MO should support LRN k param with caffe model, rather than fixed to 1 (#716)

Co-authored-by: yipengqu <yipeng.qu@intel.com>

Author: wistal

model-optimizer/extensions/front/caffe/lrn_ext.py

@@ -29,7 +29,7 @@ def extract(cls, node):
         AttributedLRN.update_node_stat(node, {
             'alpha': param.alpha,
             'beta': param.beta,
-            'bias': 1,
+            'bias': param.k,
             'local_size': param.local_size,
             'region': region,
         })