Hi,
Does ResizeImage/Squeeze (within Preprocessor) of SSDMobilenet operation is supported by snpe 1.8.0 ?
Issue: my program got stuck at (for almost 30 mins):
INFO - INFO_TF_CONVERTING_SCOPES: Converting scope (Preprocessor/map/while/ResizeImage/): [u'Preprocessor/map/while/ResizeImage/ExpandDims', u'Preprocessor/map/while/ResizeImage/ResizeBilinear', u'Preprocessor/map/while/ResizeImage/Squeeze']
Model: SSD_Mobilenet
Type: Tensorflow
SNPE doesn't support SSD_Mobilenet at this time. We hope to support it in a future release.
SNPE doesn't support SSD_Mobilenet at this time. We hope to support it in a future release.
Hi Jesliger,
My doubt is does it supports "ResizeImage/Squeeze" operation ?
Executor failed to create kernel. Invalid argument: NodeDef mentions attr 'data_format' not in Op<name=DepthwiseConv2dNative; signature=input:T, filter:T -> output:T; attr=T:type,allowed=[DT_FLOAT, DT_DOUBLE]; attr=strides:list(int); attr=padding:string,allowed=["SAME", "VALID"]>; NodeDef: FeatureExtractor/MobilenetV1/MobilenetV1/Conv2d_1_depthwise/depthwise = DepthwiseConv2dNative[T=DT_FLOAT, data_format="NHWC", padding="SAME", strides=[1, 1, 1, 1], _device="/job:localhost/replica:0/task:0/cpu:0"](FeatureExtractor/MobilenetV1/MobilenetV1/Conv2d_0/Relu6, FeatureExtractor/MobilenetV1/Conv2d_1_depthwise/depthwise_weights/read)
[[Node: FeatureExtractor/MobilenetV1/MobilenetV1/Conv2d_1_depthwise/depthwise = DepthwiseConv2dNative[T=DT_FLOAT, data_format="NHWC", padding="SAME", strides=[1, 1, 1, 1], _device="/job:localhost/replica:0/task:0/cpu:0"](FeatureExtractor/MobilenetV1/MobilenetV1/Conv2d_0/Relu6, FeatureExtractor/MobilenetV1/Conv2d_1_depthwise/depthwise_weights/read)]]
2018-01-15 19:38:47,651 - 357 - WARNING - ERROR_TF_FALLBACK_TO_ONDEMAND_EVALUATION: Unable to resolve operation output shapes in single pass. Using on-demand evaluation!