search

SteveMacenski / jetson_nano_detection_and_tracking

Jetson Nano ML install scripts, automated optimization of robotics detection models, and filter-based tracking of detections
GNU Lesser General Public License v2.1
226 stars 66 forks source link

Error in `./install_scripts/install_tf_models.sh` #12

Closed zhangtemplar closed 5 years ago

zhangtemplar commented 5 years ago 
Processing dependencies for tf-trt-models==0.0
Finished processing dependencies for tf-trt-models==0.0
Testing model install...
RuntimeError: module compiled against API version 0xc but this version of numpy is 0xb
ImportError: numpy.core.multiarray failed to import
ImportError: numpy.core.umath failed to import
ImportError: numpy.core.umath failed to import
Traceback (most recent call last):
  File "object_detection/builders/model_builder_test.py", line 18, in <module>
    import tensorflow as tf
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/__init__.py", line 24, in <module>
    from tensorflow.python import pywrap_tensorflow  # pylint: disable=unused-import
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/__init__.py", line 52, in <module>
    from tensorflow.core.framework.graph_pb2 import *
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/core/framework/graph_pb2.py", line 15, in <module>
    from tensorflow.core.framework import node_def_pb2 as tensorflow_dot_core_dot_framework_dot_node__def__pb2
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/core/framework/node_def_pb2.py", line 15, in <module>
    from tensorflow.core.framework import attr_value_pb2 as tensorflow_dot_core_dot_framework_dot_attr__value__pb2
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/core/framework/attr_value_pb2.py", line 15, in <module>
    from tensorflow.core.framework import tensor_pb2 as tensorflow_dot_core_dot_framework_dot_tensor__pb2
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/core/framework/tensor_pb2.py", line 15, in <module>
    from tensorflow.core.framework import resource_handle_pb2 as tensorflow_dot_core_dot_framework_dot_resource__handle__pb2
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/core/framework/resource_handle_pb2.py", line 22, in <module>
    serialized_pb=_b('\n/tensorflow/core/framework/resource_handle.proto\x12\ntensorflow\"r\n\x13ResourceHandleProto\x12\x0e\n\x06\x64\x65vice\x18\x01 \x01(\t\x12\x11\n\tcontainer\x18\x02 \x01(\t\x12\x0c\n\x04name\x18\x03 \x01(\t\x12\x11\n\thash_code\x18\x04 \x01(\x04\x12\x17\n\x0fmaybe_type_name\x18\x05 \x01(\tBn\n\x18org.tensorflow.frameworkB\x0eResourceHandleP\x01Z=github.com/tensorflow/tensorflow/tensorflow/go/core/framework\xf8\x01\x01\x62\x06proto3')
TypeError: __new__() got an unexpected keyword argument 'serialized_options'
SteveMacenski commented 5 years ago

It looks like you’re using the wrong version of numpy

omartin2010 commented 4 years ago

I've had the same error on my jetson tx2 today... which I tried to solve by changing the version of numpy. It ended up working by downgrading protobuf to 3.6.1... And it worked (tensorflow was importable) on the first attempt when I created a virtual env, even with a more recent version of protobuf, so not sure what was wrong... but clearly something related to previous work I did (like following the nvidia 2 day tutorial which unfortunately isn't done in a container where things are isolated). I'm trying to port this to a container... for that same reason :) great work btw.