Loading nn_ds_inmem.py +70 −69 Original line number Diff line number Diff line Loading @@ -194,6 +194,7 @@ def batchLoss(out_batch, # [batch_size,(1..2)] tf_result conf_pwr = 2.0, gt_conf_offset = 0.08, gt_conf_pwr = 1.0): with tf.name_scope("BatchLoss"): """ Here confidence should be after relU. Disparity - may be also if absolute, but no activation if output is residual disparity """ Loading Loading @@ -239,8 +240,8 @@ def batchLoss(out_batch, # [batch_size,(1..2)] tf_result w_all = tf.multiply (w, nconf, name = "w_all") else: w_all = w cost2 = 0.0 cost3 = 0.0 # cost2 = 0.0 # cost3 = 0.0 # normalize weights w_sum = tf.reduce_sum(w_all, name = "w_sum") iw_sum = tf.divide(tf_1f, w_sum, name = "iw_sum") Loading Loading
nn_ds_inmem.py +70 −69 Original line number Diff line number Diff line Loading @@ -194,6 +194,7 @@ def batchLoss(out_batch, # [batch_size,(1..2)] tf_result conf_pwr = 2.0, gt_conf_offset = 0.08, gt_conf_pwr = 1.0): with tf.name_scope("BatchLoss"): """ Here confidence should be after relU. Disparity - may be also if absolute, but no activation if output is residual disparity """ Loading Loading @@ -239,8 +240,8 @@ def batchLoss(out_batch, # [batch_size,(1..2)] tf_result w_all = tf.multiply (w, nconf, name = "w_all") else: w_all = w cost2 = 0.0 cost3 = 0.0 # cost2 = 0.0 # cost3 = 0.0 # normalize weights w_sum = tf.reduce_sum(w_all, name = "w_sum") iw_sum = tf.divide(tf_1f, w_sum, name = "iw_sum") Loading