fail to open "model/pnet-symbol.json"
tianhuangli opened this issue · comments
@Seanlinx hi，when i finished train P-net (executing train_P_net.py),The results are shown below. however ,there is no problem when i train the R-net and O-net...I carefully examined the parameters of the program，but still not find any problem. Do you know why this happens? thank you for reply!
result:
INFO:root:Epoch[0] Train-Accuracy=0.998550
INFO:root:Epoch[0] Train-LogLoss=0.008678
INFO:root:Epoch[0] Train-BBOX_MSE=0.031688
INFO:root:Epoch[0] Time cost=583.256
[10:39:30] include/dmlc/logging.h:308: [10:39:30] src/io/local_filesys.cc:166: Check failed: allow_null LocalFileSystem: fail to open "model/pnet-symbol.json"
Stack trace returned 10 entries:
[bt] (0) /home/cc/tianhuangshijie/mxnet/local/lib/python2.7/site-packages/mxnet-0.12.1-py2.7.egg/mxnet/libmxnet.so(_ZN4dmlc2io15LocalFileSystem4OpenERKNS0_3URIEPKcb+0x459) [0x7fd587df1d39]
[bt] (1) /home/cc/tianhuangshijie/mxnet/local/lib/python2.7/site-packages/mxnet-0.12.1-py2.7.egg/mxnet/libmxnet.so(_ZN4dmlc6Stream6CreateEPKcS2_b+0x3a) [0x7fd587de85aa]
[bt] (2) /home/cc/tianhuangshijie/mxnet/local/lib/python2.7/site-packages/mxnet-0.12.1-py2.7.egg/mxnet/libmxnet.so(MXSymbolSaveToFile+0x25) [0x7fd5878418b5]
[bt] (3) /usr/lib/x86_64-linux-gnu/libffi.so.6(ffi_call_unix64+0x4c) [0x7fd590988c7c]
[bt] (4) /usr/lib/x86_64-linux-gnu/libffi.so.6(ffi_call+0x1fc) [0x7fd5909885ac]
[bt] (5) /home/cc/tianhuangshijie/mxnet/lib/python2.7/lib-dynload/_ctypes.x86_64-linux-gnu.so(_ctypes_callproc+0x48e) [0x7fd590b9f5fe]
[bt] (6) /home/cc/tianhuangshijie/mxnet/lib/python2.7/lib-dynload/_ctypes.x86_64-linux-gnu.so(+0x15f9e) [0x7fd590ba0f9e]
[bt] (7) python(PyEval_EvalFrameEx+0x965) [0x499be5]
[bt] (8) python(PyEval_EvalCodeEx+0x2ac) [0x4a090c]
[bt] (9) python(PyEval_EvalFrameEx+0x18c5) [0x49ab45]
Traceback (most recent call last):
File "train_P_net.py", line 54, in
args.begin_epoch, args.end_epoch, args.frequent, args.lr, args.resume)
File "train_P_net.py", line 16, in train_P_net
12, frequent, not resume, lr)
File "/home/lith/mtcnn/train.py", line 84, in train_net
arg_params=args, aux_params=auxs, begin_epoch=begin_epoch, num_epoch=end_epoch)
File "/home/cc/tianhuangshijie/mxnet/local/lib/python2.7/site-packages/mxnet-0.12.1-py2.7.egg/mxnet/module/base_module.py", line 521, in fit
callback(epoch, self.symbol, arg_params, aux_params)
File "/home/cc/tianhuangshijie/mxnet/local/lib/python2.7/site-packages/mxnet-0.12.1-py2.7.egg/mxnet/callback.py", line 89, in _callback
save_checkpoint(prefix, iter_no + 1, sym, arg, aux)
File "/home/cc/tianhuangshijie/mxnet/local/lib/python2.7/site-packages/mxnet-0.12.1-py2.7.egg/mxnet/model.py", line 387, in save_checkpoint
symbol.save('%s-symbol.json' % prefix)
File "/home/cc/tianhuangshijie/mxnet/local/lib/python2.7/site-packages/mxnet-0.12.1-py2.7.egg/mxnet/symbol/symbol.py", line 1185, in save
check_call(_LIB.MXSymbolSaveToFile(self.handle, c_str(fname)))
File "/home/cc/tianhuangshijie/mxnet/local/lib/python2.7/site-packages/mxnet-0.12.1-py2.7.egg/mxnet/base.py", line 146, in check_call
raise MXNetError(py_str(_LIB.MXGetLastError()))
mxnet.base.MXNetError: [10:39:30] src/io/local_filesys.cc:166: Check failed: allow_null LocalFileSystem: fail to open "model/pnet-symbol.json"
Stack trace returned 10 entries:
[bt] (0) /home/cc/tianhuangshijie/mxnet/local/lib/python2.7/site-packages/mxnet-0.12.1-py2.7.egg/mxnet/libmxnet.so(_ZN4dmlc2io15LocalFileSystem4OpenERKNS0_3URIEPKcb+0x459) [0x7fd587df1d39]
[bt] (1) /home/cc/tianhuangshijie/mxnet/local/lib/python2.7/site-packages/mxnet-0.12.1-py2.7.egg/mxnet/libmxnet.so(_ZN4dmlc6Stream6CreateEPKcS2_b+0x3a) [0x7fd587de85aa]
[bt] (2) /home/cc/tianhuangshijie/mxnet/local/lib/python2.7/site-packages/mxnet-0.12.1-py2.7.egg/mxnet/libmxnet.so(MXSymbolSaveToFile+0x25) [0x7fd5878418b5]
[bt] (3) /usr/lib/x86_64-linux-gnu/libffi.so.6(ffi_call_unix64+0x4c) [0x7fd590988c7c]
[bt] (4) /usr/lib/x86_64-linux-gnu/libffi.so.6(ffi_call+0x1fc) [0x7fd5909885ac]
[bt] (5) /home/cc/tianhuangshijie/mxnet/lib/python2.7/lib-dynload/_ctypes.x86_64-linux-gnu.so(_ctypes_callproc+0x48e) [0x7fd590b9f5fe]
[bt] (6) /home/cc/tianhuangshijie/mxnet/lib/python2.7/lib-dynload/_ctypes.x86_64-linux-gnu.so(+0x15f9e) [0x7fd590ba0f9e]
[bt] (7) python(PyEval_EvalFrameEx+0x965) [0x499be5]
[bt] (8) python(PyEval_EvalCodeEx+0x2ac) [0x4a090c]
[bt] (9) python(PyEval_EvalFrameEx+0x18c5) [0x49ab45]
When I was training the third time, there was no error...==
I also met this problem, can you tell me how to fix it? thanks!
I have tried several times, but it always exists.
mkdir model