LPKT.py

import logging
import numpy as np
from load_data import DATA
from EduKTM import LPKT

def generate_q_matrix(path, n_skill, n_problem, gamma=0.0):
    with open(path, 'r', encoding='utf-8') as f:
        for line in f:
            problem2skill = eval(line)
    q_matrix = np.zeros((n_problem + 1, n_skill + 1)) + gamma
    for p in problem2skill.keys():
        q_matrix[p][problem2skill[p]] = 1
    return q_matrix

batch_size = 32
n_at = 1326
n_it = 2839
n_question = 102
n_exercise = 3162
seqlen = 500
d_k = 128
d_a = 50
d_e = 128
q_gamma = 0.03
dropout = 0.2

q_matrix = generate_q_matrix(
    '../../data/anonymized_full_release_competition_dataset/problem2skill',
    n_question, n_exercise,
    q_gamma
)
dat = DATA(seqlen=seqlen, separate_char=',')

logging.getLogger().setLevel(logging.INFO)

# k-fold cross validation
k, train_auc_sum, valid_auc_sum = 5, .0, .0
for i in range(k):
    lpkt = LPKT(n_at, n_it, n_exercise, n_question, d_a, d_e, d_k, q_matrix, batch_size, dropout)
    train_data = dat.load_data('../../data/anonymized_full_release_competition_dataset/train' + str(i) + '.txt')
    valid_data = dat.load_data('../../data/anonymized_full_release_competition_dataset/valid' + str(i) + '.txt')
    best_train_auc, best_valid_auc = lpkt.train(train_data, valid_data, epoch=30, lr=0.003, lr_decay_step=10)
    print('fold %d, train auc %f, valid auc %f' % (i, best_train_auc, best_valid_auc))
    train_auc_sum += best_train_auc
    valid_auc_sum += best_valid_auc
print('%d-fold validation: avg of best train auc %f, avg of best valid auc %f' % (k, train_auc_sum / k, valid_auc_sum / k))

# train and pred
train_data = dat.load_data('../../data/anonymized_full_release_competition_dataset/train0.txt')
valid_data = dat.load_data('../../data/anonymized_full_release_competition_dataset/valid0.txt')
test_data = dat.load_data('../../data/anonymized_full_release_competition_dataset/test.txt')
lpkt = LPKT(n_at, n_it, n_exercise, n_question, d_a, d_e, d_k, q_matrix, batch_size, dropout)
lpkt.train(train_data, valid_data, epoch=30, lr=0.003, lr_decay_step=10)
lpkt.save("lpkt.params")

lpkt.load("lpkt.params")
_, auc, accuracy = lpkt.eval(test_data)
print("auc: %.6f, accuracy: %.6f" % (auc, accuracy))