vai_q_tensorflow2 Error Codes - 3.5 English

Vitis AI User Guide (UG1414)

Document ID
UG1414
Release Date
2023-09-28
Version
3.5 English
Table 1. vai_q_tensorflow2 Error Codes
Error Description Error Types Causes and Solutions
Quantizer_TF2_Unsupported_Layer Unsupported layer type Layer is not a `tf.keras.layers.Layer` or this layer is not yet supported. By default, this layer will not be quantized and will be mapped to run on the CPU. You can use the experimental support for customizing quantize strategy to define the quantization behavior of it.
Quantizer_TF2_Unsupported_Model Unsupported model type Only tf.keras sequential or functional models can be supported. Subclassing model is not supported currently. Convert it to a sequential or functional model and try again.
Quantizer_TF2_Invalid_Input_Shape Invalid input shape The input_shape parameter is not valid. Check and set the correct value for it.
Quantizer_TF2_Invalid_Calib_Dataset Invalid calibration dataset The calibration dataset is not valid. Check and set the value for it.
Quantizer_TF2_Invalid_Target Invalid target The target parameter is not valid. Check and set the correct value for it.