train: added cd_binary_gaussian

src/rbm/train.py

@@ -42,6 +42,25 @@ def cd_jens(entity: Entity, v_states: Mat):
 
 	return dw, dbv, dbh
 
+# %%%%%%%%% START POSITIVE PHASE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+#   data = batchdata(:,:,batch);
+#   poshidprobs = 1./(1 + exp(-data*vishid - repmat(hidbiases,numcases,1)));
+#   batchposhidprobs(:,:,batch)=poshidprobs;
+#   posprods    = data' * poshidprobs;
+#   poshidact   = sum(poshidprobs);
+#   posvisact = sum(data);
+#
+# %%%%%%%%% END OF POSITIVE PHASE  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+#   poshidstates = poshidprobs > rand(numcases,numhid);
+#
+# %%%%%%%%% START NEGATIVE PHASE  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+#   negdata = 1./(1 + exp(-poshidstates*vishid' - repmat(visbiases,numcases,1)));
+#   neghidprobs = 1./(1 + exp(-negdata*vishid - repmat(hidbiases,numcases,1)));
+#   negprods  = negdata'*neghidprobs;
+#   neghidact = sum(neghidprobs);
+#   negvisact = sum(negdata);
+#
+# %%%%%%%%% END OF NEGATIVE PHASE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
 def cd_binary_binary(entity: Entity, data_pos: Mat):
 	params = entity.training_params
 	# Positive phase
@@ -118,6 +137,56 @@ def cd_gaussian_gaussian(entity: Entity, data_pos: Mat):
 
 	return dw, dbv, dbh
 
+# %%%%%%%%%START POSITIVE PHASE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+# data = batchdata(:,:, batch);
+# poshidprobs = (data * vishid) + repmat(hidbiases, numcases, 1);
+# batchposhidprobs(:,:, batch)=poshidprobs;
+#
+# posprods = data' * poshidprobs;
+# poshidact = sum(poshidprobs);
+# posvisact = sum(data);
+#
+# %%%%%%%%%END OF POSITIVE PHASE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+# poshidstates = poshidprobs + randn(numcases, numhid);
+#
+# %%%%%%%%%START NEGATIVE PHASE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+# negdata = 1. / (1 + exp(-poshidstates * vishid' - repmat(visbiases,numcases,1)));
+# neghidprobs = (negdata * vishid) + repmat(hidbiases, numcases, 1);
+# negprods = negdata'*neghidprobs;
+# neghidact = sum(neghidprobs);
+# negvisact = sum(negdata);
+#
+# %%%%%%%%%END OF NEGATIVE PHASE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+def cd_binary_gaussian(entity: Entity, data_pos: Mat):
+	params = entity.training_params
+	# Positive phase
+	h_probs_pos = entity.h_given_v(data_pos)
+
+	# Update weights (positive phase)
+	dw = np.dot(np.transpose(data_pos), h_probs_pos)
+	dbh = np.sum(h_probs_pos, 0)
+	dbv = np.sum(data_pos, 0)
+
+	# Sample hidden states
+	if not params.do_rao_blackwell:
+		h_probs_pos += sample_gaussian(h_probs_pos)
+
+	# Negative phase
+	data_neg = prob(entity.v_given_h(h_probs_pos))
+	h_probs_neg = entity.h_given_v(data_neg)
+
+	# Gibbs	sampling
+	for _ in range(params.num_gibbs_samples-1):
+		data_neg = prob(entity.v_given_h(h_probs_neg))
+		h_probs_neg = entity.h_given_v(data_neg)
+
+	# Update weights (negative phase)
+	dw -= np.dot(np.transpose(data_neg), h_probs_neg)
+	dbh -= np.sum(h_probs_neg, 0)
+	dbv -= np.sum(data_neg, 0)
+
+	return dw, dbv, dbh
+
 def to_mini_batch(batch: Mat, mini_batch_size: int):
 	mini_batches = []
 	remain = batch.shape[0]
@@ -147,6 +216,8 @@ def train(entity: Entity, batch: Mat, status: Status):
 		cd_func = cd_binary_binary
 	if entity.type == Entity.Type.GG_RBM:
 		cd_func = cd_gaussian_gaussian
+	if entity.type == Entity.Type.BG_RBM:
+		cd_func = cd_binary_gaussian
 
 	for mini_batch in to_mini_batch(batch, mini_batch_size):
 		if not keep_running: