btp mtp allocation divyam

import pandas as pd

# 1. Load the input file
file_path = '/Users/divyamgoel/Desktop/ERP_MAIN/ERP/include/python_codes/old/BTP-ORDER.csv'
#sample - https://www.x.com/scl/fi/0vqoeszcdfpw3623b9z9f/BTP-ORDER.csv?rlkey=afg5et38epsucg9nctoajv4uq&dl=0
df = pd.read_csv(file_path)
OUTPUT_FILE = str(file_path.split('.')[0])+"_allocation.csv"
# Added the new output file path
PREF_COUNT_FILE = str(file_path.split('.')[0])+"_pref_count.csv"

print(OUTPUT_FILE)
print(PREF_COUNT_FILE)

# 2. Sort students by CGPA in decreasing order
df_sorted = df.sort_values(by='CGPA', ascending=False, kind='stable').reset_index(drop=True)

# 3. Identify professor preference columns
prof_cols = [col for col in df.columns if 'Supervisor Preference [' in col]
num_profs = len(prof_cols)

allocation_results = []
available_profs = set()

# 4. Allocation Logic (Unchanged)
for i, row in df_sorted.iterrows():
    if i % num_profs == 0:
        available_profs = set(prof_cols)

    student_prefs = row[prof_cols].sort_values().index.tolist()

    allotted_prof_col = None
    rank_of_allotment = None

    for prof_col in student_prefs:
        if prof_col in available_profs:
            allotted_prof_col = prof_col
            rank_of_allotment = row[prof_col]
            available_profs.remove(prof_col)
            break

    prof_name = allotted_prof_col.replace('Supervisor Preference [', '').replace(']', '')

    allocation_results.append({
        'Roll Number': row['Roll Number'],
        'Full Name': row['Full Name'],
        'CGPA': row['CGPA'],
        'Allotted Professor': prof_name,
        'Preference Rank': int(rank_of_allotment)
    })

# 5. Save the final allocation output to CSV
output_df = pd.DataFrame(allocation_results)
output_df.to_csv(OUTPUT_FILE, index=False)

# --- NEW SECTION: Preference Count Logic ---

# Calculate how many times each rank (1, 2, 3...) appears for each professor
# We transpose it so professors are rows and ranks (1, 2, 3...) are columns
pref_counts = df[prof_cols].apply(pd.Series.value_counts).fillna(0).astype(int).T

# Clean the index (Professor names) to match your previous formatting
pref_counts.index = [name.replace('Supervisor Preference [', '').replace(']', '') for name in pref_counts.index]
pref_counts.index.name = 'Professor'

# Save the preference counts to CSV
pref_counts.to_csv(PREF_COUNT_FILE)

# --- END NEW SECTION ---

print(f"Allocation complete. Results saved to {OUTPUT_FILE}")
print(f"Preference counts saved to {PREF_COUNT_FILE}")