def test_rbm_pcd_gibbs(x_test,
Whv,
bh,
bv,
p_target=0.5,
n_gibbs_steps=5000,
thinning=10,
burnin=20):
rbm = BernoulliRBM(n_components=Whv.shape[0], learning_rate=0.0)
rbm.components_, rbm.intercept_hidden_, rbm.intercept_visible_ = Whv, bh, bv
evidence_mask = np.random.binomial(
1, p_target, x_test.shape) # 0: target node, 1: evidence node,
V = np.random.binomial(1, p_target, x_test.shape)
V = x_test * evidence_mask + V * (1 - evidence_mask)
prob1 = np.zeros_like(V)
count = 0
for it in range(n_gibbs_steps):
V = rbm.gibbs(V)
V = x_test * evidence_mask + V * (1 - evidence_mask)
if (it + 1) % thinning == 0 and it > burnin:
prob1 += V
count += 1
prob1 /= count
prob1_clipped = prob1.clip(1e-15, 1 - 1e-15)
target_mask = 1 - evidence_mask
logp = x_test * np.log(prob1_clipped) + (
1 - x_test) * np.log(1 - prob1_clipped)
logp *= target_mask
return -logp.sum() / target_mask.sum() / np.log(2)
def classify(self,X):
transformed = self.first_pipeline.transform(X)
transformed = np.concatenate((transformed,[[0]*10]*len(transformed)),axis=1)
# The inverse of rbm_3 to go from hidden layer to visible layer
rbm_aux = BernoulliRBM()
rbm_aux.intercept_hidden_ = self.rbm_3.intercept_visible_
rbm_aux.intercept_visible_ = self.rbm_3.intercept_hidden_
rbm_aux.components_ = np.transpose(self.rbm_3.components_)
results = rbm_aux.transform(self.rbm_3.transform(transformed))
results = results[:,-10:]
return np.argmax(results,axis=1)
def classify(self, X):
transformed = self.first_pipeline.transform(X)
transformed = np.concatenate(
(transformed, [[0] * 10] * len(transformed)), axis=1)
# The inverse of rbm_3 to go from hidden layer to visible layer
rbm_aux = BernoulliRBM()
rbm_aux.intercept_hidden_ = self.rbm_3.intercept_visible_
rbm_aux.intercept_visible_ = self.rbm_3.intercept_hidden_
rbm_aux.components_ = np.transpose(self.rbm_3.components_)
results = rbm_aux.transform(self.rbm_3.transform(transformed))
results = results[:, -10:]
return np.argmax(results, axis=1)
image_size = [28, 28]
train_imgs, train_lbls, test_imgs, test_lbls = read_mnist(dim=image_size,
n_train=6000,
n_test=10000)
n_iter = 10
v_h = BernoulliRBM(n_components=500,
learning_rate=0.01,
batch_size=20,
verbose=1,
n_iter=n_iter)
try:
v_h.components_, v_h.intercept_hidden_, v_h.intercept_visible_ = load_stuff(
"v_h_sklearn.npz")
except FileNotFoundError:
v_h.fit(train_imgs)
save_stuff(
"v_h_sklearn.npz",
[v_h.components_, v_h.intercept_hidden_, v_h.intercept_visible_])
v_h_p_hv = sigmoid(test_imgs @ v_h.components_.T + v_h.intercept_hidden_)
v_h_p_vh = sigmoid(
sample_binary(v_h_p_hv) @ v_h.components_ + v_h.intercept_visible_)
print("Layer1 MSE:", mean_squared_error(test_imgs, v_h_p_vh))
h_p = BernoulliRBM(n_components=500,
learning_rate=0.01,
batch_size=20,
def get_likelihood(data, W, vb, hb):
rbm = BernoulliRBM(n_components=W.shape[1])
rbm.components_ = W.T
rbm.intercept_hidden_ = hb
rbm.intercept_visible_ = vb
return rbm.score_samples(data).mean()