An open source AutoML toolkit for automate machine learning lifecycle, including feature engineering, neural architecture search, model compression and hyper-parameter tuning.
Describe the issue:
When running quantization_speedup.py in the tutorial file (I did not change anything), I got an error as below.
[2022-07-20 01:41:46] Model state_dict saved to ./log/mnist_model.pth
[2022-07-20 01:41:46] Mask dict saved to ./log/mnist_calibration.pth
[07/20/2022-01:41:49] [TRT] [W] DynamicRange(min: -0.424213, max: 2.82149). Dynamic range should be symmetric for better accuracy.
Traceback (most recent call last):
File "quantization_speedup.py", line 114, in <module>
engine.compress()
File "/opt/conda/lib/python3.7/site-packages/nni/compression/pytorch/quantization_speedup/integrated_tensorrt.py", line 298, in compress
context = self._tensorrt_build_withoutcalib(self.onnx_path)
File "/opt/conda/lib/python3.7/site-packages/nni/compression/pytorch/quantization_speedup/integrated_tensorrt.py", line 348, in _tensorrt_build_withoutcalib
engine = build_engine(onnx_path, self.onnx_config, self.extra_layer_bits, self.strict_datatype)
File "/opt/conda/lib/python3.7/site-packages/nni/compression/pytorch/quantization_speedup/integrated_tensorrt.py", line 198, in build_engine
handle_gemm(network, i, config)
File "/opt/conda/lib/python3.7/site-packages/nni/compression/pytorch/quantization_speedup/integrated_tensorrt.py", line 82, in handle_gemm
pre_in_tensor.dynamic_range = (tracked_min_input, tracked_max_input)
AttributeError: 'NoneType' object has no attribute 'dynamic_range'
Environment:
NNI version: 2.8
Training service (local|remote|pai|aml|etc): local
Describe the issue: When running quantization_speedup.py in the tutorial file (I did not change anything), I got an error as below.
Environment:
Configuration:
Log message:
How to reproduce it?: