Vai_q_pytorch prints warning message on screen when there is
issue may causing the quantization result has problem or incomplete (check according to
the message text), but the process can be performed to its end, the format of this kind
of message is "[VAIQ_WARN][MESSAGE_ID]: message text"
List important warning messages in the following table:
| Message ID | Description |
|---|---|
| QUANTIZER_TORCH_BATCHNORM_AFFINE | BatchNorm OP attribute affine=False has been replaced by affine=True when parsing the model. |
| QUANTIZER_TORCH_BITWIDTH_MISMATCH | Bit width setting in configuration file is conflict with that from torch_quantizer API, will use that in configuration file. |
| QUANTIZER_TORCH_CONVERT_XMODEL | Convert to xmodel failed. Check message text to locate the reason. |
| QUANTIZER_TORCH_CUDA_UNAVAILABLE | CUDA (HIP) is not available, change device to CPU |
| QUANTIZER_TORCH_DATA_PARALLEL | Data parallel is not supported. The wrapper 'torch.nn.DataParallel' has been removed in vai_q_pytorch. |
| QUANTIZER_TORCH_DEPLOY_MODEL | Only quantization aware training process has deployable model. |
| QUANTIZER_TORCH_DEVICE_MISMATCH | The Device of input arguments mismatch with quantizer device type. |
| QUANTIZER_TORCH_EXPORT_XMODEL | Failed to generate xmodel due to some reasons. Refer to the message text. |
| QUANTIZER_TORCH_FINETUNE_IGNORED | Fast fine-tune function will be ignored in test mode! |
| QUANTIZER_TORCH_FLOAT_OP | vai_q_pytorch recognize the list OP as a float operator by default. |
| QUANTIZER_TORCH_INSPECTOR_PATTERN | The OP may be fused by compiler and will be assigned to DPU. |
| QUANTIZER_TORCH_LEAKYRELU | Force to change negative_slope of LeakyReLU to 0.1015625 because DPU only supports this value. It is recommended to change all negative_slope of LeakyReLU to 0.1015625 and re-train the float model for better deployed model accuracy. |
| QUANTIZER_TORCH_MATPLOTLIB | matplotlib is needed for visualization but not found. It needs to be installed. |
| QUANTIZER_TORCH_MEMORY_SHORTAGE | There is no enough memory for fast fine-tune and this process will be ignored!. Try to use a smaller calibration dataset. |
| QUANTIZER_TORCH_NO_XIR | Can't find XIR package in environment. It needs to be installed. |
| QUANTIZER_TORCH_REPLACE_RELU6 | ReLU6 has been replaced by ReLU. |
| QUANTIZER_TORCH_REPLACE_SIGMOID | Sigmoid has been replaced by Hardsigmoid. |
| QUANTIZER_TORCH_REPLACE_SILU | SiLU has been replaced by Hardswish. |
| QUANTIZER_TORCH_SHIFT_CHECK | Quantization scale is too large or too small. |
| QUANTIZER_TORCH_TENSOR_NOT_QUANTIZED | Some tensors are not quantized, please check their particularity. |
| QUANTIZER_TORCH_TENSOR_TYPE_NOT_QUANTIZABLE | The tensor type of the node cannot be quantized. Only support float32/double/float16 quantization. |
| QUANTIZER_TORCH_TENSOR_VALUE_INVALID | The tensor has "inf" or "nan" value. Quantization for this tensor is ignored. |
| QUANTIZER_TORCH_TORCH_VERSION | Only support exporting torch script with pytorch 1.10 and later version. |
| QUANTIZER_TORCH_XIR_MISMATCH | XIR version does not match current vai_q_pytorch. |
| QUANTIZER_TORCH_XMODEL_DEVICE | Not support to dump xmodel when target device is not DPU. |
| QUANTIZER_TORCH_REUSED_MODULE | Reused module may lead to low accuracy of QAT, make sure this is what you expect. Refer to the message text to locate the module with issue. |
| QUANTIZER_TORCH_DEPRECATED_ARGUMENT | The argument "device" is no longer needed. Device information is get from the model directly. |
| QUANTIZER_TORCH_SCALE_VALUE | Exported scale values are not trained. |