MACE v0.12.0 Error log - validation failed

conversion.log

# python tools/converter.py convert --config=/models/retinanet/retinanet3.yml 
CMD> bazel build //mace/proto:mace_py
WARNING: --batch mode is deprecated. Please instead explicitly shut down your Bazel server using the command "bazel shutdown".
Loading:
Loading: 0 packages loaded
Analyzing: target //mace/proto:mace_py (4 packages loaded)
INFO: Analysed target //mace/proto:mace_py (17 packages loaded).
INFO: Found 1 target...
[2 / 9] [-----] BazelWorkspaceStatusAction stable-status.txt
Target //mace/proto:mace_py up-to-date:
bazel-genfiles/mace/proto/mace_pb2.py
INFO: Elapsed time: 3.049s, Critical Path: 0.07s
INFO: 0 processes.
INFO: Build completed successfully, 1 total action
INFO: Build completed successfully, 1 total action

CMD> cp -f bazel-genfiles/mace/proto/mace_pb2.py tools/python/py_proto

CMD> bazel build //third_party/caffe:caffe_py
WARNING: --batch mode is deprecated. Please instead explicitly shut down your Bazel server using the command "bazel shutdown".
Loading:
Loading: 0 packages loaded
Analyzing: target //third_party/caffe:caffe_py (3 packages loaded)
Analyzing: target //third_party/caffe:caffe_py (12 packages loaded)
INFO: Analysed target //third_party/caffe:caffe_py (17 packages loaded).
INFO: Found 1 target...
[0 / 1] [-----] BazelWorkspaceStatusAction stable-status.txt
Target //third_party/caffe:caffe_py up-to-date:
bazel-genfiles/third_party/caffe/caffe_pb2.py
INFO: Elapsed time: 3.617s, Critical Path: 0.14s
INFO: 0 processes.
INFO: Build completed successfully, 1 total action
INFO: Build completed successfully, 1 total action

CMD> cp -f bazel-genfiles/third_party/caffe/caffe_pb2.py tools/python/py_proto

{'library_name': 'libretinanet', 'target_abis': ['arm64-v8a'], 'model_graph_format': 'code', 'model_data_format': 'code', 'models': {'retinanet': {'platform': 'caffe', 'model_file_path': '/models/retinanet/retinanet3.prototxt', 'weight_file_path': '/models/retinanet/retinanet3.caffemodel', 'model_sha256_checksum': '638e05fc466737c3b8fc36261adaaff40cbd4de5a8c72a46b37f2b00f01180e1', 'weight_sha256_checksum': '6222910a773c693c23b4765baba4ed8427e9f3c11781060918e6282a297a7437', 'subgraphs': [{'input_tensors': ['data'], 'input_shapes': ['1,3,320,320'], 'input_data_formats': ['NCHW'], 'output_tensors': ['face_rpn_cls_prob_reshape_stride32', 'face_rpn_bbox_pred_stride32', 'face_rpn_landmark_pred_stride32', 'face_rpn_cls_prob_reshape_stride16', 'face_rpn_bbox_pred_stride16', 'face_rpn_landmark_pred_stride16', 'face_rpn_cls_prob_reshape_stride8', 'face_rpn_bbox_pred_stride8', 'face_rpn_landmark_pred_stride8'], 'output_shapes': ['1,4,10,10', '1,8,10,10', '1,20,10,10', '1,4,20,20', '1,8,20,20', '1,20,20,20', '1,4,40,40', '1,8,40,40', '1,20,40,40'], 'output_data_formats': ['NCHW', 'NCHW', 'NCHW', 'NCHW', 'NCHW', 'NCHW', 'NCHW', 'NCHW', 'NCHW']}], 'obfuscate': 0, 'runtime': 'gpu', 'winograd': 4}}}
* Generate mace engine creator source
Generate mace engine creator source done!

{'platform': 'caffe', 'model_file_path': '/models/retinanet/retinanet3.prototxt', 'weight_file_path': '/models/retinanet/retinanet3.caffemodel', 'model_sha256_checksum': '638e05fc466737c3b8fc36261adaaff40cbd4de5a8c72a46b37f2b00f01180e1', 'weight_sha256_checksum': '6222910a773c693c23b4765baba4ed8427e9f3c11781060918e6282a297a7437', 'obfuscate': 0, 'runtime': 'gpu', 'winograd': 4, 'input_tensors': ['data'], 'input_shapes': ['1,3,320,320'], 'input_data_formats': ['NCHW'], 'output_tensors': ['face_rpn_cls_prob_reshape_stride32', 'face_rpn_bbox_pred_stride32', 'face_rpn_landmark_pred_stride32', 'face_rpn_cls_prob_reshape_stride16', 'face_rpn_bbox_pred_stride16', 'face_rpn_landmark_pred_stride16', 'face_rpn_cls_prob_reshape_stride8', 'face_rpn_bbox_pred_stride8', 'face_rpn_landmark_pred_stride8'], 'output_shapes': ['1,4,10,10', '1,8,10,10', '1,20,10,10', '1,4,20,20', '1,8,20,20', '1,20,20,20', '1,4,40,40', '1,8,40,40', '1,20,40,40'], 'output_data_formats': ['NCHW', 'NCHW', 'NCHW', 'NCHW', 'NCHW', 'NCHW', 'NCHW', 'NCHW', 'NCHW']}
{'platform': <Platform.CAFFE: 1>, 'model_file_path': '/models/retinanet/retinanet3.prototxt', 'weight_file_path': '/models/retinanet/retinanet3.caffemodel', 'model_sha256_checksum': '638e05fc466737c3b8fc36261adaaff40cbd4de5a8c72a46b37f2b00f01180e1', 'weight_sha256_checksum': '6222910a773c693c23b4765baba4ed8427e9f3c11781060918e6282a297a7437', 'obfuscate': 0, 'runtime': <DeviceType.GPU: 2>, 'winograd': 4, 'input_tensors': ['data'], 'input_shapes': [[1, 3, 320, 320]], 'input_data_formats': [<DataFormat.NCHW: 2>], 'output_tensors': ['face_rpn_cls_prob_reshape_stride32', 'face_rpn_bbox_pred_stride32', 'face_rpn_landmark_pred_stride32', 'face_rpn_cls_prob_reshape_stride16', 'face_rpn_bbox_pred_stride16', 'face_rpn_landmark_pred_stride16', 'face_rpn_cls_prob_reshape_stride8', 'face_rpn_bbox_pred_stride8', 'face_rpn_landmark_pred_stride8'], 'output_shapes': [[1, 4, 10, 10], [1, 8, 10, 10], [1, 20, 10, 10], [1, 4, 20, 20], [1, 8, 20, 20], [1, 20, 20, 20], [1, 4, 40, 40], [1, 8, 40, 40], [1, 20, 40, 40]], 'output_data_formats': [<DataFormat.NCHW: 2>, <DataFormat.NCHW: 2>, <DataFormat.NCHW: 2>, <DataFormat.NCHW: 2>, <DataFormat.NCHW: 2>, <DataFormat.NCHW: 2>, <DataFormat.NCHW: 2>, <DataFormat.NCHW: 2>, <DataFormat.NCHW: 2>], 'data_type': 3, 'input_data_types': [1], 'input_ranges': [[-1.0, 1.0]], 'output_data_types': [1, 1, 1, 1, 1, 1, 1, 1, 1]}
Transform model to one that can better run on device
deconv layer rf_c3_upsampling (DepthwiseDeconv2d) input:[ 1 64 10 10] filter:[64, 1, 4, 4] output:[ 1 64 20 20]
deconv layer rf_c2_upsampling (DepthwiseDeconv2d) input:[ 1 64 20 20] filter:[64, 1, 4, 4] output:[ 1 64 40 40]
Transform Caffe Reshape
Transform Caffe Reshape
Transform Caffe Reshape
Transform Caffe Reshape
Transform Caffe Reshape
Transform Caffe Reshape
Fold conv and bn: mobilenetv21_features_conv0_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck0_conv1_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck1_conv0_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck1_conv2_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck2_conv0_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck2_conv2_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck3_conv0_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck3_conv2_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck4_conv0_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck4_conv2_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck5_conv0_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck5_conv2_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck6_conv0_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck6_conv2_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck7_conv0_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck7_conv2_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck8_conv0_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck8_conv2_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck9_conv0_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck9_conv2_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck10_conv0_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck10_conv2_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck11_conv0_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck11_conv2_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck12_conv0_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck12_conv2_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck13_conv0_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck13_conv2_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck14_conv0_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck14_conv2_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck15_conv0_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck15_conv2_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck16_conv0_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_linearbottleneck16_conv2_fwd(Conv2D)
Fold conv and bn: mobilenetv21_features_conv1_fwd(Conv2D)
Fold conv and bn: rf_c3_lateral(Conv2D)
Fold conv and bn: rf_c3_det_conv1(Conv2D)
Fold conv and bn: rf_c3_det_context_conv1(Conv2D)
Fold conv and bn: rf_c3_det_context_conv2(Conv2D)
Fold conv and bn: rf_c3_det_context_conv3_1(Conv2D)
Fold conv and bn: rf_c3_det_context_conv3_2(Conv2D)
Fold conv and bn: rf_c2_lateral(Conv2D)
Fold conv and bn: rf_c2_aggr(Conv2D)
Fold conv and bn: rf_c2_det_conv1(Conv2D)
Fold conv and bn: rf_c2_det_context_conv1(Conv2D)
Fold conv and bn: rf_c2_det_context_conv2(Conv2D)
Fold conv and bn: rf_c2_det_context_conv3_1(Conv2D)
Fold conv and bn: rf_c2_det_context_conv3_2(Conv2D)
Fold conv and bn: rf_c1_red_conv(Conv2D)
Fold conv and bn: rf_c1_aggr(Conv2D)
Fold conv and bn: rf_c1_det_conv1(Conv2D)
Fold conv and bn: rf_c1_det_context_conv1(Conv2D)
Fold conv and bn: rf_c1_det_context_conv2(Conv2D)
Fold conv and bn: rf_c1_det_context_conv3_1(Conv2D)
Fold conv and bn: rf_c1_det_context_conv3_2(Conv2D)
Fold depthwise conv and bn: mobilenetv21_features_linearbottleneck0_conv0_fwd(DepthwiseConv2d)
Fold depthwise conv and bn: mobilenetv21_features_linearbottleneck1_conv1_fwd(DepthwiseConv2d)
Fold depthwise conv and bn: mobilenetv21_features_linearbottleneck2_conv1_fwd(DepthwiseConv2d)
Fold depthwise conv and bn: mobilenetv21_features_linearbottleneck3_conv1_fwd(DepthwiseConv2d)
Fold depthwise conv and bn: mobilenetv21_features_linearbottleneck4_conv1_fwd(DepthwiseConv2d)
Fold depthwise conv and bn: mobilenetv21_features_linearbottleneck5_conv1_fwd(DepthwiseConv2d)
Fold depthwise conv and bn: mobilenetv21_features_linearbottleneck6_conv1_fwd(DepthwiseConv2d)
Fold depthwise conv and bn: mobilenetv21_features_linearbottleneck7_conv1_fwd(DepthwiseConv2d)
Fold depthwise conv and bn: mobilenetv21_features_linearbottleneck8_conv1_fwd(DepthwiseConv2d)
Fold depthwise conv and bn: mobilenetv21_features_linearbottleneck9_conv1_fwd(DepthwiseConv2d)
Fold depthwise conv and bn: mobilenetv21_features_linearbottleneck10_conv1_fwd(DepthwiseConv2d)
Fold depthwise conv and bn: mobilenetv21_features_linearbottleneck11_conv1_fwd(DepthwiseConv2d)
Fold depthwise conv and bn: mobilenetv21_features_linearbottleneck12_conv1_fwd(DepthwiseConv2d)
Fold depthwise conv and bn: mobilenetv21_features_linearbottleneck13_conv1_fwd(DepthwiseConv2d)
Fold depthwise conv and bn: mobilenetv21_features_linearbottleneck14_conv1_fwd(DepthwiseConv2d)
Fold depthwise conv and bn: mobilenetv21_features_linearbottleneck15_conv1_fwd(DepthwiseConv2d)
Fold depthwise conv and bn: mobilenetv21_features_linearbottleneck16_conv1_fwd(DepthwiseConv2d)
Fold activation: mobilenetv21_features_conv0_fwd(Conv2D)
Fold activation: mobilenetv21_features_linearbottleneck0_conv0_fwd(DepthwiseConv2d)
Fold activation: mobilenetv21_features_linearbottleneck1_conv0_fwd(Conv2D)
Fold activation: mobilenetv21_features_linearbottleneck1_conv1_fwd(DepthwiseConv2d)
Fold activation: mobilenetv21_features_linearbottleneck2_conv0_fwd(Conv2D)
Fold activation: mobilenetv21_features_linearbottleneck2_conv1_fwd(DepthwiseConv2d)
Fold activation: mobilenetv21_features_linearbottleneck3_conv0_fwd(Conv2D)
Fold activation: mobilenetv21_features_linearbottleneck3_conv1_fwd(DepthwiseConv2d)
Fold activation: mobilenetv21_features_linearbottleneck4_conv0_fwd(Conv2D)
Fold activation: mobilenetv21_features_linearbottleneck4_conv1_fwd(DepthwiseConv2d)
Fold activation: mobilenetv21_features_linearbottleneck5_conv0_fwd(Conv2D)
Fold activation: mobilenetv21_features_linearbottleneck5_conv1_fwd(DepthwiseConv2d)
Fold activation: mobilenetv21_features_linearbottleneck6_conv0_fwd(Conv2D)
Fold activation: mobilenetv21_features_linearbottleneck6_conv1_fwd(DepthwiseConv2d)
Fold activation: mobilenetv21_features_linearbottleneck7_conv0_fwd(Conv2D)
Fold activation: mobilenetv21_features_linearbottleneck7_conv1_fwd(DepthwiseConv2d)
Fold activation: mobilenetv21_features_linearbottleneck8_conv0_fwd(Conv2D)
Fold activation: mobilenetv21_features_linearbottleneck8_conv1_fwd(DepthwiseConv2d)
Fold activation: mobilenetv21_features_linearbottleneck9_conv0_fwd(Conv2D)
Fold activation: mobilenetv21_features_linearbottleneck9_conv1_fwd(DepthwiseConv2d)
Fold activation: mobilenetv21_features_linearbottleneck10_conv0_fwd(Conv2D)
Fold activation: mobilenetv21_features_linearbottleneck10_conv1_fwd(DepthwiseConv2d)
Fold activation: mobilenetv21_features_linearbottleneck11_conv0_fwd(Conv2D)
Fold activation: mobilenetv21_features_linearbottleneck11_conv1_fwd(DepthwiseConv2d)
Fold activation: mobilenetv21_features_linearbottleneck12_conv0_fwd(Conv2D)
Fold activation: mobilenetv21_features_linearbottleneck12_conv1_fwd(DepthwiseConv2d)
Fold activation: mobilenetv21_features_linearbottleneck13_conv0_fwd(Conv2D)
Fold activation: mobilenetv21_features_linearbottleneck13_conv1_fwd(DepthwiseConv2d)
Fold activation: mobilenetv21_features_linearbottleneck14_conv0_fwd(Conv2D)
Fold activation: mobilenetv21_features_linearbottleneck14_conv1_fwd(DepthwiseConv2d)
Fold activation: mobilenetv21_features_linearbottleneck15_conv0_fwd(Conv2D)
Fold activation: mobilenetv21_features_linearbottleneck15_conv1_fwd(DepthwiseConv2d)
Fold activation: mobilenetv21_features_linearbottleneck16_conv0_fwd(Conv2D)
Fold activation: mobilenetv21_features_linearbottleneck16_conv1_fwd(DepthwiseConv2d)
Fold activation: mobilenetv21_features_conv1_fwd(Conv2D)
Fold activation: rf_c3_lateral(Conv2D)
Fold activation: rf_c3_det_context_conv1(Conv2D)
Fold activation: rf_c3_det_context_conv3_1(Conv2D)
Fold activation: rf_c2_lateral(Conv2D)
Fold activation: rf_c2_aggr(Conv2D)
Fold activation: rf_c2_det_context_conv1(Conv2D)
Fold activation: rf_c2_det_context_conv3_1(Conv2D)
Fold activation: rf_c1_red_conv(Conv2D)
Fold activation: rf_c1_aggr(Conv2D)
Fold activation: rf_c1_det_context_conv1(Conv2D)
Fold activation: rf_c1_det_context_conv3_1(Conv2D)
update op with float data type
Add OpenCL informations
Sort by execution
Final ops:
mobilenetv21_features_relu60_relu6 (Conv2D, index:0): [[1, 32, 160, 160]]
mobilenetv21_features_linearbottleneck0_relu60_relu6 (DepthwiseConv2d, index:1): [[1, 32, 160, 160]]
mobilenetv21_features_linearbottleneck0_conv1_fwd (Conv2D, index:2): [[1, 16, 160, 160]]
mobilenetv21_features_linearbottleneck1_relu60_relu6 (Conv2D, index:3): [[1, 96, 160, 160]]
mobilenetv21_features_linearbottleneck1_relu61_relu6 (DepthwiseConv2d, index:4): [[1, 96, 80, 80]]
mobilenetv21_features_linearbottleneck1_conv2_fwd (Conv2D, index:5): [[1, 24, 80, 80]]
mobilenetv21_features_linearbottleneck2_relu60_relu6 (Conv2D, index:6): [[1, 144, 80, 80]]
mobilenetv21_features_linearbottleneck2_relu61_relu6 (DepthwiseConv2d, index:7): [[1, 144, 80, 80]]
mobilenetv21_features_linearbottleneck2_conv2_fwd (Conv2D, index:8): [[1, 24, 80, 80]]
mobilenetv21_features_linearbottleneck2_elemwise_add0 (Eltwise, index:9): [[1, 24, 80, 80]]
mobilenetv21_features_linearbottleneck3_relu60_relu6 (Conv2D, index:10): [[1, 144, 80, 80]]
mobilenetv21_features_linearbottleneck3_relu61_relu6 (DepthwiseConv2d, index:11): [[1, 144, 40, 40]]
mobilenetv21_features_linearbottleneck3_conv2_fwd (Conv2D, index:12): [[1, 32, 40, 40]]
mobilenetv21_features_linearbottleneck4_relu60_relu6 (Conv2D, index:13): [[1, 192, 40, 40]]
mobilenetv21_features_linearbottleneck4_relu61_relu6 (DepthwiseConv2d, index:14): [[1, 192, 40, 40]]
mobilenetv21_features_linearbottleneck4_conv2_fwd (Conv2D, index:15): [[1, 32, 40, 40]]
mobilenetv21_features_linearbottleneck4_elemwise_add0 (Eltwise, index:16): [[1, 32, 40, 40]]
mobilenetv21_features_linearbottleneck5_relu60_relu6 (Conv2D, index:17): [[1, 192, 40, 40]]
mobilenetv21_features_linearbottleneck5_relu61_relu6 (DepthwiseConv2d, index:18): [[1, 192, 40, 40]]
mobilenetv21_features_linearbottleneck5_conv2_fwd (Conv2D, index:19): [[1, 32, 40, 40]]
mobilenetv21_features_linearbottleneck5_elemwise_add0 (Eltwise, index:20): [[1, 32, 40, 40]]
mobilenetv21_features_linearbottleneck6_relu60_relu6 (Conv2D, index:21): [[1, 192, 40, 40]]
mobilenetv21_features_linearbottleneck6_relu61_relu6 (DepthwiseConv2d, index:22): [[1, 192, 20, 20]]
mobilenetv21_features_linearbottleneck6_conv2_fwd (Conv2D, index:23): [[1, 64, 20, 20]]
mobilenetv21_features_linearbottleneck7_relu60_relu6 (Conv2D, index:24): [[1, 384, 20, 20]]
mobilenetv21_features_linearbottleneck7_relu61_relu6 (DepthwiseConv2d, index:25): [[1, 384, 20, 20]]
mobilenetv21_features_linearbottleneck7_conv2_fwd (Conv2D, index:26): [[1, 64, 20, 20]]
mobilenetv21_features_linearbottleneck7_elemwise_add0 (Eltwise, index:27): [[1, 64, 20, 20]]
mobilenetv21_features_linearbottleneck8_relu60_relu6 (Conv2D, index:28): [[1, 384, 20, 20]]
mobilenetv21_features_linearbottleneck8_relu61_relu6 (DepthwiseConv2d, index:29): [[1, 384, 20, 20]]
mobilenetv21_features_linearbottleneck8_conv2_fwd (Conv2D, index:30): [[1, 64, 20, 20]]
mobilenetv21_features_linearbottleneck8_elemwise_add0 (Eltwise, index:31): [[1, 64, 20, 20]]
mobilenetv21_features_linearbottleneck9_relu60_relu6 (Conv2D, index:32): [[1, 384, 20, 20]]
mobilenetv21_features_linearbottleneck9_relu61_relu6 (DepthwiseConv2d, index:33): [[1, 384, 20, 20]]
mobilenetv21_features_linearbottleneck9_conv2_fwd (Conv2D, index:34): [[1, 64, 20, 20]]
mobilenetv21_features_linearbottleneck9_elemwise_add0 (Eltwise, index:35): [[1, 64, 20, 20]]
mobilenetv21_features_linearbottleneck10_relu60_relu6 (Conv2D, index:36): [[1, 384, 20, 20]]
mobilenetv21_features_linearbottleneck10_relu61_relu6 (DepthwiseConv2d, index:37): [[1, 384, 20, 20]]
mobilenetv21_features_linearbottleneck10_conv2_fwd (Conv2D, index:38): [[1, 96, 20, 20]]
mobilenetv21_features_linearbottleneck11_relu60_relu6 (Conv2D, index:39): [[1, 576, 20, 20]]
mobilenetv21_features_linearbottleneck11_relu61_relu6 (DepthwiseConv2d, index:40): [[1, 576, 20, 20]]
mobilenetv21_features_linearbottleneck11_conv2_fwd (Conv2D, index:41): [[1, 96, 20, 20]]
mobilenetv21_features_linearbottleneck11_elemwise_add0 (Eltwise, index:42): [[1, 96, 20, 20]]
mobilenetv21_features_linearbottleneck12_relu60_relu6 (Conv2D, index:43): [[1, 576, 20, 20]]
mobilenetv21_features_linearbottleneck12_relu61_relu6 (DepthwiseConv2d, index:44): [[1, 576, 20, 20]]
mobilenetv21_features_linearbottleneck12_conv2_fwd (Conv2D, index:45): [[1, 96, 20, 20]]
mobilenetv21_features_linearbottleneck12_elemwise_add0 (Eltwise, index:46): [[1, 96, 20, 20]]
mobilenetv21_features_linearbottleneck13_relu60_relu6 (Conv2D, index:47): [[1, 576, 20, 20]]
mobilenetv21_features_linearbottleneck13_relu61_relu6 (DepthwiseConv2d, index:48): [[1, 576, 10, 10]]
mobilenetv21_features_linearbottleneck13_conv2_fwd (Conv2D, index:49): [[1, 160, 10, 10]]
mobilenetv21_features_linearbottleneck14_relu60_relu6 (Conv2D, index:50): [[1, 960, 10, 10]]
mobilenetv21_features_linearbottleneck14_relu61_relu6 (DepthwiseConv2d, index:51): [[1, 960, 10, 10]]
mobilenetv21_features_linearbottleneck14_conv2_fwd (Conv2D, index:52): [[1, 160, 10, 10]]
mobilenetv21_features_linearbottleneck14_elemwise_add0 (Eltwise, index:53): [[1, 160, 10, 10]]
mobilenetv21_features_linearbottleneck15_relu60_relu6 (Conv2D, index:54): [[1, 960, 10, 10]]
mobilenetv21_features_linearbottleneck15_relu61_relu6 (DepthwiseConv2d, index:55): [[1, 960, 10, 10]]
mobilenetv21_features_linearbottleneck15_conv2_fwd (Conv2D, index:56): [[1, 160, 10, 10]]
mobilenetv21_features_linearbottleneck15_elemwise_add0 (Eltwise, index:57): [[1, 160, 10, 10]]
mobilenetv21_features_linearbottleneck16_relu60_relu6 (Conv2D, index:58): [[1, 960, 10, 10]]
mobilenetv21_features_linearbottleneck16_relu61_relu6 (DepthwiseConv2d, index:59): [[1, 960, 10, 10]]
mobilenetv21_features_linearbottleneck16_conv2_fwd (Conv2D, index:60): [[1, 320, 10, 10]]
mobilenetv21_features_relu61_relu6 (Conv2D, index:61): [[1, 1280, 10, 10]]
rf_c3_lateral_relu (Conv2D, index:62): [[1, 64, 10, 10]]
rf_c3_det_conv1 (Conv2D, index:63): [[1, 32, 10, 10]]
rf_c3_det_context_conv1_relu (Conv2D, index:64): [[1, 16, 10, 10]]
rf_c3_det_context_conv2 (Conv2D, index:65): [[1, 16, 10, 10]]
rf_c3_det_context_conv3_1_relu (Conv2D, index:66): [[1, 16, 10, 10]]
rf_c3_det_context_conv3_2 (Conv2D, index:67): [[1, 16, 10, 10]]
rf_c3_det_concat (Concat, index:68): [[1, 64, 10, 10]]
rf_c3_det_concat_relu (Activation, index:69): [[1, 64, 10, 10]]
face_rpn_cls_score_stride32 (Conv2D, index:70): [[1, 4, 10, 10]]
face_rpn_cls_score_reshape_stride32 (Reshape, index:71): [[1, 2, 20, 10]]
face_rpn_cls_prob_stride32 (Softmax, index:72): [[1, 2, 20, 10]]
face_rpn_cls_prob_reshape_stride32 (Reshape, index:73): [[1, 4, 10, 10]]
face_rpn_bbox_pred_stride32 (Conv2D, index:74): [[1, 8, 10, 10]]
face_rpn_landmark_pred_stride32 (Conv2D, index:75): [[1, 20, 10, 10]]
rf_c2_lateral_relu (Conv2D, index:76): [[1, 64, 20, 20]]
rf_c3_upsampling (DepthwiseDeconv2d, index:77): [[1, 64, 20, 20]]
crop0 (Crop, index:78): [[1, 64, 20, 20]]
_plus0 (Eltwise, index:79): [[1, 64, 20, 20]]
rf_c2_aggr_relu (Conv2D, index:80): [[1, 64, 20, 20]]
rf_c2_det_conv1 (Conv2D, index:81): [[1, 32, 20, 20]]
rf_c2_det_context_conv1_relu (Conv2D, index:82): [[1, 16, 20, 20]]
rf_c2_det_context_conv2 (Conv2D, index:83): [[1, 16, 20, 20]]
rf_c2_det_context_conv3_1_relu (Conv2D, index:84): [[1, 16, 20, 20]]
rf_c2_det_context_conv3_2 (Conv2D, index:85): [[1, 16, 20, 20]]
rf_c2_det_concat (Concat, index:86): [[1, 64, 20, 20]]
rf_c2_det_concat_relu (Activation, index:87): [[1, 64, 20, 20]]
face_rpn_cls_score_stride16 (Conv2D, index:88): [[1, 4, 20, 20]]
face_rpn_cls_score_reshape_stride16 (Reshape, index:89): [[1, 2, 40, 20]]
face_rpn_cls_prob_stride16 (Softmax, index:90): [[1, 2, 40, 20]]
face_rpn_cls_prob_reshape_stride16 (Reshape, index:91): [[1, 4, 20, 20]]
face_rpn_bbox_pred_stride16 (Conv2D, index:92): [[1, 8, 20, 20]]
face_rpn_landmark_pred_stride16 (Conv2D, index:93): [[1, 20, 20, 20]]
rf_c1_red_conv_relu (Conv2D, index:94): [[1, 64, 40, 40]]
rf_c2_upsampling (DepthwiseDeconv2d, index:95): [[1, 64, 40, 40]]
crop1 (Crop, index:96): [[1, 64, 40, 40]]
_plus1 (Eltwise, index:97): [[1, 64, 40, 40]]
rf_c1_aggr_relu (Conv2D, index:98): [[1, 64, 40, 40]]
rf_c1_det_conv1 (Conv2D, index:99): [[1, 32, 40, 40]]
rf_c1_det_context_conv1_relu (Conv2D, index:100): [[1, 16, 40, 40]]
rf_c1_det_context_conv2 (Conv2D, index:101): [[1, 16, 40, 40]]
rf_c1_det_context_conv3_1_relu (Conv2D, index:102): [[1, 16, 40, 40]]
rf_c1_det_context_conv3_2 (Conv2D, index:103): [[1, 16, 40, 40]]
rf_c1_det_concat (Concat, index:104): [[1, 64, 40, 40]]
rf_c1_det_concat_relu (Activation, index:105): [[1, 64, 40, 40]]
face_rpn_cls_score_stride8 (Conv2D, index:106): [[1, 4, 40, 40]]
face_rpn_cls_score_reshape_stride8 (Reshape, index:107): [[1, 2, 80, 40]]
face_rpn_cls_prob_stride8 (Softmax, index:108): [[1, 2, 80, 40]]
face_rpn_cls_prob_reshape_stride8 (Reshape, index:109): [[1, 4, 40, 40]]
face_rpn_bbox_pred_stride8 (Conv2D, index:110): [[1, 8, 40, 40]]
face_rpn_landmark_pred_stride8 (Conv2D, index:111): [[1, 20, 40, 40]]
update data format
Transpose arguments based on data format
Transpose output shapes: mobilenetv21_features_relu60_relu6(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck0_relu60_relu6(DepthwiseConv2d)
Transpose output shapes: mobilenetv21_features_linearbottleneck0_conv1_fwd(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck1_relu60_relu6(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck1_relu61_relu6(DepthwiseConv2d)
Transpose output shapes: mobilenetv21_features_linearbottleneck1_conv2_fwd(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck2_relu60_relu6(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck2_relu61_relu6(DepthwiseConv2d)
Transpose output shapes: mobilenetv21_features_linearbottleneck2_conv2_fwd(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck2_elemwise_add0(Eltwise)
Transpose output shapes: mobilenetv21_features_linearbottleneck3_relu60_relu6(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck3_relu61_relu6(DepthwiseConv2d)
Transpose output shapes: mobilenetv21_features_linearbottleneck3_conv2_fwd(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck4_relu60_relu6(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck4_relu61_relu6(DepthwiseConv2d)
Transpose output shapes: mobilenetv21_features_linearbottleneck4_conv2_fwd(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck4_elemwise_add0(Eltwise)
Transpose output shapes: mobilenetv21_features_linearbottleneck5_relu60_relu6(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck5_relu61_relu6(DepthwiseConv2d)
Transpose output shapes: mobilenetv21_features_linearbottleneck5_conv2_fwd(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck5_elemwise_add0(Eltwise)
Transpose output shapes: mobilenetv21_features_linearbottleneck6_relu60_relu6(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck6_relu61_relu6(DepthwiseConv2d)
Transpose output shapes: mobilenetv21_features_linearbottleneck6_conv2_fwd(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck7_relu60_relu6(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck7_relu61_relu6(DepthwiseConv2d)
Transpose output shapes: mobilenetv21_features_linearbottleneck7_conv2_fwd(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck7_elemwise_add0(Eltwise)
Transpose output shapes: mobilenetv21_features_linearbottleneck8_relu60_relu6(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck8_relu61_relu6(DepthwiseConv2d)
Transpose output shapes: mobilenetv21_features_linearbottleneck8_conv2_fwd(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck8_elemwise_add0(Eltwise)
Transpose output shapes: mobilenetv21_features_linearbottleneck9_relu60_relu6(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck9_relu61_relu6(DepthwiseConv2d)
Transpose output shapes: mobilenetv21_features_linearbottleneck9_conv2_fwd(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck9_elemwise_add0(Eltwise)
Transpose output shapes: mobilenetv21_features_linearbottleneck10_relu60_relu6(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck10_relu61_relu6(DepthwiseConv2d)
Transpose output shapes: mobilenetv21_features_linearbottleneck10_conv2_fwd(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck11_relu60_relu6(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck11_relu61_relu6(DepthwiseConv2d)
Transpose output shapes: mobilenetv21_features_linearbottleneck11_conv2_fwd(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck11_elemwise_add0(Eltwise)
Transpose output shapes: mobilenetv21_features_linearbottleneck12_relu60_relu6(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck12_relu61_relu6(DepthwiseConv2d)
Transpose output shapes: mobilenetv21_features_linearbottleneck12_conv2_fwd(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck12_elemwise_add0(Eltwise)
Transpose output shapes: mobilenetv21_features_linearbottleneck13_relu60_relu6(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck13_relu61_relu6(DepthwiseConv2d)
Transpose output shapes: mobilenetv21_features_linearbottleneck13_conv2_fwd(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck14_relu60_relu6(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck14_relu61_relu6(DepthwiseConv2d)
Transpose output shapes: mobilenetv21_features_linearbottleneck14_conv2_fwd(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck14_elemwise_add0(Eltwise)
Transpose output shapes: mobilenetv21_features_linearbottleneck15_relu60_relu6(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck15_relu61_relu6(DepthwiseConv2d)
Transpose output shapes: mobilenetv21_features_linearbottleneck15_conv2_fwd(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck15_elemwise_add0(Eltwise)
Transpose output shapes: mobilenetv21_features_linearbottleneck16_relu60_relu6(Conv2D)
Transpose output shapes: mobilenetv21_features_linearbottleneck16_relu61_relu6(DepthwiseConv2d)
Transpose output shapes: mobilenetv21_features_linearbottleneck16_conv2_fwd(Conv2D)
Transpose output shapes: mobilenetv21_features_relu61_relu6(Conv2D)
Transpose output shapes: rf_c3_lateral_relu(Conv2D)
Transpose output shapes: rf_c3_det_conv1(Conv2D)
Transpose output shapes: rf_c3_det_context_conv1_relu(Conv2D)
Transpose output shapes: rf_c3_det_context_conv2(Conv2D)
Transpose output shapes: rf_c3_det_context_conv3_1_relu(Conv2D)
Transpose output shapes: rf_c3_det_context_conv3_2(Conv2D)
Transpose concat/split args: rf_c3_det_concat(Concat)
Transpose output shapes: rf_c3_det_concat(Concat)
Transpose output shapes: rf_c3_det_concat_relu(Activation)
Transpose output shapes: face_rpn_cls_score_stride32(Conv2D)
Transpose output shapes: face_rpn_cls_score_reshape_stride32(Reshape)
Transpose output shapes: face_rpn_cls_prob_stride32(Softmax)
Transpose output shapes: face_rpn_cls_prob_reshape_stride32(Reshape)
Transpose output shapes: face_rpn_bbox_pred_stride32(Conv2D)
Transpose output shapes: face_rpn_landmark_pred_stride32(Conv2D)
Transpose output shapes: rf_c2_lateral_relu(Conv2D)
Transpose output shapes: rf_c3_upsampling(DepthwiseDeconv2d)
Transpose crop args: crop0(Crop)
Transpose output shapes: crop0(Crop)
Transpose output shapes: _plus0(Eltwise)
Transpose output shapes: rf_c2_aggr_relu(Conv2D)
Transpose output shapes: rf_c2_det_conv1(Conv2D)
Transpose output shapes: rf_c2_det_context_conv1_relu(Conv2D)
Transpose output shapes: rf_c2_det_context_conv2(Conv2D)
Transpose output shapes: rf_c2_det_context_conv3_1_relu(Conv2D)
Transpose output shapes: rf_c2_det_context_conv3_2(Conv2D)
Transpose concat/split args: rf_c2_det_concat(Concat)
Transpose output shapes: rf_c2_det_concat(Concat)
Transpose output shapes: rf_c2_det_concat_relu(Activation)
Transpose output shapes: face_rpn_cls_score_stride16(Conv2D)
Transpose output shapes: face_rpn_cls_score_reshape_stride16(Reshape)
Transpose output shapes: face_rpn_cls_prob_stride16(Softmax)
Transpose output shapes: face_rpn_cls_prob_reshape_stride16(Reshape)
Transpose output shapes: face_rpn_bbox_pred_stride16(Conv2D)
Transpose output shapes: face_rpn_landmark_pred_stride16(Conv2D)
Transpose output shapes: rf_c1_red_conv_relu(Conv2D)
Transpose output shapes: rf_c2_upsampling(DepthwiseDeconv2d)
Transpose crop args: crop1(Crop)
Transpose output shapes: crop1(Crop)
Transpose output shapes: _plus1(Eltwise)
Transpose output shapes: rf_c1_aggr_relu(Conv2D)
Transpose output shapes: rf_c1_det_conv1(Conv2D)
Transpose output shapes: rf_c1_det_context_conv1_relu(Conv2D)
Transpose output shapes: rf_c1_det_context_conv2(Conv2D)
Transpose output shapes: rf_c1_det_context_conv3_1_relu(Conv2D)
Transpose output shapes: rf_c1_det_context_conv3_2(Conv2D)
Transpose concat/split args: rf_c1_det_concat(Concat)
Transpose output shapes: rf_c1_det_concat(Concat)
Transpose output shapes: rf_c1_det_concat_relu(Activation)
Transpose output shapes: face_rpn_cls_score_stride8(Conv2D)
Transpose output shapes: face_rpn_cls_score_reshape_stride8(Reshape)
Transpose output shapes: face_rpn_cls_prob_stride8(Softmax)
Transpose output shapes: face_rpn_cls_prob_reshape_stride8(Reshape)
Transpose output shapes: face_rpn_bbox_pred_stride8(Conv2D)
Transpose output shapes: face_rpn_landmark_pred_stride8(Conv2D)
*********************************************
          Model retinanet converted          
*********************************************

******************************************
          Building model library          
******************************************

* Build //mace/codegen:generated_models with ABI arm64-v8a
WARNING: --batch mode is deprecated. Please instead explicitly shut down your Bazel server using the command "bazel shutdown".
WARNING: The major revision of the Android NDK referenced by android_ndk_repository rule 'androidndk' is 18. The major revisions supported by Bazel are [10, 11, 12, 13, 14, 15, 16]. Bazel will attempt to treat the NDK as if it was r16. This may cause compilation and linkage problems. Please download a supported NDK version.
INFO: Analysed target //mace/codegen:generated_models (28 packages loaded).
INFO: Found 1 target...
Target //mace/codegen:generated_models up-to-date:
  bazel-bin/mace/codegen/libgenerated_models.a
INFO: Elapsed time: 130.568s, Critical Path: 35.58s
INFO: 185 processes: 185 local.
INFO: Build completed successfully, 186 total actions
('build', '//mace/codegen:generated_models', '--config', 'android', '--cpu=arm64-v8a', '--define', 'neon=true', '--define', 'openmp=false', '--define', 'opencl=true', '--define', 'quantize=false', '--define', 'hexagon=false', '--define', 'hta=false', '--define', 'apu=false', '--config', 'optimization', '--config', 'symbol_hidden')
Build done!

-----------------------------------------------------------------
                             Library                             
-----------------------------------------------------------------
|          key          |                 value                 |
=================================================================
|        MACE Model Path|               build/libretinanet/model|
-----------------------------------------------------------------
| MACE Model Header Path| build/libretinanet/include/mace/public|
-----------------------------------------------------------------

validation.log

# python tools/converter.py run --config=/models/retinanet/retinanet3.yml --validate
CMD> bazel build //mace/proto:mace_py
WARNING: --batch mode is deprecated. Please instead explicitly shut down your Bazel server using the command "bazel shutdown".
Loading:
Loading: 0 packages loaded
Analyzing: target //mace/proto:mace_py (6 packages loaded)
INFO: Analysed target //mace/proto:mace_py (17 packages loaded).
INFO: Found 1 target...
[1 / 7] [-----] BazelWorkspaceStatusAction stable-status.txt
Target //mace/proto:mace_py up-to-date:
bazel-genfiles/mace/proto/mace_pb2.py
INFO: Elapsed time: 2.361s, Critical Path: 0.07s
INFO: 0 processes.
INFO: Build completed successfully, 1 total action
INFO: Build completed successfully, 1 total action

CMD> cp -f bazel-genfiles/mace/proto/mace_pb2.py tools/python/py_proto

CMD> bazel build //third_party/caffe:caffe_py
WARNING: --batch mode is deprecated. Please instead explicitly shut down your Bazel server using the command "bazel shutdown".
Loading:
Loading: 0 packages loaded
Analyzing: target //third_party/caffe:caffe_py (5 packages loaded)
INFO: Analysed target //third_party/caffe:caffe_py (17 packages loaded).
INFO: Found 1 target...
[0 / 4] [-----] BazelWorkspaceStatusAction stable-status.txt
Target //third_party/caffe:caffe_py up-to-date:
bazel-genfiles/third_party/caffe/caffe_pb2.py
INFO: Elapsed time: 2.541s, Critical Path: 0.09s
INFO: 0 processes.
INFO: Build completed successfully, 1 total action
INFO: Build completed successfully, 1 total action

CMD> cp -f bazel-genfiles/third_party/caffe/caffe_pb2.py tools/python/py_proto

* Build //mace/tools:mace_run_static with ABI arm64-v8a
('build', '//mace/tools:mace_run_static', '--config', 'android', '--cpu=arm64-v8a', '--define', 'neon=true', '--define', 'openmp=false', '--define', 'opencl=true', '--define', 'quantize=false', '--define', 'hexagon=false', '--define', 'hta=false', '--define', 'apu=false', '--config', 'optimization', '--config', 'symbol_hidden', '--per_file_copt=mace/tools/mace_run.cc@-DMODEL_GRAPH_FORMAT_CODE')
WARNING: --batch mode is deprecated. Please instead explicitly shut down your Bazel server using the command "bazel shutdown".
WARNING: The major revision of the Android NDK referenced by android_ndk_repository rule 'androidndk' is 18. The major revisions supported by Bazel are [10, 11, 12, 13, 14, 15, 16]. Bazel will attempt to treat the NDK as if it was r16. This may cause compilation and linkage problems. Please download a supported NDK version.
INFO: Analysed target //mace/tools:mace_run_static (32 packages loaded).
INFO: Found 1 target...
Target //mace/tools:mace_run_static up-to-date:
  bazel-bin/mace/tools/mace_run_static
INFO: Elapsed time: 18.539s, Critical Path: 2.45s
INFO: 0 processes.
INFO: Build completed successfully, 1 total action
('build', '//mace/tools:mace_run_static', '--config', 'android', '--cpu=arm64-v8a', '--define', 'neon=true', '--define', 'openmp=false', '--define', 'opencl=true', '--define', 'quantize=false', '--define', 'hexagon=false', '--define', 'hta=false', '--define', 'apu=false', '--config', 'optimization', '--config', 'symbol_hidden', '--per_file_copt=mace/tools/mace_run.cc@-DMODEL_GRAPH_FORMAT_CODE')
Build done!

**********************************************
          Run model retinanet on MI9          
**********************************************

Generate input file:  build/libretinanet/_tmp/retinanet/14a062bf9f488e2a38c1fb60b18a80de/MI9_msmnile/arm64-v8a/model_input_data
Generate input file done.
* Run 'retinanet' with round=1, restart_round=1, tuning=False, out_of_range_check=False, omp_num_threads=(-1,), cpu_affinity_policy=(1,), gpu_perf_hint=(3,), gpu_priority_hint=(3,)
Push build/libretinanet/_tmp/retinanet/14a062bf9f488e2a38c1fb60b18a80de/MI9_msmnile/arm64-v8a/model_input_data to /data/local/tmp/mace_run
Push build/libretinanet/_tmp/arm64-v8a/mace_run_static to /data/local/tmp/mace_run
Push /tmp/cmd_file-retinanet-1581388386.7015445 to /data/local/tmp/mace_run/cmd_file-retinanet-1581388386.7015445
I mace/tools/mace_run.cc:527] model name: retinanet
I mace/tools/mace_run.cc:528] mace version: v0.12.0-0-ga610d50
I mace/tools/mace_run.cc:529] input node: data
I mace/tools/mace_run.cc:530] input shape: 1,3,320,320
I mace/tools/mace_run.cc:531] output node: face_rpn_cls_prob_reshape_stride32,face_rpn_bbox_pred_stride32,face_rpn_landmark_pred_stride32,face_rpn_cls_prob_reshape_stride16,face_rpn_bbox_pred_stride16,face_rpn_landmark_pred_stride16,face_rpn_cls_prob_reshape_stride8,face_rpn_bbox_pred_stride8,face_rpn_landmark_pred_stride8
I mace/tools/mace_run.cc:532] output shape: 1,4,10,10:1,8,10,10:1,20,10,10:1,4,20,20:1,8,20,20:1,20,20,20:1,4,40,40:1,8,40,40:1,20,40,40
I mace/tools/mace_run.cc:533] input_file: /data/local/tmp/mace_run/model_input
I mace/tools/mace_run.cc:534] output_file: /data/local/tmp/mace_run/model_out
I mace/tools/mace_run.cc:535] input dir:
I mace/tools/mace_run.cc:536] output dir:
I mace/tools/mace_run.cc:537] model_data_file:
I mace/tools/mace_run.cc:538] model_file:
I mace/tools/mace_run.cc:539] device: GPU
I mace/tools/mace_run.cc:540] round: 1
I mace/tools/mace_run.cc:541] restart_round: 1
I mace/tools/mace_run.cc:542] gpu_perf_hint: 3
I mace/tools/mace_run.cc:543] gpu_priority_hint: 3
I mace/tools/mace_run.cc:544] omp_num_threads: -1
I mace/tools/mace_run.cc:545] cpu_affinity_policy: 1
I mace/tools/mace_run.cc:548] limit_opencl_kernel_time: 0
I mace/tools/mace_run.cc:553] opencl_queue_window_size: 0
I mace/libmace/mace.cc:464] Creating MaceEngine, MACE version: v0.12.0-0-ga610d50
I mace/libmace/mace.cc:503] Initializing MaceEngine
I mace/libmace/mace.cc:636] Destroying MaceEngine
I mace/tools/mace_run.cc:596] restart round 0
W ./mace/utils/tuner.h:201] Failed to read tuned param file: /data/local/tmp/mace_run/libretinanet_tuned_opencl_parameter.MI9.msmnile.bin
I mace/libmace/mace.cc:464] Creating MaceEngine, MACE version: v0.12.0-0-ga610d50
W mace/core/kv_storage.cc:109] Failed to read kv store file: /data/local/tmp/mace_run/interior//mace_cl_compiled_program.bin
W mace/core/runtime/opencl/opencl_runtime.cc:382] Load OpenCL cached compiled kernel file failed. Please make sure the storage directory exist and you have Write&Read permission
I mace/libmace/mace.cc:503] Initializing MaceEngine
I mace/tools/mace_run.cc:269] Create Mace Engine latency: 877.206 ms
I mace/tools/mace_run.cc:276] Total init latency: 877.331 ms
I mace/tools/mace_run.cc:370] Warm up run
I mace/tools/mace_run.cc:406] 1st warm up run latency: 1323.14 ms
I mace/tools/mace_run.cc:414] Run model
I mace/tools/mace_run.cc:476] Average latency: 16.371 ms
I mace/tools/mace_run.cc:491] Write output file /data/local/tmp/mace_run/model_out_face_rpn_cls_prob_reshape_stride32 with size 1600 done.
I mace/tools/mace_run.cc:491] Write output file /data/local/tmp/mace_run/model_out_face_rpn_bbox_pred_stride32 with size 3200 done.
I mace/tools/mace_run.cc:491] Write output file /data/local/tmp/mace_run/model_out_face_rpn_landmark_pred_stride32 with size 8000 done.
I mace/tools/mace_run.cc:491] Write output file /data/local/tmp/mace_run/model_out_face_rpn_cls_prob_reshape_stride16 with size 6400 done.
I mace/tools/mace_run.cc:491] Write output file /data/local/tmp/mace_run/model_out_face_rpn_bbox_pred_stride16 with size 12800 done.
I mace/tools/mace_run.cc:491] Write output file /data/local/tmp/mace_run/model_out_face_rpn_landmark_pred_stride16 with size 32000 done.
I mace/tools/mace_run.cc:491] Write output file /data/local/tmp/mace_run/model_out_face_rpn_cls_prob_reshape_stride8 with size 25600 done.
I mace/tools/mace_run.cc:491] Write output file /data/local/tmp/mace_run/model_out_face_rpn_bbox_pred_stride8 with size 51200 done.
I mace/tools/mace_run.cc:491] Write output file /data/local/tmp/mace_run/model_out_face_rpn_landmark_pred_stride8 with size 128000 done.
========================================================
     capability(CPU)        init      warmup     run_avg
========================================================
time           7.484     877.331    1323.142      16.371
I mace/libmace/mace.cc:636] Destroying MaceEngine
Running finished!

* Validate with caffe
Pull /data/local/tmp/mace_run/model_out_face_rpn_cls_prob_reshape_stride32 to build/libretinanet/_tmp/retinanet/14a062bf9f488e2a38c1fb60b18a80de/MI9_msmnile/arm64-v8a
Pull /data/local/tmp/mace_run/model_out_face_rpn_bbox_pred_stride32 to build/libretinanet/_tmp/retinanet/14a062bf9f488e2a38c1fb60b18a80de/MI9_msmnile/arm64-v8a
Pull /data/local/tmp/mace_run/model_out_face_rpn_landmark_pred_stride32 to build/libretinanet/_tmp/retinanet/14a062bf9f488e2a38c1fb60b18a80de/MI9_msmnile/arm64-v8a
Pull /data/local/tmp/mace_run/model_out_face_rpn_cls_prob_reshape_stride16 to build/libretinanet/_tmp/retinanet/14a062bf9f488e2a38c1fb60b18a80de/MI9_msmnile/arm64-v8a
Pull /data/local/tmp/mace_run/model_out_face_rpn_bbox_pred_stride16 to build/libretinanet/_tmp/retinanet/14a062bf9f488e2a38c1fb60b18a80de/MI9_msmnile/arm64-v8a
Pull /data/local/tmp/mace_run/model_out_face_rpn_landmark_pred_stride16 to build/libretinanet/_tmp/retinanet/14a062bf9f488e2a38c1fb60b18a80de/MI9_msmnile/arm64-v8a
Pull /data/local/tmp/mace_run/model_out_face_rpn_cls_prob_reshape_stride8 to build/libretinanet/_tmp/retinanet/14a062bf9f488e2a38c1fb60b18a80de/MI9_msmnile/arm64-v8a
Pull /data/local/tmp/mace_run/model_out_face_rpn_bbox_pred_stride8 to build/libretinanet/_tmp/retinanet/14a062bf9f488e2a38c1fb60b18a80de/MI9_msmnile/arm64-v8a
Pull /data/local/tmp/mace_run/model_out_face_rpn_landmark_pred_stride8 to build/libretinanet/_tmp/retinanet/14a062bf9f488e2a38c1fb60b18a80de/MI9_msmnile/arm64-v8a
Traceback (most recent call last):
  File "/mace/validate.py", line 459, in <module>
face_rpn_cls_prob_reshape_stride32 MACE VS CAFFE similarity: 0.7051022200187764 , sqnr: 1.988659806883121 , pixel_accuracy: 0.4
    FLAGS.log_file)
  File "/mace/validate.py", line 371, in validate
    validation_threshold, log_file)
  File "/mace/validate.py", line 262, in validate_caffe_model
    value, validation_threshold, log_file)
  File "/mace/validate.py", line 113, in compare_output
    "", common.StringFormatter.block("Similarity Test Failed"))
TypeError: summary() takes exactly 1 argument (2 given)
Traceback (most recent call last):
  File "tools/converter.py", line 1151, in <module>
    flags.func(flags)
  File "tools/converter.py", line 938, in run_mace
    device.run_specify_abi(flags, configs, target_abi)
  File "/mace/tools/device.py", line 782, in run_specify_abi
    log_file=log_file,
  File "/mace/tools/sh_commands.py", line 756, in validate_model
    _fg=True)
  File "/root/.pyenv/versions/3.6.3/lib/python3.6/site-packages/sh.py", line 1413, in __call__
    raise exc
sh.ErrorReturnCode_1: 

  RAN: /usr/bin/docker exec mace_caffe_lastest_validator python -u /mace/validate.py --platform=caffe --model_file=/mace/retinanet3.prototxt --weight_file=/mace/retinanet3.caffemodel --input_file=/mace/model_input --mace_out_file=/mace/model_out --device_type=GPU --input_node=data --output_node=face_rpn_cls_prob_reshape_stride32,face_rpn_bbox_pred_stride32,face_rpn_landmark_pred_stride32,face_rpn_cls_prob_reshape_stride16,face_rpn_bbox_pred_stride16,face_rpn_landmark_pred_stride16,face_rpn_cls_prob_reshape_stride8,face_rpn_bbox_pred_stride8,face_rpn_landmark_pred_stride8 --input_shape=1,3,320,320 --output_shape=1,4,10,10:1,8,10,10:1,20,10,10:1,4,20,20:1,8,20,20:1,20,20,20:1,4,40,40:1,8,40,40:1,20,40,40 --input_data_format=NCHW --output_data_format=NCHW,NCHW,NCHW,NCHW,NCHW,NCHW,NCHW,NCHW,NCHW --validation_threshold=0.995000 --input_data_type=float32 --backend=tensorflow --validation_outputs_data= --log_file=

  STDOUT:


  STDERR: