, ranking_allocations):
+    allocations_items = []
+    for project, students in allocations.items():
+        temp = []
+        temp.append(project)
+        for student in students:
+            temp.append(student)
+        allocations_items.append(temp)
+
+    save('allocations.csv', allocations_items)
+
+
+    proposals_items = []
+    for project, students in allocations.items():
+        temp = []
+        temp.append(project)
+        for student in students:
+            proposal = proposals[student]
+            temp.append(str(proposal))
+        proposals_items.append(temp)
+
+    save('proposals.csv', proposals_items)
+
+    ranking_allocations_items = []
+    for project, students in allocations.items():
+        temp = []
+        temp.append(project)
+        for student in students:
+            ranking_allocation = ranking_allocations[student]
+            temp.append(str(ranking_allocation))
+        ranking_allocations_items.append(temp)
+
+    save('ranking_allocations.csv', ranking_allocations_items)
+
+
+def save(filename, items):
+    with open(PATH + '/output1/' + filename, 'w', newline='') as file:
+        writer = csv.writer(file)
+        for item in items:
+            writer.writerow(item)
+
+
+if __name__ == '__main__':
+    students, projects, max_per_projects, preferences, rankings, ranking_map = read_data_and_clean()
+
+    allocations, proposals, ranking_allocations, unassigned_students = matching_first_round(students, projects, max_per_projects, preferences, rankings, ranking_map)
+    # checking(projects, max_per_projects, preferences, allocations, proposals, ranking_allocations, unassigned_students)
+
+    write_csv(allocations, proposals, ranking_allocations)