Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

allow_build_at_runtime=False broken for TrtGraphConverterV2 #216

Open
wmdrichards opened this issue Oct 11, 2020 · 1 comment
Open

allow_build_at_runtime=False broken for TrtGraphConverterV2 #216

wmdrichards opened this issue Oct 11, 2020 · 1 comment
Assignees

Comments

@wmdrichards
Copy link

setting allow_build_at_runtime=False in TrtConversionParams gives the following error:

ValueError: NodeDef mentions attr 'allow_build_at_runtime' not in Op<name=TRTEngineOp; signature=in_tensor: -> out_tensor:; attr=serialized_segment:string; attr=segment_func:func,default=[]; attr=InT:list(type),min=1,allowed=[DT_INT8, DT_HALF, DT_FLOAT, DT_INT32]; attr=OutT:list(type),min=1,allowed=[DT_INT8, DT_HALF, DT_FLOAT, DT_INT32]; attr=input_shapes:list(shape),default=[]; attr=max_cached_engines_count:int,default=1; attr=workspace_size_bytes:int; attr=precision_mode:string,allowed=["FP32", "FP16", "INT8"]; attr=calibration_data:string,default=""; attr=use_calibration:bool,default=true; attr=segment_funcdef_name:string,default=""; attr=cached_engine_batches:list(int),default=[],min=0; attr=fixed_input_size:bool,default=true; attr=output_shapes:list(shape),default=[]; attr=static_engine:bool,default=true>; NodeDef: {{node TRTEngineOp29e3f093_19b1_4530_9eed_0f4a8c0dbaf3_0_2}}. (Check whether your GraphDef-interpreting binary is up to date with your GraphDef-generating binary.).

Per tensorflow/tensorflow#34919 (comment), the change node.attr["allow_build_at_runtime"].b = False to node.attr["_allow_build_at_runtime"].b = False in trt_convert.py fixes the issue

@devalexqt
Copy link

Same shit:

ValueError: NodeDef mentions attr 'allow_build_at_runtime' not in Op<name=TRTEngineOp; signature=in_tensor: -> out_tensor:; attr=serialized_segment:string; attr=segment_func:func,default=[]; attr=InT:list(type),min=1,allowed=[DT_INT8, DT_HALF, DT_FLOAT, DT_INT32]; attr=OutT:list(type),min=1,allowed=[DT_INT8, DT_HALF, DT_FLOAT, DT_INT32]; attr=input_shapes:list(shape),default=[]; attr=max_cached_engines_count:int,default=1; attr=max_batch_size:int,default=1; attr=workspace_size_bytes:int; attr=precision_mode:string,allowed=["FP32", "FP16", "INT8"]; attr=calibration_data:string,default=""; attr=use_calibration:bool,default=true; attr=segment_funcdef_name:string,default=""; attr=cached_engine_batches:list(int),default=[],min=0; attr=fixed_input_size:bool,default=true; attr=output_shapes:list(shape),default=[]; attr=static_engine:bool,default=true>; NodeDef: {{node TRTEngineOp_0_0}}. (Check whether your GraphDef-interpreting binary is up to date with your GraphDef-generating binary.).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants