BUG: Unexpected exception TypeError("_post_eval_hook() got an unexpected keyword argument 'accuracy'") happened during tuning.

[johnsGuo]

When I use tf2.5 with NIC, I want to use tensorboard, and my conf is

version: 1.0

model:                                               # mandatory. used to specify model specific information.
  name: origin_model
  framework: tensorflow                              # mandatory. supported values are tensorflow, pytorch, pytorch_ipex, onnxrt_integer, onnxrt_qlinear or mxnet; allow new framework backend extension.

quantization:                                        # optional. tuning constraints on model-wise for advance user to reduce tuning space.
  calibration:
    sampling_size: 10000                              # optional. default value is 100. used to set how many samples should be used in calibration.
  model_wise:                                        # optional. tuning constraints on model-wise for advance user to reduce tuning space.
    activation:
      algorithm: minmax
#  op_wise: {
#             'import/dnn/hiddenlayer_0/MatMul': {
#               'activation':  {'dtype': ['uint8'], 'algorithm': ['minmax'], 'scheme':['asym']},
#             }
#           }

tuning:
  tensorboard: true
  accuracy_criterion:
    relative:  0.01                                  # optional. default value is relative, other value is absolute. this example allows relative accuracy loss: 1%.
  exit_policy:
    timeout: 0                                       # optional. tuning timeout (seconds). default value is 0 which means early stop. combine with max_trials field to decide when to exit.
    max_trials: 100                                  # optional. max tune times. default value is 100. combine with timeout field to decide when to exit.
  random_seed: 9527  

Then I got this error, Is a Bug 2022-04-07 11:03:00 [ERROR] Unexpected exception TypeError("_post_eval_hook() got an unexpected keyword argument 'accuracy'") happened during tuning. Traceback (most recent call last): File "/root/anaconda3/envs/tf2/lib/python3.7/site-packages/neural_compressor/experimental/quantization.py", line 151, in execute self.strategy.traverse() File "/root/anaconda3/envs/tf2/lib/python3.7/site-packages/neural_compressor/strategy/strategy.py", line 323, in traverse self.baseline = self._evaluate(self.model) File "/root/anaconda3/envs/tf2/lib/python3.7/site-packages/neural_compressor/strategy/strategy.py", line 463, in _evaluate self.adaptor._post_eval_hook(model, accuracy=val[0])

[johnsGuo]

The tensorflow adaptor not has accuracy params, So raise the execption, You have pass it, So tensorboard in NIC is not avaliable now!

[ftian1]

thanks for the feedback. we will take a look on that

[qgao007]

Hi @GuoGuiRong,

Thank you for reporting the bug. It's fixed in our next release.