n_items = len(unique_sid)
def load_data(csv_file, shape=(n_users, n_items)):
tp = pd.read_csv(csv_file)
rows, cols, vals = np.array(tp['uid']), np.array(
tp['sid']), np.array(tp['rating'])
data = sparse.csr_matrix((vals, (rows, cols)),
dtype=np.float32,
shape=shape)
return data
train_data = load_data(os.path.join(DATA_DIR, 'train.csv'))
vad_data = load_data(os.path.join(DATA_DIR, 'validation.csv'))
test_data = load_data(os.path.join(DATA_DIR, 'test_full.csv'))
train_data_imp = exp_to_imp(train_data, 0.5)
vad_data_imp = exp_to_imp(vad_data, 0.5)
dims = np.array([1, 2, 5, 10, 20, 50, 100])
train_data_coo = train_data_imp.tocoo()
row_tr, col_tr = train_data_coo.row, train_data_coo.col
vad_data_coo = vad_data_imp.tocoo()
row_vd, col_vd = vad_data_coo.row, vad_data_coo.col
for i, dim in enumerate(dims):
print("dim", dim)
pf = pmf.PoissonMF(n_components=dim,
random_state=98765,
verbose=True,
for dim in dims:
U = np.loadtxt(CAUSEFIT_DIR + '/cause_pmf_k' + str(dim) + '_U.csv')
V = np.loadtxt(CAUSEFIT_DIR + '/cause_pmf_k' + str(dim) + '_V.csv')
U = (np.atleast_2d(U.T).T)
V = (np.atleast_2d(V.T).T)
reconstr_cau = U.dot(V.T)
for i, K in enumerate(outdims):
print("K0", dim, "K", K)
D = train_data.shape[0]
N = train_data.shape[1]
weights = train_data * alpha
cau = exp_to_imp(train_data)
tf.reset_default_graph()
sess = tf.InteractiveSession()
idx_ph = tf.placeholder(tf.int32, M)
cau_ph = tf.placeholder(tf.float32, [M, N])
sd_ph = tf.placeholder(tf.float32, [M, N])
reconstr_cau_ph = tf.placeholder(tf.float32, [M, N])
U = Gamma(0.3 * tf.ones([M, K]), 0.3 * tf.ones([M, K]))
V = Gamma(0.3 * tf.ones([N, K]), 0.3 * tf.ones([N, K]))
gamma = Gamma(tf.ones([M, 1]), tf.ones([M, 1]))
beta0 = Gamma(0.3 * tf.ones([1, 1]), 0.3 * tf.ones([1, 1]))
x = Poisson(tf.add(tf.matmul(U, V, transpose_b=True),\
data_te = sparse.csr_matrix(
(ratings_te, (rows_te, cols_te)),
dtype='float64',
shape=(end_idx - start_idx + 1, n_items))
return data_tr, data_te
train_data = load_train_data(os.path.join(DATA_DIR,
'train.csv')).tocsr()
vad_data_tr, vad_data_te = load_tr_te_data(
os.path.join(DATA_DIR, 'validation_tr.csv'),
os.path.join(DATA_DIR, 'validation_te.csv'))
test_data_tr, test_data_te = load_tr_te_data(
os.path.join(DATA_DIR, 'test_tr.csv'),
os.path.join(DATA_DIR, 'test_te.csv'))
train_data = exp_to_imp(train_data, 0.5)
vad_data_tr, vad_data_te = exp_to_imp(vad_data_tr, 0.5), exp_to_imp(
vad_data_tr, 0.5)
test_data_tr, test_data_te = exp_to_imp(test_data_tr, 0.5), exp_to_imp(
test_data_tr, 0.5)
dims = np.array([1, 2, 5, 10, 20, 50, 100])
train_data_coo = train_data.tocoo()
vad_data_tr_coo = vad_data_tr.tocoo()
test_data_tr_coo = test_data_tr.tocoo()
for i, dim in enumerate(dims):
print("dim", dim)
pf = pmf.PoissonMF(n_components=dim,
random_state=98765,
for dim in dims:
U = np.loadtxt(CAUSEFIT_DIR + '/cause_pmf_k' + str(dim) +
'_trainU.csv')
V = np.loadtxt(CAUSEFIT_DIR + '/cause_pmf_k' + str(dim) + '_V.csv')
U = (np.atleast_2d(U.T).T)
V = (np.atleast_2d(V.T).T)
reconstr_cau_train = U.dot(V.T)
for i, K in enumerate(outdims):
print("K0", dim, "K", K)
D = train_data.shape[0]
N = train_data.shape[1]
weights = train_data * alpha
cau = exp_to_imp(train_data)
tf.reset_default_graph()
sess = tf.InteractiveSession()
idx_ph = tf.placeholder(tf.int32, M)
cau_ph = tf.placeholder(tf.float32, [M, N])
sd_ph = tf.placeholder(tf.float32, [M, N])
U = Normal(loc=tf.zeros([M, K]), scale=pri_U * tf.ones([M, K]))
V = Normal(loc=tf.zeros([N, K]), scale=pri_V * tf.ones([N, K]))
x = Normal(loc=tf.multiply(cau_ph, tf.matmul(U,
V,
transpose_b=True)),
scale=tf.ones([M, N]))
shape=(end_idx - start_idx + 1, n_items))
return data_tr, data_te
train_data = load_train_data(os.path.join(DATA_DIR, 'train.csv')).tocsr()
vad_data_tr, vad_data_te = load_tr_te_data(
os.path.join(DATA_DIR, 'validation_tr.csv'),
os.path.join(DATA_DIR, 'validation_te.csv'))
test_data_tr, test_data_te = load_tr_te_data(
os.path.join(DATA_DIR, 'test_tr.csv'),
os.path.join(DATA_DIR, 'test_te.csv'))
alpha = args.alpha
print("alpha", alpha)
cau = exp_to_imp(train_data, 0.5)
if binary > 0:
train_data = binarize_rating(train_data)
vad_data_tr, vad_data_te = binarize_rating(
vad_data_tr), binarize_rating(vad_data_te)
test_data_tr, test_data_te = binarize_rating(
test_data_tr), binarize_rating(test_data_te)
train_data_coo = train_data.tocoo()
vad_data_tr_coo = vad_data_tr.tocoo()
test_data_tr_coo = test_data_tr.tocoo()
model_name = 'sg_pmf_obs'
print("model", model_name)