#data_type = 'random'
#full_data,Z_gen = generate_data(feature_count,data_count + held_out,T,sig,data_type)
#Y = full_data[:data_count,:]
#held_out = full_data[data_count:,:]
iterate = 5000
small_x = 3
small_y = 3
big_x = 3
big_y = 3
data_dim = (small_x,small_y,big_x,big_y)
feature_count = big_x*big_y
T = small_x*small_y*big_x*big_y
data_type = 'corr'
flag = 'nine'
#full_data,Z_gen = generate_data(feature_count,data_count + held_out,T,sig,data_type)
full_data,Z_gen = construct_data(small_x,small_y,big_x,big_y,data_count + held_out,sig,data_type,corr_value=2)
Y = full_data[:data_count,:]
held_out = full_data[data_count:,:]
K = 1 #start with K features
ext = 1 #draw one new feature per iteration
# profile.run('ugibbs_sample(log_res,hold,Y,ext,sig,sig_w,iterate,K,valid)')
runs = 1
ll_data = np.zeros((runs,iterate))
ll_time = np.zeros((runs,iterate))
feature = np.zeros((runs,iterate))
lapse_data = np.zeros((runs,iterate))
#pred_data = np.zeros((runs,iterate))
valid = 0
sig_test = sig
small_x = 3
small_y = 3
big_x = 2
big_y = 2
data_dim = (small_x, small_y, big_x, big_y)
feature_count = big_x * big_y
T = small_x * small_y * big_x * big_y
#data_type = 'random'
#data_type = 'special'
#data_type = 'random'
#data_type = 'debug'
data_type = 'debug-four'
flag = 'four'
#full_data,Z_gen = generate_data(feature_count,data_count + held_out,T,sig,data_type)
full_data, Z_gen = construct_data(data_dim, data_count + held_out, sig,
data_type)
Y = full_data[:data_count, :]
held_out = full_data[data_count:, :]
#we observe half the signal and recover the other half
observe = held_out[:, :T / 2]
trunc = 12 #truncate active features
if algorithm == 'IBP':
iterate = 200
alpha = 2.0
select = 12
Z, W, ll_set, pred_ll = ugibbs_sampler(Y, held_out, alpha, sig_test, sig_w,
iterate, select, trunc, data_dim)
Z_trunc, W_trunc = truncate(Z, W, select)
print_posterior(Z_trunc, W_trunc, data_dim)
recover_ll = recover_IBP(held_out, observe, Z, W, sig_test)
held_out = 50
sig = 0.1 #noise
sig_w = 0.2 #feature deviation
small_x = 3
small_y = 3
big_x = 3
big_y = 3
data_dim = (small_x, small_y, big_x, big_y)
feature_count = big_x * big_y
T = small_x * small_y * big_x * big_y
select = 12
data_type = 'corr'
#full_data,Z_gen = generate_data(feature_count,data_count + held_out,T,sig,data_type)
full_data, Z_gen = construct_data(data_dim,
data_count + held_out,
sig,
data_type,
corr_value=2)
Y = full_data[:data_count, :]
held_out = full_data[data_count:, :]
Z, W, ll_set, pred_ll = ugibbs_sampler(Y, held_out, alpha, sig, sig_w,
iterate, select)
approx = np.dot(Z, W)
for i in range(10):
print("sample: " + str(i))
print("features probability")
#print(prob_matrix[i,:])
print("features selected")
print(Z[i, :])