def test_predict(self):
data = np.array([[0, 0, 5], [3, 0, 5], [3, 1, 2]])
ratings = csr_matrix(data)
train = ds.array(x=ratings, block_size=(1, 1))
als = ALS(tol=0.01, random_state=666, n_f=5, verbose=False)
als.fit(train)
predictions = als.predict_user(user_id=0)
# Check that the ratings for user 0 are similar to user 1 because they
# share preferences (third movie), thus it is expected that user 0
# will rate movie 1 similarly to user 1.
self.assertTrue(2.75 < predictions[0] < 3.25 and
predictions[1] < 1 and
predictions[2] > 4.5)
def main():
n_blocks = 384
data = "/gpfs/projects/bsc19/COMPSs_DATASETS/dislib/recommendation" \
"/netflix/netflix_data_libsvm.txt"
n_factors = 100
n_features = 480189
block_size = (int(ceil(17770 / n_blocks)),
int(ceil(n_features / n_blocks)))
x, y = ds.load_svmlight_file(data, block_size=block_size,
n_features=n_features, store_sparse=True)
als = ALS(tol=0.0001, random_state=676, n_f=n_factors, max_iter=10,
verbose=False)
performance.measure("ALS", "Netflix", als, x)
def test_fit(self):
train, test = load_movielens()
als = ALS(tol=0.01, random_state=666, n_f=100, verbose=False,
check_convergence=True)
als.fit(train, test)
self.assertTrue(als.converged)
als.fit(train)
self.assertTrue(als.converged)
def test_init_params(self):
# Test all parameters
seed = 666
n_f = 100
lambda_ = 0.001
convergence_threshold = 0.1
max_iter = 10
verbose = True
arity = 12
als = ALS(random_state=seed, n_f=n_f, lambda_=lambda_,
tol=convergence_threshold,
max_iter=max_iter, verbose=verbose, arity=arity)
self.assertEqual(als.random_state, seed)
self.assertEqual(als.n_f, n_f)
self.assertEqual(als.lambda_, lambda_)
self.assertEqual(als.tol, convergence_threshold)
self.assertEqual(als.max_iter, max_iter)
self.assertEqual(als.verbose, verbose)
self.assertEqual(als.arity, arity)
def main():
n_blocks = 384
data = "/fefs/scratch/bsc19/bsc19029/PERFORMANCE/datasets/" \
"netflix_data_libsvm.txt"
n_factors = 100
n_features = 480189
block_size = (int(ceil(17770 / n_blocks)),
int(ceil(n_features / n_blocks)))
x, y = ds.load_svmlight_file(data,
block_size=block_size,
n_features=n_features,
store_sparse=True)
als = ALS(tol=0.0001,
random_state=676,
n_f=n_factors,
max_iter=10,
verbose=False)
performance.measure("ALS", "Netflix", als.fit, x)
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument("--num_subsets", type=int, default=48)
parser.add_argument("--num_factors", type=int, default=100)
parser.add_argument("--data_path", type=str, default='./tests/files/')
args = parser.parse_args()
num_subsets = args.num_subsets
data_path = args.data_path
n_f = args.num_factors
train, test = load_movielens(data_path=data_path)
exec_start = time()
als = ALS(tol=0.0001, n_f=n_f, max_iter=2, verbose=True)
# Fit using training data to check convergence
# als.fit(train)
# Fit using test data to check convergence
als.fit(train, test)
exec_end = time()
print("Ratings for user 0:\n%s" % als.predict_user(0))
print("Execution time: %.2f" % (exec_end - exec_start))