refactor for op::Result

[mbrookhart]

Man, op::result is a lot of memcpy.

[ashokei]

seeing below issue

python example/image-classification/train_mnist.py
INFO:root:start with arguments Namespace(add_stn=False, batch_size=64, disp_batches=100, dtype='float32', gc_threshold=0.5, gc_type='none', gpus=None, is_nnp=False, kv_store='device', load_epoch=None, lr=0.05, lr_factor=0.1, lr_step_epoch
s='10', model_prefix=None, mom=0.9, monitor=0, network='mlp', num_classes=10, num_epochs=20, num_examples=60000, num_layers=None, optimizer='sgd', test_io=0, top_k=0, wd=0.0001)
example/image-classification/train_mnist.py:38: DeprecationWarning: The binary mode of fromstring is deprecated, as it behaves surprisingly on unicode inputs. Use frombuffer instead
  label = np.fromstring(flbl.read(), dtype=np.int8)  
example/image-classification/train_mnist.py:41: DeprecationWarning: The binary mode of fromstring is deprecated, as it behaves surprisingly on unicode inputs. Use frombuffer instead
  image = np.fromstring(fimg.read(), dtype=np.uint8).reshape(len(label), rows, cols)
terminate called after throwing an instance of 'std::out_of_range'
  what():  deque::_M_range_check: __n (which is 1)>= this->size() (which is 1)
fish: “python example/image-classifica…” terminated by signal SIGABRT (Abort)

[ashokei]

could be ngraph-cpp that i used, may have to retest with different version.

[ashokei]

LGTM,. the crash is resolved in my testing.