tflite 模型转rknn 错误

[复制链接] · 发表于 2022-12-30 15:04:10

1、Kernel_dequantize' has no buffer, init to zeros.是模型中有些算子不支持吗？
W Tensor b'conv2d_16/Kernel_dequantize' has no buffer, init to zeros.
W Tensor b'conv2d_16/Bias_dequantize' has no buffer, init to zeros.
D Convert layer add
D Convert layer relu
D Convert layer dequantize
D Convert layer dequantize
D Convert layer conv_2d
W Tensor b'classificator_8/Kernel_dequantize' has no buffer, init to zeros.
W Tensor b'classificator_8/Bias_dequantize' has no buffer, init to zeros.
D Convert layer dequantize
D Convert layer dequantize
D Convert layer conv_2d
W Tensor b'classificator_16/Kernel_dequantize' has no buffer, init to zeros.
W Tensor b'classificator_16/Bias_dequantize' has no buffer, init to zeros.
D Convert layer dequantize
D Convert layer dequantize
D Convert layer conv_2d
W Tensor b'regressor_8/Kernel_dequantize' has no buffer, init to zeros.
W Tensor b'regressor_8/Bias_dequantize' has no buffer, init to zeros.
D Convert layer dequantize
D Convert layer dequantize
D Convert layer conv_2d
W Tensor b'regressor_16/Kernel_dequantize' has no buffer, init to zeros.
W Tensor b'regressor_16/Bias_dequantize' has no buffer, init to zeros.
D Convert layer reshape
D Convert layer reshape
D Convert layer reshape
D Convert layer reshape
D Convert layer concatenation
D Convert layer concatenation
E Invalid tensor id(1), tensor(@conv2d/Kernel_dequantize_2

ut0)
W ----------------Warning(75)----------------
E Catch exception when loading tflite model: face_detection_short.tflite!
E Traceback (most recent call last):
E File "rknn/api/rknn_base.py", line 439, in rknn.api.rknn_base.RKNNBase.load_tflite
E File "rknn/base/RKNNlib/converter/tflite_loader.py", line 78, in rknn.base.RKNNlib.converter.tflite_loader.TFLiteLoader.load
E File "rknn/base/RKNNlib/converter/tflite_loader.py", line 418, in rknn.base.RKNNlib.converter.tflite_loader.ModelParser.parse
E File "rknn/base/RKNNlib/converter/tflite_loader.py", line 398, in rknn.base.RKNNlib.converter.tflite_loader.ModelParser._build_connections
E File "rknn/base/RKNNlib/layer/RKNNlayer.py", line 141, in rknn.base.RKNNlib.layer.RKNNlayer.RKNNLayer.add_input
E File "rknn/base/RKNNlib/layer/RKNNlayer.py", line 26, in rknn.base.RKNNlib.layer.RKNNlayer.IoStruct.add
E File "rknn/api/rknn_log.py", line 312, in rknn.api.rknn_log.RKNNLog.e
E ValueError: Invalid tensor id(1), tensor(@conv2d/Kernel_dequantize_2

ut0)

2、错误2，为什么会有关于下标的错误
W Verbose file path is invalid, debug info will not dump to file.
--> Config model
done
--> Loading model
I Model: face_landmark
I Version: 3
I Description: facemesh-lite.tflite.no_meta
I Subgraphs: 1
D import clients finished
D Convert layer conv_2d
D Convert layer prelu
E Catch exception when loading tflite model: face_landmark.tflite!
E Traceback (most recent call last):
E File "rknn/api/rknn_base.py", line 439, in rknn.api.rknn_base.RKNNBase.load_tflite
E File "rknn/base/RKNNlib/converter/tflite_loader.py", line 78, in rknn.base.RKNNlib.converter.tflite_loader.TFLiteLoader.load
E File "rknn/base/RKNNlib/converter/tflite_loader.py", line 416, in rknn.base.RKNNlib.converter.tflite_loader.ModelParser.parse
E File "rknn/base/RKNNlib/converter/tflite_loader.py", line 365, in rknn.base.RKNNlib.converter.tflite_loader.ModelParser._build_layers
E File "rknn/base/RKNNlib/converter/tflite_loader.py", line 118, in rknn.base.RKNNlib.converter.tflite_loader.ModelParser._new_layer
E IndexError: only integers, slices (`:`), ellipsis (`...`), numpy.newaxis (`None`) and integer or boolean arrays are valid indices
Load model failed!