You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The error has been discovered thanks to level 3 logging enabled by GLOG_v environmental variable.
GLOG has reported, that:
Some operators use the same variables for reading/writing output. For example, when the fp32_model has been ran, one could observe that scale_op as well transpose2_op accept transpose_4.tmp_0 as their input to the operator (while according to the original graph they do not)
Subpart of a GLOG error documenting this:
As far as I understand that, this is a bug, since variable names should be unique (as long as they are enclosed in the same scope).
To illustrate the problem, please see the following figure depicting a different model (ernie_quant) which suffers from the same problem:
This is a blocking issue for INT8 Ernie quantization task, since our quantization system associates scales with variable names. And if the variable repeats in several places, we have end up with the same scales where we didn't mean to.
Reproduction
based on 8da0cd5
-CPU: including MKLDNN version v.20
-OS Platform Ubuntu 16.04
-Cmake orders -DCMAKE_BUILD_TYPE=RelWithDebInfo -DWITH_GPU=OFF -DON_INFER=ON -DWITH_MKLDNN=ON -DWITH_TESTING=ON -DWITH_PROFILER=ON -DWITH_STYLE_CHECK=OFF -DWITH_INFERENCE_API_TEST=ON
-API information To Reproduce
The graph is malformed when the memory_optimize_pass is enabled (e.g. via EnableMemoryOptim() method). With the pass being disabled, the graph of the model looks fine.
Current behaviour
The error has been discovered thanks to level 3 logging enabled by GLOG_v environmental variable.
GLOG has reported, that:
Some operators use the same variables for reading/writing output. For example, when the
fp32_model
has been ran, one could observe thatscale_op
as welltranspose2_op
accepttranspose_4.tmp_0
as their input to the operator (while according to the original graph they do not)Subpart of a GLOG error documenting this:
As far as I understand that, this is a bug, since variable names should be unique (as long as they are enclosed in the same scope).
To illustrate the problem, please see the following figure depicting a different model (ernie_quant) which suffers from the same problem:
This is a blocking issue for INT8 Ernie quantization task, since our quantization system associates scales with variable names. And if the variable repeats in several places, we have end up with the same scales where we didn't mean to.
Reproduction
-CPU: including MKLDNN version v.20
-OS Platform Ubuntu 16.04
-Cmake orders
-DCMAKE_BUILD_TYPE=RelWithDebInfo -DWITH_GPU=OFF -DON_INFER=ON -DWITH_MKLDNN=ON -DWITH_TESTING=ON -DWITH_PROFILER=ON -DWITH_STYLE_CHECK=OFF -DWITH_INFERENCE_API_TEST=ON
-API information
To Reproduce
Run any 4-input ernie model.
@luotao1 Could you please assign someone to help solving this issue?
The text was updated successfully, but these errors were encountered: