Closed wenhe-jia closed 6 years ago
When I tried to visualize the mask rcnn network, I used the script below:
import mxnet as mx sym = mx.sym.load('rcnn1-symbol.json') mx.viz.plot_network(symbol=sym, node_attrs={"shape": 'oval', "fixedsize": 'false'}).view()
But a error is reported:
/home/jiawenhe/workspace/mxnet/dmlc-core/include/dmlc/logging.h:308: [20:47:29] src/operator/custom/custom.cc:78: Check failed: Registry::Get()->Find(params.op_type) != nullptr Cannot find custom operator MaskOutput Stack trace returned 10 entries: [bt] (0) /usr/local/lib/python2.7/dist-packages/mxnet-0.12.0-py2.7.egg/mxnet/libmxnet.so(_ZN4dmlc15LogMessageFatalD1Ev+0x3c) [0x7f4e99987f2c] [bt] (1) /usr/local/lib/python2.7/dist-packages/mxnet-0.12.0-py2.7.egg/mxnet/libmxnet.so(_ZN5mxnet2op6custom10AttrParserEPN4nnvm9NodeAttrsE+0xaac) [0x7f4e99a5658c] [bt] (2) /usr/local/lib/python2.7/dist-packages/mxnet-0.12.0-py2.7.egg/mxnet/libmxnet.so(_ZN5mxnet17UpgradeJSON_ParseEN4nnvm5GraphE+0x330) [0x7f4e9b9e1cd0] [bt] (3) /usr/local/lib/python2.7/dist-packages/mxnet-0.12.0-py2.7.egg/mxnet/libmxnet.so(_ZNSt17_Function_handlerIFN4nnvm5GraphES1_EPS2_E9_M_invokeERKSt9_Any_dataS1_+0x11f) [0x7f4e9b9e538f] [bt] (4) /usr/local/lib/python2.7/dist-packages/mxnet-0.12.0-py2.7.egg/mxnet/libmxnet.so(_ZN5mxnet18LoadLegacyJSONPassEN4nnvm5GraphE+0x333) [0x7f4e9b9df9c3] [bt] (5) /usr/local/lib/python2.7/dist-packages/mxnet-0.12.0-py2.7.egg/mxnet/libmxnet.so(_ZNSt17_Function_handlerIFN4nnvm5GraphES1_EPS2_E9_M_invokeERKSt9_Any_dataS1_+0x11f) [0x7f4e9b9e538f] [bt] (6) /usr/local/lib/python2.7/dist-packages/mxnet-0.12.0-py2.7.egg/mxnet/libmxnet.so(_ZN4nnvm11ApplyPassesENS_5GraphERKSt6vectorISsSaISsEE+0x518) [0x7f4e9cf24448] [bt] (7) /usr/local/lib/python2.7/dist-packages/mxnet-0.12.0-py2.7.egg/mxnet/libmxnet.so(_ZN4nnvm9ApplyPassENS_5GraphERKSs+0x8e) [0x7f4e9b7c46ae] [bt] (8) /usr/local/lib/python2.7/dist-packages/mxnet-0.12.0-py2.7.egg/mxnet/libmxnet.so(MXSymbolCreateFromFile+0x469) [0x7f4e9b7be799] [bt] (9) /usr/lib/x86_64-linux-gnu/libffi.so.6(ffi_call_unix64+0x4c) [0x7f4ec7f07adc] Traceback (most recent call last): File "../visual_net.py", line 4, in <module> sym = mx.sym.load('/home/jiawenhe/workspace/mx-maskrcnn/rcnn1-symbol.json') File "/usr/local/lib/python2.7/dist-packages/mxnet-0.12.0-py2.7.egg/mxnet/symbol/symbol.py", line 2494, in load check_call(_LIB.MXSymbolCreateFromFile(c_str(fname), ctypes.byref(handle))) File "/usr/local/lib/python2.7/dist-packages/mxnet-0.12.0-py2.7.egg/mxnet/base.py", line 146, in check_call raise MXNetError(py_str(_LIB.MXGetLastError())) mxnet.base.MXNetError: [20:47:29] src/operator/custom/custom.cc:78: Check failed: Registry::Get()->Find(params.op_type) != nullptr Cannot find custom operator MaskOutput
This problem also occured with annother custom operator such as RPN, both RPN and MaskOutput are layers implemented by Python. How to fix this?
When I tried to visualize the mask rcnn network, I used the script below:
But a error is reported:
This problem also occured with annother custom operator such as RPN, both RPN and MaskOutput are layers implemented by Python. How to fix this?