Last active
November 11, 2020 17:28
-
-
Save ibmua/7e6c3cd365eefc64fb562f2ae354de31 to your computer and use it in GitHub Desktop.
3080 ai-benchmark.com [run under NGC 20.10-tf1]
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
First goes 3080 Gigabyte Gaming OC, then 3080 MSI Gaming X Trio | |
>> AI-Benchmark-v.0.1.2 | |
>> Let the AI Games begin.. | |
2020-11-11 17:04:23.002305: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 2599990000 Hz | |
2020-11-11 17:04:23.003399: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x5b4a090 initialized for platform Host (this does not guarantee that XLA will be used). Devices: | |
2020-11-11 17:04:23.003415: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Host, Default Version | |
2020-11-11 17:04:23.005150: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcuda.so.1 | |
2020-11-11 17:04:23.108471: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x5b4c500 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices: | |
2020-11-11 17:04:23.108495: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): GeForce RTX 3080, Compute Capability 8.6 | |
2020-11-11 17:04:23.109148: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:04:23.109175: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:04:23.111404: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:04:23.112215: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:04:23.112437: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:04:23.114817: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:04:23.115382: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:04:23.115493: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:04:23.116567: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:04:23.116596: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:04:23.377793: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:04:23.377830: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:04:23.377835: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:04:23.379053: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
2020-11-11 17:04:23.380394: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:04:23.380422: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:04:23.380440: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:04:23.380448: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:04:23.380456: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:04:23.380466: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:04:23.380475: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:04:23.380484: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:04:23.381485: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:04:23.381502: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:04:23.381508: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:04:23.381513: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:04:23.382536: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
2020-11-11 17:04:23.383258: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:04:23.383278: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:04:23.383296: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:04:23.383305: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:04:23.383315: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:04:23.383325: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:04:23.383334: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:04:23.383345: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:04:23.384334: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:04:23.384348: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:04:23.384353: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:04:23.384358: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:04:23.385369: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
* TF Version: 1.15.4 | |
* Platform: Linux-5.4.0-52-generic-x86_64-with-Ubuntu-18.04-bionic | |
* CPU: N/A | |
* CPU RAM: 63 GB | |
* GPU/0: GeForce RTX 3080 | |
* GPU RAM: 8.0 GB | |
* CUDA Version: 11.1 | |
* CUDA Build: V11.1.74 | |
The benchmark is running... | |
The tests might take up to 20 minutes | |
Please don't interrupt the script | |
1/19. MobileNet-V2 | |
2020-11-11 17:04:30.372403: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:04:30.372431: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:04:30.372452: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:04:30.372462: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:04:30.372472: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:04:30.372483: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:04:30.372495: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:04:30.372506: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:04:30.373540: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:04:30.374516: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:04:30.374537: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:04:30.374552: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:04:30.374562: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:04:30.374570: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:04:30.374578: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:04:30.374587: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:04:30.374597: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:04:30.375622: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:04:30.375648: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:04:30.375655: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:04:30.375661: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:04:30.376711: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
2020-11-11 17:04:31.686660: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:04:33.047448: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
1.1 - inference | batch=50, size=224x224: 63.0 ± 144.9 ms | |
1.2 - training | batch=50, size=224x224: 240 ± 248 ms | |
2/19. Inception-V3 | |
2020-11-11 17:04:56.395729: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:04:56.395766: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:04:56.395790: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:04:56.395805: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:04:56.395816: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:04:56.395825: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:04:56.395834: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:04:56.395843: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:04:56.396605: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:04:56.396633: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:04:56.396641: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:04:56.396648: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:04:56.397427: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
2.1 - inference | batch=20, size=346x346: 48.5 ± 5.4 ms | |
2.2 - training | batch=20, size=346x346: 131 ± 2 ms | |
3/19. Inception-V4 | |
2020-11-11 17:05:12.093305: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:05:12.093339: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:05:12.093359: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:05:12.093367: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:05:12.093376: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:05:12.093385: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:05:12.093393: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:05:12.093406: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:05:12.094175: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:05:12.094201: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:05:12.094210: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:05:12.094216: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:05:12.094994: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
3.1 - inference | batch=10, size=346x346: 38.2 ± 0.5 ms | |
3.2 - training | batch=10, size=346x346: 162.6 ± 0.5 ms | |
4/19. Inception-ResNet-V2 | |
2020-11-11 17:05:30.995238: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:05:30.995274: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:05:30.995294: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:05:30.995302: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:05:30.995311: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:05:30.995319: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:05:30.995327: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:05:30.995344: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:05:30.996114: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:05:30.996139: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:05:30.996145: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:05:30.996153: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:05:30.996935: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
4.1 - inference | batch=10, size=346x346: 52.7 ± 0.9 ms | |
4.2 - training | batch=8, size=346x346: 173.7 ± 1.0 ms | |
5/19. ResNet-V2-50 | |
2020-11-11 17:05:59.566744: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:05:59.566783: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:05:59.566809: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:05:59.566820: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:05:59.566830: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:05:59.566840: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:05:59.566851: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:05:59.566861: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:05:59.567614: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:05:59.567639: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:05:59.567645: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:05:59.567651: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:05:59.568424: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
5.1 - inference | batch=10, size=346x346: 28.3 ± 0.9 ms | |
5.2 - training | batch=10, size=346x346: 81.9 ± 0.3 ms | |
6/19. ResNet-V2-152 | |
2020-11-11 17:06:09.858758: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:06:09.858796: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:06:09.858815: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:06:09.858824: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:06:09.858834: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:06:09.858843: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:06:09.858852: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:06:09.858867: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:06:09.859616: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:06:09.859642: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:06:09.859648: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:06:09.859654: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:06:09.860423: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
6.1 - inference | batch=10, size=256x256: 36.8 ± 0.4 ms | |
6.2 - training | batch=10, size=256x256: 147.2 ± 0.7 ms | |
7/19. VGG-16 | |
2020-11-11 17:06:32.150505: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:06:32.150543: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:06:32.150566: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:06:32.150574: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:06:32.150583: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:06:32.150592: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:06:32.150600: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:06:32.150609: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:06:32.151356: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:06:32.151381: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:06:32.151388: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:06:32.151393: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:06:32.152162: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
7.1 - inference | batch=20, size=224x224: 48.0 ± 0.0 ms | |
7.2 - training | batch=2, size=224x224: 64.8 ± 0.5 ms | |
8/19. SRCNN 9-5-5 | |
2020-11-11 17:06:41.515397: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:06:41.515432: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:06:41.515457: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:06:41.515467: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:06:41.515477: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:06:41.515488: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:06:41.515504: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:06:41.515515: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:06:41.516264: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:06:41.516284: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:06:41.516290: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:06:41.516295: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:06:41.517056: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
8.1 - inference | batch=10, size=512x512: 58.9 ± 3.6 ms | |
8.2 - inference | batch=1, size=1536x1536: 58.5 ± 0.5 ms | |
8.3 - training | batch=10, size=512x512: 141 ± 5 ms | |
9/19. VGG-19 Super-Res | |
2020-11-11 17:07:04.397808: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:07:04.397842: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:07:04.397862: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:07:04.397870: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:07:04.397878: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:07:04.397885: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:07:04.397893: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:07:04.397908: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:07:04.398725: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:07:04.398744: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:07:04.398750: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:07:04.398756: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:07:04.399517: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
9.1 - inference | batch=10, size=256x256: 60.9 ± 0.9 ms | |
9.2 - inference | batch=1, size=1024x1024: 104.0 ± 0.5 ms | |
9.3 - training | batch=10, size=224x224: 176.5 ± 0.6 ms | |
10/19. ResNet-SRGAN | |
2020-11-11 17:07:27.008873: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:07:27.008910: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:07:27.008933: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:07:27.008943: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:07:27.008954: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:07:27.008964: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:07:27.008974: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:07:27.008998: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:07:27.009745: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:07:27.009767: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:07:27.009773: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:07:27.009779: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:07:27.010548: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
10.1 - inference | batch=10, size=512x512: 69.6 ± 1.8 ms | |
10.2 - inference | batch=1, size=1536x1536: 63.3 ± 0.5 ms | |
10.3 - training | batch=5, size=512x512: 103.0 ± 0.3 ms | |
11/19. ResNet-DPED | |
2020-11-11 17:07:45.780247: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:07:45.780281: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:07:45.780305: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:07:45.780316: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:07:45.780326: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:07:45.780336: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:07:45.780346: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:07:45.780357: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:07:45.781111: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:07:45.781132: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:07:45.781139: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:07:45.781144: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:07:45.781917: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
11.1 - inference | batch=10, size=256x256: 67.2 ± 0.7 ms | |
11.2 - inference | batch=1, size=1024x1024: 124.8 ± 0.8 ms | |
11.3 - training | batch=15, size=128x128: 105.6 ± 0.5 ms | |
12/19. U-Net | |
2020-11-11 17:08:12.942391: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:08:12.942423: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:08:12.942442: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:08:12.942449: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:08:12.942457: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:08:12.942467: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:08:12.942476: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:08:12.942493: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:08:12.943237: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:08:12.943258: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:08:12.943265: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:08:12.943270: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:08:12.944092: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
12.1 - inference | batch=4, size=512x512: 118.8 ± 0.9 ms | |
12.2 - inference | batch=1, size=1024x1024: 143.2 ± 0.7 ms | |
12.3 - training | batch=4, size=256x256: 130.1 ± 0.3 ms | |
13/19. Nvidia-SPADE | |
2020-11-11 17:08:34.354428: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:08:34.354464: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:08:34.354484: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:08:34.354494: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:08:34.354503: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:08:34.354512: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:08:34.354521: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:08:34.354531: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:08:34.355285: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:08:34.355307: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:08:34.355313: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:08:34.355319: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:08:34.356082: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
13.1 - inference | batch=5, size=128x128: 51.8 ± 0.4 ms | |
13.2 - training | batch=1, size=128x128: 88.6 ± 0.7 ms | |
14/19. ICNet | |
2020-11-11 17:08:50.175883: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:08:50.175918: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:08:50.175939: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:08:50.175948: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:08:50.175957: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:08:50.175966: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:08:50.175974: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:08:50.175984: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:08:50.176737: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:08:50.176763: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:08:50.176769: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:08:50.176774: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:08:50.177539: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
14.1 - inference | batch=5, size=1024x1536: 146.6 ± 0.7 ms | |
14.2 - training | batch=10, size=1024x1536: 450.4 ± 0.5 ms | |
15/19. PSPNet | |
2020-11-11 17:09:19.453292: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:09:19.453328: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:09:19.453348: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:09:19.453357: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:09:19.453366: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:09:19.453374: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:09:19.453383: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:09:19.453392: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:09:19.454144: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:09:19.454166: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:09:19.454172: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:09:19.454178: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:09:19.454948: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
15.1 - inference | batch=5, size=720x720: 204.0 ± 0.7 ms | |
15.2 - training | batch=1, size=512x512: 93.3 ± 0.6 ms | |
16/19. DeepLab | |
2020-11-11 17:09:36.270615: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:09:36.270657: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:09:36.270684: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:09:36.270696: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:09:36.270708: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:09:36.270719: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:09:36.270730: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:09:36.270741: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:09:36.271504: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:09:36.271529: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:09:36.271536: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:09:36.271543: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:09:36.272319: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
16.1 - inference | batch=2, size=512x512: 68.7 ± 0.5 ms | |
16.2 - training | batch=1, size=384x384: 81.4 ± 0.5 ms | |
17/19. Pixel-RNN | |
2020-11-11 17:09:52.425709: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:09:52.425745: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:09:52.425768: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:09:52.425778: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:09:52.425789: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:09:52.425799: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:09:52.425810: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:09:52.425827: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:09:52.426604: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:09:52.426631: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:09:52.426638: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:09:52.426646: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:09:52.427425: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
17.1 - inference | batch=50, size=64x64: 305 ± 8 ms | |
17.2 - training | batch=10, size=64x64: 1015 ± 12 ms | |
18/19. LSTM-Sentiment | |
2020-11-11 17:11:25.841494: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:11:25.841536: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:11:25.841561: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:11:25.841571: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:11:25.841589: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:11:25.841605: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:11:25.841613: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:11:25.841622: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:11:25.844645: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:11:25.844680: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:11:25.844686: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:11:25.844691: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:11:25.845501: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
18.1 - inference | batch=100, size=1024x300: 330 ± 1 ms | |
18.2 - training | batch=10, size=1024x300: 430 ± 12 ms | |
19/19. GNMT-Translation | |
2020-11-11 17:11:48.564397: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.815 | |
pciBusID: 0000:19:00.0 | |
2020-11-11 17:11:48.564432: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:11:48.564455: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:11:48.564464: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:11:48.564475: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:11:48.564486: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:11:48.564501: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:11:48.564512: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:11:48.565257: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:11:48.565275: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:11:48.565281: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:11:48.565286: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:11:48.566055: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8145 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:19:00.0, compute capability: 8.6) | |
19.1 - inference | batch=1, size=1x20: 96.2 ± 0.9 ms | |
Device Inference Score: 17645 | |
Device Training Score: 18980 | |
Device AI Score: 36625 | |
>> AI-Benchmark-v.0.1.2 | |
>> Let the AI Games begin.. | |
2020-11-11 17:13:19.658308: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 2599990000 Hz | |
2020-11-11 17:13:19.659461: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x42495d0 initialized for platform Host (this does not guarantee that XLA will be used). Devices: | |
2020-11-11 17:13:19.659476: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Host, Default Version | |
2020-11-11 17:13:19.661119: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcuda.so.1 | |
2020-11-11 17:13:19.762619: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x424ba40 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices: | |
2020-11-11 17:13:19.762647: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): GeForce RTX 3080, Compute Capability 8.6 | |
2020-11-11 17:13:19.763312: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:13:19.763341: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:13:19.765522: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:13:19.766334: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:13:19.766554: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:13:19.768943: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:13:19.769488: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:13:19.769601: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:13:19.770685: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:13:19.770712: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:13:20.034534: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:13:20.034571: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:13:20.034576: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:13:20.035792: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
2020-11-11 17:13:20.037047: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:13:20.037075: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:13:20.037095: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:13:20.037104: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:13:20.037112: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:13:20.037120: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:13:20.037129: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:13:20.037137: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:13:20.038153: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:13:20.038171: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:13:20.038176: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:13:20.038182: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:13:20.039206: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
2020-11-11 17:13:20.039941: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:13:20.039960: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:13:20.039975: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:13:20.039983: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:13:20.039993: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:13:20.040003: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:13:20.040011: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:13:20.040020: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:13:20.041033: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:13:20.041047: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:13:20.041053: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:13:20.041058: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:13:20.042079: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
* TF Version: 1.15.4 | |
* Platform: Linux-5.4.0-52-generic-x86_64-with-Ubuntu-18.04-bionic | |
* CPU: N/A | |
* CPU RAM: 63 GB | |
* GPU/0: GeForce RTX 3080 | |
* GPU RAM: 7.8 GB | |
* CUDA Version: 11.1 | |
* CUDA Build: V11.1.74 | |
The benchmark is running... | |
The tests might take up to 20 minutes | |
Please don't interrupt the script | |
1/19. MobileNet-V2 | |
2020-11-11 17:13:26.991308: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:13:26.991341: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:13:26.991358: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:13:26.991367: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:13:26.991375: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:13:26.991385: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:13:26.991394: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:13:26.991412: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:13:26.992452: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:13:26.993437: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:13:26.993461: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:13:26.993476: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:13:26.993485: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:13:26.993493: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:13:26.993501: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:13:26.993510: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:13:26.993523: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:13:26.994578: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:13:26.994607: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:13:26.994614: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:13:26.994620: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:13:26.995695: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
2020-11-11 17:13:28.233417: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:13:29.596103: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
1.1 - inference | batch=50, size=224x224: 64.3 ± 142.0 ms | |
1.2 - training | batch=50, size=224x224: 218 ± 210 ms | |
2/19. Inception-V3 | |
2020-11-11 17:13:52.136430: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:13:52.136468: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:13:52.136490: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:13:52.136499: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:13:52.136508: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:13:52.136517: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:13:52.136526: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:13:52.136535: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:13:52.137288: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:13:52.137313: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:13:52.137319: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:13:52.137325: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:13:52.138098: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
2.1 - inference | batch=20, size=346x346: 51.6 ± 3.5 ms | |
2.2 - training | batch=20, size=346x346: 126.6 ± 0.5 ms | |
3/19. Inception-V4 | |
2020-11-11 17:14:07.102147: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:14:07.102184: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:14:07.102207: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:14:07.102216: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:14:07.102225: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:14:07.102233: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:14:07.102241: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:14:07.102253: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:14:07.103045: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:14:07.103068: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:14:07.103074: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:14:07.103079: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:14:07.103877: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
3.1 - inference | batch=10, size=346x346: 38.1 ± 0.3 ms | |
3.2 - training | batch=10, size=346x346: 160.2 ± 0.4 ms | |
4/19. Inception-ResNet-V2 | |
2020-11-11 17:14:25.256862: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:14:25.256897: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:14:25.256917: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:14:25.256925: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:14:25.256934: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:14:25.256942: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:14:25.256950: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:14:25.256959: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:14:25.257745: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:14:25.257772: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:14:25.257780: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:14:25.257784: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:14:25.258593: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
4.1 - inference | batch=10, size=346x346: 51.5 ± 0.7 ms | |
4.2 - training | batch=8, size=346x346: 167.3 ± 0.5 ms | |
5/19. ResNet-V2-50 | |
2020-11-11 17:14:52.537824: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:14:52.537859: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:14:52.537881: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:14:52.537889: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:14:52.537903: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:14:52.537912: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:14:52.537919: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:14:52.537928: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:14:52.538728: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:14:52.538754: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:14:52.538761: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:14:52.538766: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:14:52.539565: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
5.1 - inference | batch=10, size=346x346: 27.3 ± 0.7 ms | |
5.2 - training | batch=10, size=346x346: 82.3 ± 0.5 ms | |
6/19. ResNet-V2-152 | |
2020-11-11 17:15:02.476257: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:15:02.476289: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:15:02.476309: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:15:02.476317: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:15:02.476324: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:15:02.476333: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:15:02.476341: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:15:02.476358: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:15:02.477132: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:15:02.477158: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:15:02.477164: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:15:02.477170: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:15:02.477972: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
6.1 - inference | batch=10, size=256x256: 36.0 ± 0.2 ms | |
6.2 - training | batch=10, size=256x256: 145.5 ± 0.7 ms | |
7/19. VGG-16 | |
2020-11-11 17:15:23.695433: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:15:23.695471: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:15:23.695493: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:15:23.695502: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:15:23.695511: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:15:23.695519: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:15:23.695530: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:15:23.695543: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:15:23.696334: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:15:23.696360: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:15:23.696366: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:15:23.696372: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:15:23.697166: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
7.1 - inference | batch=20, size=224x224: 46.3 ± 0.6 ms | |
2020-11-11 17:15:27.327238: W tensorflow/core/framework/cpu_allocator_impl.cc:81] Allocation of 411041792 exceeds 10% of system memory. | |
2020-11-11 17:15:27.483195: W tensorflow/core/framework/cpu_allocator_impl.cc:81] Allocation of 411041792 exceeds 10% of system memory. | |
2020-11-11 17:15:27.789007: W tensorflow/core/framework/cpu_allocator_impl.cc:81] Allocation of 411041792 exceeds 10% of system memory. | |
2020-11-11 17:15:27.944590: W tensorflow/core/framework/cpu_allocator_impl.cc:81] Allocation of 411041792 exceeds 10% of system memory. | |
2020-11-11 17:15:28.169532: W tensorflow/core/framework/cpu_allocator_impl.cc:81] Allocation of 411041792 exceeds 10% of system memory. | |
7.2 - training | batch=2, size=224x224: 64.3 ± 0.5 ms | |
8/19. SRCNN 9-5-5 | |
2020-11-11 17:15:32.938901: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:15:32.938940: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:15:32.938960: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:15:32.938969: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:15:32.938977: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:15:32.938985: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:15:32.938993: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:15:32.939002: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:15:32.939769: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:15:32.939792: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:15:32.939798: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:15:32.939804: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:15:32.940588: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
8.1 - inference | batch=10, size=512x512: 54.5 ± 4.0 ms | |
8.2 - inference | batch=1, size=1536x1536: 54.2 ± 0.4 ms | |
8.3 - training | batch=10, size=512x512: 135 ± 7 ms | |
9/19. VGG-19 Super-Res | |
2020-11-11 17:15:55.678941: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:15:55.678972: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:15:55.678992: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:15:55.679000: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:15:55.679008: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:15:55.679015: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:15:55.679023: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:15:55.679032: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:15:55.679797: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:15:55.679818: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:15:55.679825: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:15:55.679830: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:15:55.680624: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
9.1 - inference | batch=10, size=256x256: 59.8 ± 0.8 ms | |
9.2 - inference | batch=1, size=1024x1024: 102.8 ± 0.6 ms | |
9.3 - training | batch=10, size=224x224: 176.3 ± 0.9 ms | |
10/19. ResNet-SRGAN | |
2020-11-11 17:16:18.186756: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:16:18.186788: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:16:18.186807: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:16:18.186815: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:16:18.186822: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:16:18.186829: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:16:18.186837: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:16:18.186845: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:16:18.187604: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:16:18.187627: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:16:18.187633: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:16:18.187639: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:16:18.188435: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
10.1 - inference | batch=10, size=512x512: 66.5 ± 1.8 ms | |
10.2 - inference | batch=1, size=1536x1536: 63.3 ± 1.4 ms | |
10.3 - training | batch=5, size=512x512: 103.1 ± 0.9 ms | |
11/19. ResNet-DPED | |
2020-11-11 17:16:36.702263: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:16:36.702296: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:16:36.702315: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:16:36.702323: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:16:36.702331: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:16:36.702340: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:16:36.702348: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:16:36.702357: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:16:36.703212: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:16:36.703239: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:16:36.703245: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:16:36.703251: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:16:36.704043: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
11.1 - inference | batch=10, size=256x256: 67.2 ± 0.9 ms | |
11.2 - inference | batch=1, size=1024x1024: 122.4 ± 0.5 ms | |
11.3 - training | batch=15, size=128x128: 105.4 ± 0.5 ms | |
12/19. U-Net | |
2020-11-11 17:17:03.678634: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:17:03.678668: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:17:03.678686: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:17:03.678693: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:17:03.678701: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:17:03.678710: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:17:03.678719: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:17:03.678727: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:17:03.679576: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:17:03.679599: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:17:03.679605: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:17:03.679611: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:17:03.680480: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
12.1 - inference | batch=4, size=512x512: 117.3 ± 0.5 ms | |
12.2 - inference | batch=1, size=1024x1024: 141.1 ± 0.6 ms | |
12.3 - training | batch=4, size=256x256: 130.4 ± 0.5 ms | |
13/19. Nvidia-SPADE | |
2020-11-11 17:17:24.647385: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:17:24.647419: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:17:24.647437: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:17:24.647445: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:17:24.647452: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:17:24.647461: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:17:24.647469: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:17:24.647477: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:17:24.648242: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:17:24.648265: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:17:24.648271: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:17:24.648276: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:17:24.649062: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
13.1 - inference | batch=5, size=128x128: 50.9 ± 0.3 ms | |
13.2 - training | batch=1, size=128x128: 87.0 ± 0.6 ms | |
14/19. ICNet | |
2020-11-11 17:17:40.429806: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:17:40.429844: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:17:40.429872: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:17:40.429882: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:17:40.429901: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:17:40.429915: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:17:40.429929: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:17:40.429944: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:17:40.430740: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:17:40.430765: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:17:40.430772: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:17:40.430777: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:17:40.431564: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
14.1 - inference | batch=5, size=1024x1536: 138 ± 1 ms | |
14.2 - training | batch=10, size=1024x1536: 442 ± 3 ms | |
15/19. PSPNet | |
2020-11-11 17:18:09.403981: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:18:09.404015: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:18:09.404035: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:18:09.404043: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:18:09.404052: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:18:09.404061: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:18:09.404069: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:18:09.404078: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:18:09.404869: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:18:09.404895: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:18:09.404901: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:18:09.404907: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:18:09.405696: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
15.1 - inference | batch=5, size=720x720: 202.1 ± 0.7 ms | |
15.2 - training | batch=1, size=512x512: 91.5 ± 0.5 ms | |
16/19. DeepLab | |
2020-11-11 17:18:26.213337: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:18:26.213370: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:18:26.213390: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:18:26.213398: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:18:26.213405: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:18:26.213412: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:18:26.213420: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:18:26.213429: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:18:26.214210: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:18:26.214238: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:18:26.214245: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:18:26.214250: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:18:26.215050: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
16.1 - inference | batch=2, size=512x512: 68.5 ± 0.5 ms | |
16.2 - training | batch=1, size=384x384: 81.5 ± 0.5 ms | |
17/19. Pixel-RNN | |
2020-11-11 17:18:41.661295: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:18:41.661329: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:18:41.661349: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:18:41.661357: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:18:41.661366: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:18:41.661375: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:18:41.661383: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:18:41.661399: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:18:41.662184: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:18:41.662210: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:18:41.662217: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:18:41.662222: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:18:41.663025: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
17.1 - inference | batch=50, size=64x64: 294 ± 7 ms | |
17.2 - training | batch=10, size=64x64: 961 ± 14 ms | |
18/19. LSTM-Sentiment | |
2020-11-11 17:20:11.907017: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:20:11.907063: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:20:11.907094: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:20:11.907105: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:20:11.907115: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:20:11.907125: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:20:11.907135: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:20:11.907157: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:20:11.910236: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:20:11.910270: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:20:11.910277: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:20:11.910283: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:20:11.911359: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
18.1 - inference | batch=100, size=1024x300: 324 ± 2 ms | |
18.2 - training | batch=10, size=1024x300: 412 ± 6 ms | |
19/19. GNMT-Translation | |
2020-11-11 17:20:34.213415: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1665] Found device 0 with properties: | |
name: GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.8 | |
pciBusID: 0000:69:00.0 | |
2020-11-11 17:20:34.213452: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0 | |
2020-11-11 17:20:34.213478: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11 | |
2020-11-11 17:20:34.213488: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10 | |
2020-11-11 17:20:34.213498: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10 | |
2020-11-11 17:20:34.213508: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11 | |
2020-11-11 17:20:34.213518: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11 | |
2020-11-11 17:20:34.213529: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8 | |
2020-11-11 17:20:34.214316: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1793] Adding visible gpu devices: 0 | |
2020-11-11 17:20:34.214341: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix: | |
2020-11-11 17:20:34.214349: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0 | |
2020-11-11 17:20:34.214356: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N | |
2020-11-11 17:20:34.215152: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8034 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3080, pci bus id: 0000:69:00.0, compute capability: 8.6) | |
19.1 - inference | batch=1, size=1x20: 91.5 ± 0.9 ms | |
Device Inference Score: 18020 | |
Device Training Score: 19392 | |
Device AI Score: 37412 | |
For more information and results, please visit http://ai-benchmark.com/alpha |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment