2020-06-02 15:59:14.452430: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.0 2020-06-02 15:59:19.807146: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libnvinfer.so.6 2020-06-02 15:59:19.825787: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libnvinfer_plugin.so.6 Converting to TF-TRT FP16... 2020-06-02 15:59:25.487306: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libnvinfer.so.6 2020-06-02 15:59:53.418308: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcuda.so.1 2020-06-02 15:59:53.458410: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:948] ARM64 does not support NUMA - returning NUMA node zero 2020-06-02 15:59:53.458914: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1555] Found device 0 with properties: pciBusID: 0000:00:00.0 name: NVIDIA Tegra X2 computeCapability: 6.2 coreClock: 1.3GHz coreCount: 2 deviceMemorySize: 7.68GiB deviceMemoryBandwidth: 38.74GiB/s 2020-06-02 15:59:53.459025: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.0 2020-06-02 15:59:53.459143: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10.0 2020-06-02 15:59:53.490077: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcufft.so.10.0 2020-06-02 15:59:53.529126: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcurand.so.10.0 2020-06-02 15:59:53.572489: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusolver.so.10.0 2020-06-02 15:59:53.599328: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusparse.so.10.0 2020-06-02 15:59:53.599633: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7 2020-06-02 15:59:53.600013: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:948] ARM64 does not support NUMA - returning NUMA node zero 2020-06-02 15:59:53.600532: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:948] ARM64 does not support NUMA - returning NUMA node zero 2020-06-02 15:59:53.600761: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1697] Adding visible gpu devices: 0 2020-06-02 15:59:53.636548: W tensorflow/core/platform/profile_utils/cpu_utils.cc:98] Failed to find bogomips in /proc/cpuinfo; cannot determine CPU frequency 2020-06-02 15:59:53.638150: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x5248a90 initialized for platform Host (this does not guarantee that XLA will be used). Devices: 2020-06-02 15:59:53.638309: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Host, Default Version 2020-06-02 15:59:53.748662: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:948] ARM64 does not support NUMA - returning NUMA node zero 2020-06-02 15:59:53.749283: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x51aee10 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices: 2020-06-02 15:59:53.749363: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): NVIDIA Tegra X2, Compute Capability 6.2 2020-06-02 15:59:53.751956: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:948] ARM64 does not support NUMA - returning NUMA node zero 2020-06-02 15:59:53.752353: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1555] Found device 0 with properties: pciBusID: 0000:00:00.0 name: NVIDIA Tegra X2 computeCapability: 6.2 coreClock: 1.3GHz coreCount: 2 deviceMemorySize: 7.68GiB deviceMemoryBandwidth: 38.74GiB/s 2020-06-02 15:59:53.752464: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.0 2020-06-02 15:59:53.752535: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10.0 2020-06-02 15:59:53.752642: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcufft.so.10.0 2020-06-02 15:59:53.752719: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcurand.so.10.0 2020-06-02 15:59:53.752793: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusolver.so.10.0 2020-06-02 15:59:53.752939: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusparse.so.10.0 2020-06-02 15:59:53.753002: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7 2020-06-02 15:59:53.753407: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:948] ARM64 does not support NUMA - returning NUMA node zero 2020-06-02 15:59:53.753870: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:948] ARM64 does not support NUMA - returning NUMA node zero 2020-06-02 15:59:53.754220: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1697] Adding visible gpu devices: 0 2020-06-02 15:59:53.754487: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.0 2020-06-02 15:59:59.398571: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1096] Device interconnect StreamExecutor with strength 1 edge matrix: 2020-06-02 15:59:59.398716: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1102] 0 2020-06-02 15:59:59.398769: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] 0: N 2020-06-02 15:59:59.399541: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:948] ARM64 does not support NUMA - returning NUMA node zero 2020-06-02 15:59:59.400059: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:948] ARM64 does not support NUMA - returning NUMA node zero 2020-06-02 15:59:59.400525: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1241] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 2182 MB memory) -> physical GPU (device: 0, name: NVIDIA Tegra X2, pci bus id: 0000:00:00.0, compute capability: 6.2) 2020-06-02 16:02:20.031931: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:948] ARM64 does not support NUMA - returning NUMA node zero 2020-06-02 16:02:20.032268: I tensorflow/core/grappler/devices.cc:55] Number of eligible GPUs (core count >= 8, compute capability >= 0.0): 0 2020-06-02 16:02:20.032530: I tensorflow/core/grappler/clusters/single_machine.cc:356] Starting new session 2020-06-02 16:02:20.034472: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:948] ARM64 does not support NUMA - returning NUMA node zero 2020-06-02 16:02:20.034806: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1555] Found device 0 with properties: pciBusID: 0000:00:00.0 name: NVIDIA Tegra X2 computeCapability: 6.2 coreClock: 1.3GHz coreCount: 2 deviceMemorySize: 7.68GiB deviceMemoryBandwidth: 38.74GiB/s 2020-06-02 16:02:20.034908: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.0 2020-06-02 16:02:20.035007: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10.0 2020-06-02 16:02:20.035093: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcufft.so.10.0 2020-06-02 16:02:20.035182: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcurand.so.10.0 2020-06-02 16:02:20.035254: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusolver.so.10.0 2020-06-02 16:02:20.035325: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusparse.so.10.0 2020-06-02 16:02:20.035404: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7 2020-06-02 16:02:20.035682: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:948] ARM64 does not support NUMA - returning NUMA node zero 2020-06-02 16:02:20.036650: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:948] ARM64 does not support NUMA - returning NUMA node zero 2020-06-02 16:02:20.036886: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1697] Adding visible gpu devices: 0 2020-06-02 16:02:20.036995: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1096] Device interconnect StreamExecutor with strength 1 edge matrix: 2020-06-02 16:02:20.037046: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1102] 0 2020-06-02 16:02:20.037091: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] 0: N 2020-06-02 16:02:20.037471: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:948] ARM64 does not support NUMA - returning NUMA node zero 2020-06-02 16:02:20.037935: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:948] ARM64 does not support NUMA - returning NUMA node zero 2020-06-02 16:02:20.038207: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1241] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 2182 MB memory) -> physical GPU (device: 0, name: NVIDIA Tegra X2, pci bus id: 0000:00:00.0, compute capability: 6.2) 2020-06-02 16:02:21.124660: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:841] Optimization results for grappler item: graph_to_optimize 2020-06-02 16:02:21.124824: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:843] function_optimizer: Graph size after: 1254 nodes (931), 2553 edges (2230), time = 130.154ms. 2020-06-02 16:02:21.124882: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:843] function_optimizer: function_optimizer did nothing. time = 2.75ms. 2020-06-02 16:02:40.259354: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:948] ARM64 does not support NUMA - returning NUMA node zero 2020-06-02 16:02:40.260093: I tensorflow/core/grappler/devices.cc:55] Number of eligible GPUs (core count >= 8, compute capability >= 0.0): 0 2020-06-02 16:02:40.260758: I tensorflow/core/grappler/clusters/single_machine.cc:356] Starting new session 2020-06-02 16:02:40.263882: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:948] ARM64 does not support NUMA - returning NUMA node zero 2020-06-02 16:02:40.264653: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1555] Found device 0 with properties: pciBusID: 0000:00:00.0 name: NVIDIA Tegra X2 computeCapability: 6.2 coreClock: 1.3GHz coreCount: 2 deviceMemorySize: 7.68GiB deviceMemoryBandwidth: 38.74GiB/s 2020-06-02 16:02:40.274228: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.0 2020-06-02 16:02:40.274420: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10.0 2020-06-02 16:02:40.274683: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcufft.so.10.0 2020-06-02 16:02:40.274847: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcurand.so.10.0 2020-06-02 16:02:40.274965: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusolver.so.10.0 2020-06-02 16:02:40.275082: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusparse.so.10.0 2020-06-02 16:02:40.275183: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7 2020-06-02 16:02:40.275525: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:948] ARM64 does not support NUMA - returning NUMA node zero 2020-06-02 16:02:40.276041: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:948] ARM64 does not support NUMA - returning NUMA node zero 2020-06-02 16:02:40.276300: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1697] Adding visible gpu devices: 0 2020-06-02 16:02:40.276726: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1096] Device interconnect StreamExecutor with strength 1 edge matrix: 2020-06-02 16:02:40.276796: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1102] 0 2020-06-02 16:02:40.276855: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] 0: N 2020-06-02 16:02:40.277449: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:948] ARM64 does not support NUMA - returning NUMA node zero 2020-06-02 16:02:40.278126: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:948] ARM64 does not support NUMA - returning NUMA node zero 2020-06-02 16:02:40.278487: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1241] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 2182 MB memory) -> physical GPU (device: 0, name: NVIDIA Tegra X2, pci bus id: 0000:00:00.0, compute capability: 6.2) 2020-06-02 16:02:55.759777: I tensorflow/compiler/tf2tensorrt/segment/segment.cc:460] There are 6 ops of 3 different types in the graph that are not converted to TensorRT: Identity, NoOp, Placeholder, (For more information see https://docs.nvidia.com/deeplearning/frameworks/tf-trt-user-guide/index.html#supported-ops). 2020-06-02 16:03:05.349833: I tensorflow/compiler/tf2tensorrt/convert/convert_graph.cc:636] Number of TensorRT candidate segments: 1 2020-06-02 16:03:13.534169: I tensorflow/compiler/tf2tensorrt/convert/convert_graph.cc:737] Replaced segment 0 consisting of 507 nodes by TRTEngineOp_0. 2020-06-02 16:03:15.704845: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:841] Optimization results for grappler item: tf_graph 2020-06-02 16:03:15.705159: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:843] constant_folding: Graph size after: 881 nodes (-320), 1860 edges (-640), time = 9185.21289ms. 2020-06-02 16:03:15.705254: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:843] layout: Graph size after: 885 nodes (4), 1864 edges (4), time = 722.016ms. 2020-06-02 16:03:15.705305: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:843] constant_folding: Graph size after: 883 nodes (-2), 1862 edges (-2), time = 573.888ms. 2020-06-02 16:03:15.705352: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:843] TensorRTOptimizer: Graph size after: 377 nodes (-506), 430 edges (-1432), time = 20652.3242ms. 2020-06-02 16:03:15.705397: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:843] constant_folding: Graph size after: 377 nodes (0), 430 edges (0), time = 330.734ms. 2020-06-02 16:03:15.705443: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:841] Optimization results for grappler item: TRTEngineOp_0_native_segment 2020-06-02 16:03:15.705516: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:843] constant_folding: Graph size after: 509 nodes (0), 792 edges (0), time = 261.479ms. 2020-06-02 16:03:15.705563: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:843] layout: Graph size after: 509 nodes (0), 792 edges (0), time = 242.867ms. 2020-06-02 16:03:15.705608: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:843] constant_folding: Graph size after: 509 nodes (0), 792 edges (0), time = 194.613ms. 2020-06-02 16:03:15.705652: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:843] TensorRTOptimizer: Graph size after: 509 nodes (0), 792 edges (0), time = 23.941ms. 2020-06-02 16:03:15.705695: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:843] constant_folding: Graph size after: 509 nodes (0), 792 edges (0), time = 195.56ms. 2020-06-02 17:56:16.885674: W tensorflow/core/framework/op_kernel.cc:1655] OP_REQUIRES failed at trt_engine_resource_ops.cc:183 : Not found: Container TF-TRT does not exist. (Could not find resource: TF-TRT/TRTEngineOp_0) WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/ops/resource_variable_ops.py:1786: calling BaseResourceVariable.__init__ (from tensorflow.python.ops.resource_variable_ops) with constraint is deprecated and will be removed in a future version. Instructions for updating: If using Keras pass *_constraint arguments to layers. Done Converting to TF-TRT FP16