今天在用efficientDet跑模型的时候，出现了下面的警告：
WARNING:tensorflow:Entity <bound method RegressBoxes.call of <layers.RegressBoxes object at 0x7f41c3f35a90>> could not be transformed and will be executed as-is. Please report this to the AutgoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: converting <bound method RegressBoxes.call of <layers.RegressBoxes object at 0x7f41c3f35a90>>: AssertionError: Bad argument number for Name: 3, expecting 4
ERROR:tensorflow:Error converting <bound method ClipBoxes.call of <layers.ClipBoxes object at 0x7f41c3fad9e8>>
Traceback (most recent call last):
File "/home/eric/anaconda3/envs/pointnet/lib/python3.6/site-packages/tensorflow/python/autograph/impl/api.py", line 524, in to_graph
return conversion.convert(entity, program_ctx)
File "/home/eric/anaconda3/envs/pointnet/lib/python3.6/site-packages/tensorflow/python/autograph/impl/conversion.py", line 306, in convert
entity, program_ctx, free_nonglobal_var_names)
File "/home/eric/anaconda3/envs/pointnet/lib/python3.6/site-packages/tensorflow/python/autograph/impl/conversion.py", line 229, in _convert_with_cache
entity, program_ctx)
File "/home/eric/anaconda3/envs/pointnet/lib/python3.6/site-packages/tensorflow/python/autograph/impl/conversion.py", line 433, in convert_entity_to_ast
nodes, name, entity_info = convert_func_to_ast(o, program_ctx)
File "/home/eric/anaconda3/envs/pointnet/lib/python3.6/site-packages/tensorflow/python/autograph/impl/conversion.py", line 624, in convert_func_to_ast
node = node_to_graph(node, context)
File "/home/eric/anaconda3/envs/pointnet/lib/python3.6/site-packages/tensorflow/python/autograph/impl/conversion.py", line 657, in node_to_graph
node = converter.standard_analysis(node, context, is_initial=True)
File "/home/eric/anaconda3/envs/pointnet/lib/python3.6/site-packages/tensorflow/python/autograph/core/converter.py", line 354, in standard_analysis
node = qual_names.resolve(node)
File "/home/eric/anaconda3/envs/pointnet/lib/python3.6/site-packages/tensorflow/python/autograph/pyct/qual_names.py", line 254, in resolve
return QnResolver().visit(node)
File "/home/eric/anaconda3/envs/pointnet/lib/python3.6/ast.py", line 253, in visit
return visitor(node)
File "/home/eric/anaconda3/envs/pointnet/lib/python3.6/ast.py", line 308, in generic_visit
value = self.visit(value)
File "/home/eric/anaconda3/envs/pointnet/lib/python3.6/ast.py", line 253, in visit
return visitor(node)
File "/home/eric/anaconda3/envs/pointnet/lib/python3.6/ast.py", line 317, in generic_visit
new_node = self.visit(old_value)
File "/home/eric/anaconda3/envs/pointnet/lib/python3.6/ast.py", line 253, in visit
return visitor(node)
File "/home/eric/anaconda3/envs/pointnet/lib/python3.6/site-packages/tensorflow/python/autograph/pyct/qual_names.py", line 236, in visit_Subscript
if isinstance(s.value, gast.Num):
AttributeError: module 'gast' has no attribute 'Num'
解决方法
pip install 'gast==0.2.2'参考文献
[1].Module 'gast' has no attribute 'Num' . https://github.com/tensorflow/tensorflow/issues/32319
