A CK-powered demo of fully automated Design Space Exploration of ML/SW/HW stacks for object detection using CK workflows and dashboards:

Performance/accuracy exploration with 12 object detection models, 5 TensorFlow backends and 6 batch sizes.

Table of Contents

Overview

This Jupyter Notebook studies performance (execution time per image, images per seconds) vs accuracy (mAP, Recall) of several Object Detection models on different size objects (large, medium and small). The experiments are performed via TensorFlow with several execution options on the CPU and the GPU.

Model Unique CK Tags (<tags>) Is Custom? mAP in %
faster_rcnn_nas_lowproposals_coco rcnn,nas,lowproposals,vcoco 0 44.340195
faster_rcnn_resnet50_lowproposals_coco rcnn,resnet50,lowproposals 0 24.241037
faster_rcnn_resnet101_lowproposals_coco rcnn,resnet101,lowproposals 0 32.594327
faster_rcnn_inception_resnet_v2_atrous_lowproposals_coco rcnn,inception-resnet-v2,lowproposals 0 36.520117
faster_rcnn_inception_v2_coco rcnn,inception-v2 0 28.309626
ssd_inception_v2_coco ssd,inception-v2 0 27.765988
ssd_mobilenet_v1_coco ssd,mobilenet-v1,non-quantized,mlperf,tf 0 23.111170
ssd_mobilenet_v1_quantized_coco ssd,mobilenet-v1,quantized,mlperf,tf 0 23.591693
ssd_mobilenet_v1_fpn_coco ssd,mobilenet-v1,fpn 0 35.353170
ssd_resnet_50_fpn_coco ssd,resnet50,fpn 0 38.341120
ssdlite_mobilenet_v2_coco ssdlite,mobilenet-v2,vcoco 0 24.281540
yolo_v3_coco yolo-v3 1 28.532508

Platform

See our Docker container for more information on the software configuration.

CPU

  • Model:
    • Intel Xeon E5-2650 v3
  • Frequency:
    • 2.3 GHz
  • Number of physical cores:
    • 10
  • Number of virtual cores (hyperthreading):
    • 20
  • RAM:
    • 32 GB
  • OS:
    • Ubuntu 16.04 LTS Linux

GPU

  • Model:
    • NVIDIA GeForce GTX 1080
  • Frequency:
    • 1.6 GHz
  • RAM:
    • 8 GB
  • CUDA version:
    • 10.2
  • Driver version:
    • 430.14

Settings

NB: Please ignore this section if you are not interested in re-running or modifying this notebook.

Includes

Standard

In [1]:
import os
import sys
import json
import re

Scientific

If some of the scientific packages are missing, please install them using:

$ python -m pip install jupyter pandas numpy matplotlib --user
In [2]:
import IPython as ip
import pandas as pd
import numpy as np
import matplotlib as mp
import seaborn as sb
In [3]:
print ('IPython version: %s' % ip.__version__)
print ('Pandas version: %s' % pd.__version__)
print ('NumPy version: %s' % np.__version__)
print ('Matplotlib version: %s' % mp.__version__)
print ('Seaborn version: %s' % sb.__version__)
IPython version: 7.5.0
Pandas version: 0.25.1
NumPy version: 1.17.2
Matplotlib version: 3.1.1
Seaborn version: 0.9.0
In [4]:
from IPython.display import Image, display
def display_in_full(df):
    pd.options.display.max_columns = len(df.columns)
    pd.options.display.max_rows = len(df.index)
    display(df)
In [5]:
import matplotlib.pyplot as plt
from matplotlib import cm
%matplotlib inline
In [6]:
default_colormap = cm.autumn
default_fontsize = 16
default_barwidth = 0.8
default_figwidth = 24
default_figheight = 3
default_figdpi = 200
default_figsize = [default_figwidth, default_figheight]
In [7]:
if mp.__version__[0]=='2': mp.style.use('classic')
mp.rcParams['figure.max_open_warning'] = 200
mp.rcParams['figure.dpi'] = default_figdpi
mp.rcParams['font.size'] = default_fontsize
mp.rcParams['legend.fontsize'] = 'medium'
In [8]:
save_fig_ext = 'png'
save_fig_dir = os.path.join(os.path.expanduser("~"), 'omnibench')
if not os.path.exists(save_fig_dir):
    os.mkdir(save_fig_dir)
In [9]:
from pprint import pprint

Collective Knowledge

If CK is not installed, please install it using:

$ python -m pip install ck --user
In [10]:
import ck.kernel as ck
print ('CK version: %s' % ck.__version__)
CK version: 1.10.3.1

Get the experimental data

Download experimental data and add CK repositories as follows:

$ wget https://www.dropbox.com/s/g0bc1exh1621k1q/ckr-dse-demo-object-detection-accuracy.zip --no-check-certificate
$ ck add repo --zip=ckr-dse-demo-object-detection-accuracy.zip

$ wget https://www.dropbox.com/s/g0bc1exh1621k1q/ckr-dse-demo-object-detection-performance-docker.zip --no-check-certificate
$ ck add repo --zip=ckr-dse-demo-object-detection-performance-docker.zip

$ wget https://www.dropbox.com/s/g0bc1exh1621k1q/ckr-dse-demo-object-detection-performance-native.zip --no-check-certificate
$ ck add repo --zip=ckr-dse-demo-object-detection-performance-native.zip
In [11]:
repo_uoa = 'dse-demo-object-detection-accuracy'
print ("*"*80)
print (repo_uoa)
print ("*"*80)
!ck list $repo_uoa:experiment:* | sort
print ("")

perf_docker_repo_uoa = 'dse-demo-object-detection-performance-docker'
print ("*"*80)
print (perf_docker_repo_uoa)
print ("*"*80)
!ck list $perf_docker_repo_uoa:experiment:* | sort
print ("")

perf_native_repo_uoa = 'dse-demo-object-detection-performance-native'
print ("*"*80)
print (perf_native_repo_uoa)
print ("*"*80)
!ck list $perf_native_repo_uoa:experiment:* | sort
********************************************************************************
dse-demo-object-detection-accuracy
********************************************************************************
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-accuracy-model-width-height
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-accuracy-no-batch
mlperf-object-detection-rcnn-inception-v2-tf-py-accuracy-model-width-height
mlperf-object-detection-rcnn-inception-v2-tf-py-accuracy-no-batch
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-accuracy-model-width-height
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-accuracy-no-batch
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-accuracy-model-width-height
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-accuracy-no-batch
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-accuracy-model-width-height
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-accuracy-no-batch
mlperf-object-detection-ssd-inception-v2-tf-py-accuracy-model-width-height
mlperf-object-detection-ssd-inception-v2-tf-py-accuracy-no-batch
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-accuracy-model-width-height
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-accuracy-no-batch
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-accuracy-model-width-height
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-accuracy-no-batch
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-accuracy-model-width-height
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-accuracy-no-batch
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-accuracy-model-width-height
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-accuracy-no-batch
mlperf-object-detection-ssd-resnet50-fpn-tf-py-accuracy-model-width-height
mlperf-object-detection-ssd-resnet50-fpn-tf-py-accuracy-no-batch
mlperf-object-detection-yolo-v3-tf-py-accuracy-no-batch

********************************************************************************
dse-demo-object-detection-performance-docker
********************************************************************************
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-cpu-batch-size1
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-cpu-batch-size16
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-cpu-batch-size2
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-cpu-batch-size32
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-cpu-batch-size4
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-cpu-batch-size8
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-cpu-prebuilt-batch-size1
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-cpu-prebuilt-batch-size16
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-cpu-prebuilt-batch-size2
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-cpu-prebuilt-batch-size32
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-cpu-prebuilt-batch-size4
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-cpu-prebuilt-batch-size8
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-cuda-batch-size1
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-cuda-batch-size16
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-cuda-batch-size2
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-cuda-batch-size4
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-cuda-batch-size8
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-tensorrt-batch-size1
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-tensorrt-batch-size16
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-tensorrt-batch-size2
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-tensorrt-batch-size4
mlperf-object-detection-rcnn-inception-resnet-v2-lowproposals-tf-py-performance-tensorrt-batch-size8
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-cpu-batch-size1
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-cpu-batch-size16
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-cpu-batch-size2
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-cpu-batch-size32
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-cpu-batch-size4
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-cpu-batch-size8
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-cpu-prebuilt-batch-size1
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-cpu-prebuilt-batch-size16
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-cpu-prebuilt-batch-size2
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-cpu-prebuilt-batch-size32
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-cpu-prebuilt-batch-size4
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-cpu-prebuilt-batch-size8
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-cuda-batch-size1
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-cuda-batch-size16
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-cuda-batch-size2
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-cuda-batch-size32
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-cuda-batch-size4
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-cuda-batch-size8
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-tensorrt-batch-size1
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-tensorrt-batch-size16
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-tensorrt-batch-size2
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-tensorrt-batch-size32
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-tensorrt-batch-size4
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-tensorrt-batch-size8
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-tensorrt-dynamic-batch-size1
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-tensorrt-dynamic-batch-size16
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-tensorrt-dynamic-batch-size2
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-tensorrt-dynamic-batch-size4
mlperf-object-detection-rcnn-inception-v2-tf-py-performance-tensorrt-dynamic-batch-size8
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-performance-cpu-batch-size1
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-performance-cpu-batch-size16
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-performance-cpu-batch-size2
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-performance-cpu-batch-size32
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-performance-cpu-batch-size4
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-performance-cpu-batch-size8
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-performance-cpu-prebuilt-batch-size1
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-performance-cpu-prebuilt-batch-size16
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-performance-cpu-prebuilt-batch-size2
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-performance-cpu-prebuilt-batch-size32
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-performance-cpu-prebuilt-batch-size4
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-performance-cpu-prebuilt-batch-size8
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-performance-cuda-batch-size1
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-performance-cuda-batch-size2
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-performance-cuda-batch-size4
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-performance-cuda-batch-size8
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-performance-tensorrt-batch-size1
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-performance-tensorrt-batch-size2
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-performance-tensorrt-batch-size4
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-performance-tensorrt-batch-size8
mlperf-object-detection-rcnn-nas-lowproposals-tf-py-performance-tensorrt-dynamic-batch-size1
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-cpu-batch-size1
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-cpu-batch-size16
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-cpu-batch-size2
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-cpu-batch-size32
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-cpu-batch-size4
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-cpu-batch-size8
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-cpu-prebuilt-batch-size1
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-cpu-prebuilt-batch-size16
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-cpu-prebuilt-batch-size2
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-cpu-prebuilt-batch-size32
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-cpu-prebuilt-batch-size4
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-cpu-prebuilt-batch-size8
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-cuda-batch-size1
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-cuda-batch-size16
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-cuda-batch-size2
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-cuda-batch-size32
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-cuda-batch-size4
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-cuda-batch-size8
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-tensorrt-batch-size1
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-tensorrt-batch-size16
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-tensorrt-batch-size2
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-tensorrt-batch-size32
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-tensorrt-batch-size4
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-tensorrt-batch-size8
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-tensorrt-dynamic-batch-size1
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-tensorrt-dynamic-batch-size16
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-tensorrt-dynamic-batch-size2
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-tensorrt-dynamic-batch-size4
mlperf-object-detection-rcnn-resnet101-lowproposals-tf-py-performance-tensorrt-dynamic-batch-size8
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-cpu-batch-size1
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-cpu-batch-size16
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-cpu-batch-size2
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-cpu-batch-size32
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-cpu-batch-size4
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-cpu-batch-size8
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-cpu-prebuilt-batch-size1
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-cpu-prebuilt-batch-size16
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-cpu-prebuilt-batch-size2
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-cpu-prebuilt-batch-size32
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-cpu-prebuilt-batch-size4
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-cpu-prebuilt-batch-size8
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-cuda-batch-size1
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-cuda-batch-size16
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-cuda-batch-size2
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-cuda-batch-size32
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-cuda-batch-size4
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-cuda-batch-size8
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-tensorrt-batch-size1
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-tensorrt-batch-size16
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-tensorrt-batch-size2
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-tensorrt-batch-size32
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-tensorrt-batch-size4
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-tensorrt-batch-size8
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-tensorrt-dynamic-batch-size1
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-tensorrt-dynamic-batch-size16
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-tensorrt-dynamic-batch-size2
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-tensorrt-dynamic-batch-size4
mlperf-object-detection-rcnn-resnet50-lowproposals-tf-py-performance-tensorrt-dynamic-batch-size8
mlperf-object-detection-ssd-inception-v2-tf-py-performance-cpu-batch-size1
mlperf-object-detection-ssd-inception-v2-tf-py-performance-cpu-batch-size16
mlperf-object-detection-ssd-inception-v2-tf-py-performance-cpu-batch-size2
mlperf-object-detection-ssd-inception-v2-tf-py-performance-cpu-batch-size32
mlperf-object-detection-ssd-inception-v2-tf-py-performance-cpu-batch-size4
mlperf-object-detection-ssd-inception-v2-tf-py-performance-cpu-batch-size8
mlperf-object-detection-ssd-inception-v2-tf-py-performance-cpu-prebuilt-batch-size1
mlperf-object-detection-ssd-inception-v2-tf-py-performance-cpu-prebuilt-batch-size16
mlperf-object-detection-ssd-inception-v2-tf-py-performance-cpu-prebuilt-batch-size2
mlperf-object-detection-ssd-inception-v2-tf-py-performance-cpu-prebuilt-batch-size32
mlperf-object-detection-ssd-inception-v2-tf-py-performance-cpu-prebuilt-batch-size4
mlperf-object-detection-ssd-inception-v2-tf-py-performance-cpu-prebuilt-batch-size8
mlperf-object-detection-ssd-inception-v2-tf-py-performance-cuda-batch-size1
mlperf-object-detection-ssd-inception-v2-tf-py-performance-cuda-batch-size16
mlperf-object-detection-ssd-inception-v2-tf-py-performance-cuda-batch-size2
mlperf-object-detection-ssd-inception-v2-tf-py-performance-cuda-batch-size32
mlperf-object-detection-ssd-inception-v2-tf-py-performance-cuda-batch-size4
mlperf-object-detection-ssd-inception-v2-tf-py-performance-cuda-batch-size8
mlperf-object-detection-ssd-inception-v2-tf-py-performance-tensorrt-batch-size1
mlperf-object-detection-ssd-inception-v2-tf-py-performance-tensorrt-batch-size16
mlperf-object-detection-ssd-inception-v2-tf-py-performance-tensorrt-batch-size2
mlperf-object-detection-ssd-inception-v2-tf-py-performance-tensorrt-batch-size32
mlperf-object-detection-ssd-inception-v2-tf-py-performance-tensorrt-batch-size4
mlperf-object-detection-ssd-inception-v2-tf-py-performance-tensorrt-batch-size8
mlperf-object-detection-ssd-inception-v2-tf-py-performance-tensorrt-dynamic-batch-size1
mlperf-object-detection-ssd-inception-v2-tf-py-performance-tensorrt-dynamic-batch-size16
mlperf-object-detection-ssd-inception-v2-tf-py-performance-tensorrt-dynamic-batch-size2
mlperf-object-detection-ssd-inception-v2-tf-py-performance-tensorrt-dynamic-batch-size32
mlperf-object-detection-ssd-inception-v2-tf-py-performance-tensorrt-dynamic-batch-size4
mlperf-object-detection-ssd-inception-v2-tf-py-performance-tensorrt-dynamic-batch-size8
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-cpu-batch-size1
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-cpu-batch-size16
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-cpu-batch-size2
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-cpu-batch-size32
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-cpu-batch-size4
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-cpu-batch-size8
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-cpu-prebuilt-batch-size1
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-cpu-prebuilt-batch-size16
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-cpu-prebuilt-batch-size2
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-cpu-prebuilt-batch-size32
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-cpu-prebuilt-batch-size4
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-cpu-prebuilt-batch-size8
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-cuda-batch-size1
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-cuda-batch-size16
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-cuda-batch-size2
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-cuda-batch-size32
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-cuda-batch-size4
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-cuda-batch-size8
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-tensorrt-batch-size1
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-tensorrt-batch-size16
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-tensorrt-batch-size2
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-tensorrt-batch-size32
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-tensorrt-batch-size4
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-tensorrt-batch-size8
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-tensorrt-dynamic-batch-size1
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-tensorrt-dynamic-batch-size16
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-tensorrt-dynamic-batch-size2
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-tensorrt-dynamic-batch-size32
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-tensorrt-dynamic-batch-size4
mlperf-object-detection-ssdlite-mobilenet-v2-tf-py-performance-tensorrt-dynamic-batch-size8
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-cpu-batch-size1
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-cpu-batch-size16
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-cpu-batch-size2
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-cpu-batch-size32
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-cpu-batch-size4
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-cpu-batch-size8
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-cpu-prebuilt-batch-size1
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-cpu-prebuilt-batch-size16
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-cpu-prebuilt-batch-size2
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-cpu-prebuilt-batch-size32
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-cpu-prebuilt-batch-size4
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-cpu-prebuilt-batch-size8
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-cuda-batch-size1
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-cuda-batch-size16
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-cuda-batch-size2
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-cuda-batch-size32
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-cuda-batch-size4
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-cuda-batch-size8
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorrt-batch-size1
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorrt-batch-size16
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorrt-batch-size2
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorrt-batch-size32
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorrt-batch-size4
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorrt-batch-size8
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorrt-dynamic-batch-size1
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorrt-dynamic-batch-size16
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorrt-dynamic-batch-size2
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorrt-dynamic-batch-size32
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorrt-dynamic-batch-size4
mlperf-object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorrt-dynamic-batch-size8
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-cpu-batch-size1
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-cpu-batch-size16
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-cpu-batch-size2
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-cpu-batch-size32
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-cpu-batch-size4
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-cpu-batch-size8
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-cpu-prebuilt-batch-size1
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-cpu-prebuilt-batch-size16
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-cpu-prebuilt-batch-size2
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-cpu-prebuilt-batch-size32
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-cpu-prebuilt-batch-size4
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-cpu-prebuilt-batch-size8
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-cuda-batch-size1
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-cuda-batch-size16
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-cuda-batch-size2
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-cuda-batch-size32
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-cuda-batch-size4
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-cuda-batch-size8
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-tensorrt-batch-size1
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-tensorrt-batch-size16
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-tensorrt-batch-size2
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-tensorrt-batch-size32
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-tensorrt-batch-size4
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-tensorrt-batch-size8
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-tensorrt-dynamic-batch-size1
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-tensorrt-dynamic-batch-size16
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-tensorrt-dynamic-batch-size2
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-tensorrt-dynamic-batch-size32
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-tensorrt-dynamic-batch-size4
mlperf-object-detection-ssd-mobilenet-v1-non-quantized-mlperf-tf-py-performance-tensorrt-dynamic-batch-size8
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-cpu-batch-size1
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-cpu-batch-size16
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-cpu-batch-size2
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-cpu-batch-size32
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-cpu-batch-size4
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-cpu-batch-size8
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-cpu-prebuilt-batch-size1
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-cpu-prebuilt-batch-size16
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-cpu-prebuilt-batch-size2
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-cpu-prebuilt-batch-size32
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-cpu-prebuilt-batch-size4
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-cpu-prebuilt-batch-size8
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-cuda-batch-size1
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-cuda-batch-size16
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-cuda-batch-size2
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-cuda-batch-size32
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-cuda-batch-size4
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-cuda-batch-size8
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorrt-batch-size1
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorrt-batch-size16
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorrt-batch-size2
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorrt-batch-size32
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorrt-batch-size4
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorrt-batch-size8
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorrt-dynamic-batch-size1
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorrt-dynamic-batch-size16
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorrt-dynamic-batch-size2
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorrt-dynamic-batch-size32
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorrt-dynamic-batch-size4
mlperf-object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorrt-dynamic-batch-size8
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-cpu-batch-size1
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-cpu-batch-size16
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-cpu-batch-size2
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-cpu-batch-size32
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-cpu-batch-size4
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-cpu-batch-size8
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-cpu-prebuilt-batch-size1
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-cpu-prebuilt-batch-size16
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-cpu-prebuilt-batch-size2
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-cpu-prebuilt-batch-size32
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-cpu-prebuilt-batch-size4
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-cpu-prebuilt-batch-size8
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-cuda-batch-size1
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-cuda-batch-size16
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-cuda-batch-size2
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-cuda-batch-size32
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-cuda-batch-size4
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-cuda-batch-size8
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-tensorrt-batch-size1
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-tensorrt-batch-size16
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-tensorrt-batch-size2
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-tensorrt-batch-size32
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-tensorrt-batch-size4
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-tensorrt-batch-size8
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-tensorrt-dynamic-batch-size1
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-tensorrt-dynamic-batch-size16
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-tensorrt-dynamic-batch-size2
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-tensorrt-dynamic-batch-size32
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-tensorrt-dynamic-batch-size4
mlperf-object-detection-ssd-resnet50-fpn-tf-py-performance-tensorrt-dynamic-batch-size8
mlperf-object-detection-yolo-v3-tf-py-performance-cpu-batch-size1
mlperf-object-detection-yolo-v3-tf-py-performance-cpu-batch-size16
mlperf-object-detection-yolo-v3-tf-py-performance-cpu-batch-size2
mlperf-object-detection-yolo-v3-tf-py-performance-cpu-batch-size32
mlperf-object-detection-yolo-v3-tf-py-performance-cpu-batch-size4
mlperf-object-detection-yolo-v3-tf-py-performance-cpu-batch-size8
mlperf-object-detection-yolo-v3-tf-py-performance-cpu-prebuilt-batch-size1
mlperf-object-detection-yolo-v3-tf-py-performance-cpu-prebuilt-batch-size16
mlperf-object-detection-yolo-v3-tf-py-performance-cpu-prebuilt-batch-size2
mlperf-object-detection-yolo-v3-tf-py-performance-cpu-prebuilt-batch-size32
mlperf-object-detection-yolo-v3-tf-py-performance-cpu-prebuilt-batch-size4
mlperf-object-detection-yolo-v3-tf-py-performance-cpu-prebuilt-batch-size8
mlperf-object-detection-yolo-v3-tf-py-performance-cuda-batch-size1
mlperf-object-detection-yolo-v3-tf-py-performance-cuda-batch-size16
mlperf-object-detection-yolo-v3-tf-py-performance-cuda-batch-size2
mlperf-object-detection-yolo-v3-tf-py-performance-cuda-batch-size32
mlperf-object-detection-yolo-v3-tf-py-performance-cuda-batch-size4
mlperf-object-detection-yolo-v3-tf-py-performance-cuda-batch-size8
mlperf-object-detection-yolo-v3-tf-py-performance-tensorrt-batch-size1
mlperf-object-detection-yolo-v3-tf-py-performance-tensorrt-batch-size16
mlperf-object-detection-yolo-v3-tf-py-performance-tensorrt-batch-size2
mlperf-object-detection-yolo-v3-tf-py-performance-tensorrt-batch-size32
mlperf-object-detection-yolo-v3-tf-py-performance-tensorrt-batch-size4
mlperf-object-detection-yolo-v3-tf-py-performance-tensorrt-batch-size8
mlperf-object-detection-yolo-v3-tf-py-performance-tensorrt-dynamic-batch-size1
mlperf-object-detection-yolo-v3-tf-py-performance-tensorrt-dynamic-batch-size16
mlperf-object-detection-yolo-v3-tf-py-performance-tensorrt-dynamic-batch-size2
mlperf-object-detection-yolo-v3-tf-py-performance-tensorrt-dynamic-batch-size32
mlperf-object-detection-yolo-v3-tf-py-performance-tensorrt-dynamic-batch-size4
mlperf-object-detection-yolo-v3-tf-py-performance-tensorrt-dynamic-batch-size8

********************************************************************************
dse-demo-object-detection-performance-native
********************************************************************************
object-detection-rcnn-nas-lowproposals-tf-py-performance-prebuilt-cpu-batch-size1
object-detection-rcnn-nas-lowproposals-tf-py-performance-prebuilt-cpu-batch-size16
object-detection-rcnn-nas-lowproposals-tf-py-performance-prebuilt-cpu-batch-size2
object-detection-rcnn-nas-lowproposals-tf-py-performance-prebuilt-cpu-batch-size32
object-detection-rcnn-nas-lowproposals-tf-py-performance-prebuilt-cpu-batch-size4
object-detection-rcnn-nas-lowproposals-tf-py-performance-prebuilt-cpu-batch-size8
object-detection-rcnn-nas-lowproposals-tf-py-performance-source-cpu-batch-size1
object-detection-rcnn-nas-lowproposals-tf-py-performance-source-cpu-batch-size16
object-detection-rcnn-nas-lowproposals-tf-py-performance-source-cpu-batch-size2
object-detection-rcnn-nas-lowproposals-tf-py-performance-source-cpu-batch-size32
object-detection-rcnn-nas-lowproposals-tf-py-performance-source-cpu-batch-size4
object-detection-rcnn-nas-lowproposals-tf-py-performance-source-cpu-batch-size8
object-detection-rcnn-nas-lowproposals-tf-py-performance-source-cuda-batch-size1
object-detection-rcnn-nas-lowproposals-tf-py-performance-source-cuda-batch-size2
object-detection-rcnn-nas-lowproposals-tf-py-performance-source-cuda-batch-size4
object-detection-rcnn-nas-lowproposals-tf-py-performance-source-cuda-batch-size8
object-detection-rcnn-nas-lowproposals-tf-py-performance-tensorRT-batch-size1
object-detection-rcnn-nas-lowproposals-tf-py-performance-tensorRT-batch-size2
object-detection-rcnn-nas-lowproposals-tf-py-performance-tensorRT-batch-size4
object-detection-rcnn-nas-lowproposals-tf-py-performance-tensorRT-batch-size8
object-detection-rcnn-nas-lowproposals-tf-py-performance-tensorRT-dynamic-batch-size1
object-detection-rcnn-nas-lowproposals-tf-py-performance-tensorRT-dynamic-batch-size2
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-prebuilt-cpu-batch-size1
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-prebuilt-cpu-batch-size16
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-prebuilt-cpu-batch-size2
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-prebuilt-cpu-batch-size32
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-prebuilt-cpu-batch-size4
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-prebuilt-cpu-batch-size8
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-source-cpu-batch-size1
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-source-cpu-batch-size16
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-source-cpu-batch-size2
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-source-cpu-batch-size32
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-source-cpu-batch-size4
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-source-cpu-batch-size8
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-source-cuda-batch-size1
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-source-cuda-batch-size16
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-source-cuda-batch-size2
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-source-cuda-batch-size32
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-source-cuda-batch-size4
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-source-cuda-batch-size8
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorRT-batch-size1
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorRT-batch-size16
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorRT-batch-size2
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorRT-batch-size32
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorRT-batch-size4
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorRT-batch-size8
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorRT-dynamic-batch-size1
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorRT-dynamic-batch-size16
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorRT-dynamic-batch-size2
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorRT-dynamic-batch-size32
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorRT-dynamic-batch-size4
object-detection-ssd-mobilenet-v1-fpn-tf-py-performance-tensorRT-dynamic-batch-size8
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-prebuilt-cpu-batch-size1
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-prebuilt-cpu-batch-size16
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-prebuilt-cpu-batch-size2
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-prebuilt-cpu-batch-size32
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-prebuilt-cpu-batch-size4
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-prebuilt-cpu-batch-size8
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-source-cpu-batch-size1
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-source-cpu-batch-size16
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-source-cpu-batch-size2
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-source-cpu-batch-size32
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-source-cpu-batch-size4
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-source-cpu-batch-size8
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-source-cuda-batch-size1
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-source-cuda-batch-size16
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-source-cuda-batch-size2
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-source-cuda-batch-size32
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-source-cuda-batch-size4
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-source-cuda-batch-size8
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorRT-batch-size1
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorRT-batch-size16
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorRT-batch-size2
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorRT-batch-size32
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorRT-batch-size4
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorRT-batch-size8
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorRT-dynamic-batch-size1
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorRT-dynamic-batch-size16
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorRT-dynamic-batch-size2
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorRT-dynamic-batch-size32
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorRT-dynamic-batch-size4
object-detection-ssd-mobilenet-v1-quantized-mlperf-tf-py-performance-tensorRT-dynamic-batch-size8
object-detection-ssd-resnet50-fpn-tf-py-performance-prebuilt-cpu-batch-size1
object-detection-ssd-resnet50-fpn-tf-py-performance-prebuilt-cpu-batch-size16
object-detection-ssd-resnet50-fpn-tf-py-performance-prebuilt-cpu-batch-size2
object-detection-ssd-resnet50-fpn-tf-py-performance-prebuilt-cpu-batch-size32
object-detection-ssd-resnet50-fpn-tf-py-performance-prebuilt-cpu-batch-size4
object-detection-ssd-resnet50-fpn-tf-py-performance-prebuilt-cpu-batch-size8
object-detection-ssd-resnet50-fpn-tf-py-performance-source-cpu-batch-size1
object-detection-ssd-resnet50-fpn-tf-py-performance-source-cpu-batch-size16
object-detection-ssd-resnet50-fpn-tf-py-performance-source-cpu-batch-size2
object-detection-ssd-resnet50-fpn-tf-py-performance-source-cpu-batch-size32
object-detection-ssd-resnet50-fpn-tf-py-performance-source-cpu-batch-size4
object-detection-ssd-resnet50-fpn-tf-py-performance-source-cpu-batch-size8
object-detection-ssd-resnet50-fpn-tf-py-performance-source-cuda-batch-size1
object-detection-ssd-resnet50-fpn-tf-py-performance-source-cuda-batch-size16
object-detection-ssd-resnet50-fpn-tf-py-performance-source-cuda-batch-size2
object-detection-ssd-resnet50-fpn-tf-py-performance-source-cuda-batch-size32
object-detection-ssd-resnet50-fpn-tf-py-performance-source-cuda-batch-size4
object-detection-ssd-resnet50-fpn-tf-py-performance-source-cuda-batch-size8
object-detection-ssd-resnet50-fpn-tf-py-performance-tensorRT-batch-size1
object-detection-ssd-resnet50-fpn-tf-py-performance-tensorRT-batch-size16
object-detection-ssd-resnet50-fpn-tf-py-performance-tensorRT-batch-size2
object-detection-ssd-resnet50-fpn-tf-py-performance-tensorRT-batch-size32
object-detection-ssd-resnet50-fpn-tf-py-performance-tensorRT-batch-size4
object-detection-ssd-resnet50-fpn-tf-py-performance-tensorRT-batch-size8
object-detection-ssd-resnet50-fpn-tf-py-performance-tensorRT-dynamic-batch-size1
object-detection-ssd-resnet50-fpn-tf-py-performance-tensorRT-dynamic-batch-size16
object-detection-ssd-resnet50-fpn-tf-py-performance-tensorRT-dynamic-batch-size2
object-detection-ssd-resnet50-fpn-tf-py-performance-tensorRT-dynamic-batch-size32
object-detection-ssd-resnet50-fpn-tf-py-performance-tensorRT-dynamic-batch-size4
object-detection-ssd-resnet50-fpn-tf-py-performance-tensorRT-dynamic-batch-size8
object-detection-yolo-v3-tf-py-performance-prebuilt-cpu-batch-size1
object-detection-yolo-v3-tf-py-performance-prebuilt-cpu-batch-size16
object-detection-yolo-v3-tf-py-performance-prebuilt-cpu-batch-size2
object-detection-yolo-v3-tf-py-performance-prebuilt-cpu-batch-size32
object-detection-yolo-v3-tf-py-performance-prebuilt-cpu-batch-size4
object-detection-yolo-v3-tf-py-performance-prebuilt-cpu-batch-size8
object-detection-yolo-v3-tf-py-performance-source-cpu-batch-size1
object-detection-yolo-v3-tf-py-performance-source-cpu-batch-size16
object-detection-yolo-v3-tf-py-performance-source-cpu-batch-size2
object-detection-yolo-v3-tf-py-performance-source-cpu-batch-size32
object-detection-yolo-v3-tf-py-performance-source-cpu-batch-size4
object-detection-yolo-v3-tf-py-performance-source-cpu-batch-size8
object-detection-yolo-v3-tf-py-performance-source-cuda-batch-size1
object-detection-yolo-v3-tf-py-performance-source-cuda-batch-size16
object-detection-yolo-v3-tf-py-performance-source-cuda-batch-size2
object-detection-yolo-v3-tf-py-performance-source-cuda-batch-size32
object-detection-yolo-v3-tf-py-performance-source-cuda-batch-size4
object-detection-yolo-v3-tf-py-performance-source-cuda-batch-size8
object-detection-yolo-v3-tf-py-performance-tensorRT-batch-size1
object-detection-yolo-v3-tf-py-performance-tensorRT-batch-size16
object-detection-yolo-v3-tf-py-performance-tensorRT-batch-size2
object-detection-yolo-v3-tf-py-performance-tensorRT-batch-size32
object-detection-yolo-v3-tf-py-performance-tensorRT-batch-size4
object-detection-yolo-v3-tf-py-performance-tensorRT-batch-size8
object-detection-yolo-v3-tf-py-performance-tensorRT-dynamic-batch-size1
object-detection-yolo-v3-tf-py-performance-tensorRT-dynamic-batch-size16
object-detection-yolo-v3-tf-py-performance-tensorRT-dynamic-batch-size2
object-detection-yolo-v3-tf-py-performance-tensorRT-dynamic-batch-size32
object-detection-yolo-v3-tf-py-performance-tensorRT-dynamic-batch-size4
object-detection-yolo-v3-tf-py-performance-tensorRT-dynamic-batch-size8

Access the experimental data

In [12]:
def get_experimental_results(repo_uoa, module_uoa='experiment', tags='', accuracy=True):
    r = ck.access({'action':'search', 'repo_uoa':repo_uoa, 'module_uoa':module_uoa, 'tags':tags})
    #pprint (r)
    if r['return']>0:
        print('Error: %s' % r['error'])
        exit(1)
    experiments = r['lst']

    dfs = []
    for experiment in experiments:
        data_uoa = experiment['data_uoa']
        r = ck.access({'action':'list_points', 'repo_uoa':repo_uoa, 'module_uoa':module_uoa, 'data_uoa':data_uoa})
     
        pipeline_file_path = os.path.join(r['path'], 'pipeline.json')
        with open(pipeline_file_path) as pipeline_file:
            pipeline_data_raw = json.load(pipeline_file)
        weights_env  = pipeline_data_raw['dependencies']['weights']['dict']['env']
        image_width  = np.int64(weights_env.get('CK_ENV_TENSORFLOW_MODEL_DEFAULT_WIDTH',-1))
        image_height = np.int64(weights_env.get('CK_ENV_TENSORFLOW_MODEL_DEFAULT_HEIGHT',-1))
        
        tags = r['dict']['tags']
        #print (tags)
        for point in r['points']:
            point_file_path = os.path.join(r['path'], 'ckp-%s.0001.json' % point)
            with open(point_file_path) as point_file:
                point_data_raw = json.load(point_file)
            #pprint (point_data_raw['choices']['env'])
            characteristics_list = point_data_raw['characteristics_list']
            num_repetitions = len(characteristics_list)
            #platform = point_data_raw['features']['platform']['platform']['model']
            if np.int64(point_data_raw['choices']['env'].get('CK_ENABLE_BATCH',-1))==1:
                batch_enabled = True 
                batch_size = np.int64(point_data_raw['choices']['env'].get('CK_BATCH_SIZE',-1))
                batch_count = np.int64(point_data_raw['choices']['env'].get('CK_BATCH_COUNT',-1))
            else:
                batch_enabled = False
                batch_size = 1
                batch_count = np.int64(point_data_raw['choices']['env'].get('CK_BATCH_COUNT',-1)) * \
                              np.int64(point_data_raw['choices']['env'].get('CK_BATCH_SIZE',-1))

            characteristics = characteristics_list[0]
            if accuracy:
                data = [
                    {
                        'model': tags[0],
                        'backend':'cuda',
                        'batch_size': batch_size,
                        'batch_count': batch_count,
                        'batch_enabled': batch_enabled,
                        'image_height': image_height,
                        'image_width': image_width,
                        'num_reps':1,
                        # runtime characteristics
                        'Recall':     characteristics['run'].get('recall',0)*100,
                        'mAP':        characteristics['run'].get('mAP',0)*100,
                        'mAP_large':  characteristics['run']['metrics'].get('DetectionBoxes_Recall/[email protected] (large)', 0)*100,
                        'mAP_medium': characteristics['run']['metrics'].get('DetectionBoxes_Recall/[email protected] (medium)', 0)*100,
                        'mAP_small':  characteristics['run']['metrics'].get('DetectionBoxes_Recall/[email protected] (small)', 0)*100,
                    }
                ]
#                print(data[0]['model'])
            else: # performance
                ####### this conversion is still needed because some of the result have the old naming convention
                backend = 'default'
                trt = point_data_raw['choices']['env'].get('CK_ENABLE_TENSORRT',0)
                trt_dyn = point_data_raw['choices']['env'].get('CK_TENSORRT_DYNAMIC',0)
                if trt_dyn == '1':
                    backend = 'tensorrt-dynamic'
                elif trt == '1':
                    backend = 'tensorrt'
                elif tags[0] == 'tensorrt' or tags[0] =='source-cuda':
                    backend = 'cuda'
                elif tags[0] == 'tf-src-cpu' or tags[0] =='source-cpu':
                    backend = 'cpu'
                elif tags[0] == 'tf-prebuild-cpu' or tags[0] == 'prebuilt-cpu':
                    backend = 'cpu-prebuilt'
                else:
                    backend = tags[0]
                model = tags[1]
                data = [
                    {
                        'model': model,
                        'backend': backend,
                        'batch_size': batch_size,
                        'batch_count': batch_count,
                        'batch_enabled': batch_enabled,
                        'image_height': image_height,
                        'image_width': image_width,
                        'num_reps' : num_repetitions,
                        # statistical repetition
                        'repetition_id': repetition_id,
                        # runtime characteristics
                        'avg_fps': characteristics['run'].get('avg_fps', 'n/a')*batch_size,
                        'avg_time_ms': characteristics['run']['avg_time_ms']/batch_size,
                        'graph_load_time_ms': characteristics['run']['graph_load_time_s']*1e+3,
                        'images_load_time_avg_ms': characteristics['run']['images_load_time_avg_s']*1e+3,
                    }
                    for (repetition_id, characteristics) in zip(range(num_repetitions), characteristics_list)
                ]
            index = [
                'model', 'backend', 'batch_size', 'batch_count', 'batch_enabled', 'image_height', 'image_width', 'num_reps'
            ]
            # Construct a DataFrame.
            df = pd.DataFrame(data)
            df = df.set_index(index)
            # Append to the list of similarly constructed DataFrames.
            dfs.append(df)
    if dfs:
        # Concatenate all thus constructed DataFrames (i.e. stack on top of each other).
        result = pd.concat(dfs)
        result.sort_index(ascending=True, inplace=True)
    else:
        # Construct a dummy DataFrame the success status of which can be safely checked.
        result = pd.DataFrame(columns=['success?'])
    return result
In [13]:
#!ck recache repo
dfs = get_experimental_results(repo_uoa, accuracy=True)
dfs_perf = get_experimental_results(perf_docker_repo_uoa, accuracy=False)
dfs_perf_native = get_experimental_results(perf_native_repo_uoa, accuracy=False)
In [14]:
display_in_full(dfs)
display_in_full(dfs_perf)
Recall mAP mAP_large mAP_medium mAP_small
model backend batch_size batch_count batch_enabled image_height image_width num_reps
rcnn-inception-resnet-v2-lowproposals cuda 1 5000 False 1024 600 1 40.071771 36.521150 67.007142 44.677563 12.482768
True 1024 600 1 27.998174 24.014985 49.457345 28.361812 6.580855
rcnn-inception-v2 cuda 1 5000 False 600 1024 1 34.596863 28.307326 57.903100 37.694893 9.948821
True 600 1024 1 30.593184 23.891029 50.423664 32.742465 9.234102
rcnn-nas-lowproposals cuda 1 5000 False 1200 1200 1 47.510012 44.339995 73.981437 54.376904 19.925600
True 1200 1200 1 47.204636 43.983223 74.631044 53.876561 20.074105
rcnn-resnet101-lowproposals cuda 1 5000 False 600 1024 1 36.399789 32.586659 59.472751 38.935294 12.325213
True 600 1024 1 29.981835 25.969248 48.480626 30.466308 10.261601
rcnn-resnet50-lowproposals cuda 1 5000 False 600 1024 1 27.724275 24.233987 46.253149 27.783111 8.761679
True 600 1024 1 22.780471 19.241366 36.731852 21.222552 7.495941
ssd-inception-v2 cuda 1 5000 False 300 300 1 30.804457 27.766736 69.530638 23.179061 3.473937
True 300 300 1 30.708289 27.629812 69.414337 23.051823 3.399932
ssd-mobilenet-v1-fpn cuda 1 5000 False 640 640 1 42.661767 35.352120 63.507365 47.378434 17.679955
True 640 640 1 42.752999 35.362516 63.959455 47.239999 18.090466
ssd-mobilenet-v1-non-quantized-mlperf cuda 1 5000 False 300 300 1 26.303653 23.110169 60.405184 19.015698 2.285248
True 300 300 1 26.419481 23.173033 60.542022 19.144721 2.251986
ssd-mobilenet-v1-quantized-mlperf cuda 1 5000 False 300 300 1 26.817126 23.577185 61.845886 19.045377 2.524694
True 300 300 1 26.818268 23.551670 61.963231 19.108292 2.419133
ssd-resnet50-fpn cuda 1 5000 False 640 640 1 45.780129 38.340936 68.218358 51.151692 19.788320
True 640 640 1 45.894732 38.380971 69.016501 51.384803 19.662063
ssdlite-mobilenet-v2 cuda 1 5000 False 300 300 1 27.226449 24.283266 64.482049 18.944375 2.304731
True 300 300 1 27.094365 24.130926 63.986969 18.824788 2.281154
yolo-v3 cuda 1 5000 False 416 416 1 32.493981 28.532508 52.285166 34.432202 13.186537
repetition_id avg_fps avg_time_ms graph_load_time_ms images_load_time_avg_ms
model backend batch_size batch_count batch_enabled image_height image_width num_reps
rcnn-inception-resnet-v2-lowproposals cpu 1 2 True 1024 600 10 0 0.475711 2102.114439 4840.737104 9.465098
10 1 0.476848 2097.102880 4681.562185 9.582043
10 2 0.473222 2113.173485 4745.374680 9.340763
10 3 0.476904 2096.859932 4693.865299 9.480476
10 4 0.482557 2072.292805 4820.402861 9.406209
10 5 0.471203 2122.228384 4716.161489 9.643555
10 6 0.475858 2101.465940 4778.688192 9.382725
10 7 0.481224 2078.035831 4674.257755 9.570599
10 8 0.479677 2084.735870 4727.047205 9.313703
10 9 0.475473 2103.169441 4704.491615 9.448886
2 2 True 1024 600 10 0 0.469123 2131.637096 4871.593714 7.836342
10 1 0.472183 2117.823720 4632.369041 7.794440
10 2 0.471514 2120.829582 4661.816835 7.792830
10 3 0.472720 2115.418553 4615.066528 7.801950
10 4 0.466531 2143.481016 4670.040846 7.791817
10 5 0.474493 2107.514739 4742.186785 7.782817
10 6 0.473662 2111.207962 4678.366899 7.603705
10 7 0.471409 2121.299028 4791.217089 7.847309
10 8 0.468963 2132.363558 4789.848566 7.508039
10 9 0.476291 2099.555254 4646.384478 7.767916
4 2 True 1024 600 10 0 0.465838 2146.670341 4637.297392 7.425994
10 1 0.459510 2176.231623 4771.982908 7.702440
10 2 0.457189 2187.279761 4693.547964 7.557511
10 3 0.463155 2159.104466 4686.470509 7.461697
10 4 0.459437 2176.578999 4653.425932 7.388800
10 5 0.461189 2168.306887 4649.534702 7.405192
10 6 0.459851 2174.616516 4677.816391 7.625759
10 7 0.460292 2172.532856 4699.117422 7.416129
10 8 0.458029 2183.267474 4749.050379 7.668465
10 9 0.457250 2186.987460 4728.041887 7.394373
8 2 True 1024 600 10 0 0.448426 2230.021060 4881.022930 6.358564
10 1 0.447450 2234.884620 4833.306551 6.405845
10 2 0.447692 2233.678490 4714.081526 6.411314
10 3 0.449703 2223.688662 4727.843761 6.498590
10 4 0.446264 2240.827024 4821.063042 6.466031
10 5 0.447525 2234.509915 4763.653517 6.378964
10 6 0.446304 2240.626216 4656.491518 6.479472
10 7 0.448708 2228.622943 4701.173306 6.423131
10 8 0.447020 2237.035245 4839.536905 6.500378
10 9 0.446514 2239.570051 4725.246906 6.514519
16 2 True 1024 600 10 0 0.443040 2257.131487 4761.220932 6.160311
10 1 0.444964 2247.375011 4645.025969 6.135449
10 2 0.445719 2243.564099 4739.571333 6.117813
10 3 0.445926 2242.522523 4691.355467 6.323762
10 4 0.444730 2248.555779 4867.135763 6.114915
10 5 0.441781 2263.564557 4699.393034 6.088033
10 6 0.442053 2262.172401 4729.875088 6.122828
10 7 0.443121 2256.722227 4730.833530 6.306559
10 8 0.441437 2265.327543 4706.488609 6.191246
10 9 0.442450 2260.142982 4633.033752 6.211273
32 2 True 1024 600 10 0 0.441659 2264.188550 7046.988964 5.898494
10 1 0.443310 2255.756468 4752.347708 6.041717
10 2 0.441976 2262.563989 4757.130861 5.883444
10 3 0.443412 2255.240895 4838.208675 5.935721
10 4 0.442814 2258.284524 4626.413584 5.847793
10 5 0.441853 2263.196304 4953.391552 5.801924
10 6 0.443172 2256.461382 4938.735247 6.233998
10 7 0.442252 2261.153892 4810.835838 5.887762
10 8 0.441839 2263.269097 4680.061817 5.924489
10 9 0.443198 2256.326482 4646.239042 6.009314
cpu-prebuilt 1 2 True 1024 600 10 0 0.476430 2098.946333 6911.992073 9.578943
10 1 0.479329 2086.250067 4802.347898 9.509683
10 2 0.479133 2087.101221 4653.066397 9.565115
10 3 0.480091 2082.938433 4607.165098 9.295821
10 4 0.477704 2093.345881 4627.895355 9.655595
10 5 0.478506 2089.837074 4740.164995 9.519935
10 6 0.473604 2111.468315 4751.509905 9.440780
10 7 0.475716 2102.094650 4671.503067 9.498119
10 8 0.476884 2096.944571 4664.308548 9.598494
10 9 0.476331 2099.380016 4684.918880 9.502530
2 2 True 1024 600 10 0 0.468431 2134.788275 4732.982874 7.852614
10 1 0.472935 2114.454389 4649.711609 7.798254
10 2 0.470190 2126.798034 4643.195629 7.697940
10 3 0.471013 2123.083591 4773.692131 7.840872
10 4 0.475274 2104.051471 4850.495815 7.591724
10 5 0.472168 2117.888451 4685.405493 7.899523
10 6 0.468009 2136.710882 4751.292706 7.935166
10 7 0.472980 2114.253998 4710.348606 8.239985
10 8 0.474203 2108.802080 4819.133282 7.782161
10 9 0.473089 2113.765717 4710.736036 8.140326
4 2 True 1024 600 10 0 0.456413 2190.999448 4674.053669 7.327020
10 1 0.456003 2192.967653 4681.468010 7.452399
10 2 0.459294 2177.252531 4655.152798 7.265180
10 3 0.461772 2165.570319 4762.456894 7.626474
10 4 0.461604 2166.359723 4708.060980 7.452428
10 5 0.460823 2170.032084 4666.865587 7.605553
10 6 0.459412 2176.693141 4695.546150 7.397562
10 7 0.460089 2173.491180 4695.769548 7.499397
10 8 0.460982 2169.284046 4728.900194 7.480860
10 9 0.458869 2179.272771 4866.041183 7.490665
8 2 True 1024 600 10 0 0.445128 2246.544719 4824.887991 6.349862
10 1 0.446106 2241.620868 4814.151287 6.454602
10 2 0.446176 2241.268754 4751.685143 6.356791
10 3 0.445251 2245.923698 4764.530420 6.336719
10 4 0.446161 2241.344512 4746.541023 6.455719
10 5 0.446886 2237.708509 4695.195913 6.394356
10 6 0.446231 2240.990847 4730.701923 6.395116
10 7 0.444921 2247.592241 4701.094151 6.470054
10 8 0.444265 2250.911266 4799.312592 6.382376
10 9 0.444193 2251.275152 4890.583754 6.191820
16 2 True 1024 600 10 0 0.442694 2258.895636 4723.585844 6.115980
10 1 0.443003 2257.322982 4619.927168 6.121911
10 2 0.443031 2257.177547 4690.367460 6.208442
10 3 0.442958 2257.551774 4842.099667 6.250903
10 4 0.441984 2262.527600 4723.618031 6.095320
10 5 0.441009 2267.529160 4709.165812 6.330647
10 6 0.443174 2256.451771 4732.079029 6.190538
10 7 0.443248 2256.072521 4688.783884 6.279618
10 8 0.442761 2258.555487 4755.360603 6.120138
10 9 0.443000 2257.334948 4673.398972 6.245270
32 2 True 1024 600 10 0 0.445693 2243.696325 4708.714724 5.976126
10 1 0.444501 2249.715857 4799.293756 5.980864
10 2 0.443733 2253.608018 4812.208176 5.862933
10 3 0.444597 2249.227196 4619.879246 5.906854
10 4 0.442427 2260.260180 4708.615303 5.930688
10 5 0.442084 2262.011275 4752.560377 6.009258
10 6 0.441787 2263.532028 4742.196798 6.043244
10 7 0.444153 2251.477160 4678.473949 6.093286
10 8 0.443272 2255.950890 4655.450106 5.904902
10 9 0.443241 2256.108768 4884.309292 6.025005
cuda 1 2 True 1024 600 10 0 3.312149 301.918745 4553.108692 9.642363
10 1 3.239443 308.695078 4603.839397 9.423256
10 2 3.318541 301.337242 4610.033751 9.488583
10 3 3.272390 305.587053 4625.099421 9.373426
10 4 3.288376 304.101467 4622.536659 9.427190
10 5 3.278278 305.038214 4729.156017 9.627342
10 6 3.244148 308.247328 4665.069103 9.451509
10 7 3.312236 301.910877 4691.882610 9.592414
10 8 3.261540 306.603670 4566.406250 9.423494
10 9 3.224632 310.112953 4607.881784 9.596944
2 2 True 1024 600 10 0 3.552907 281.459689 4668.402672 7.642150
10 1 3.518340 284.224987 4657.329798 7.844627
10 2 3.532334 283.098936 4700.743914 7.920146
10 3 3.539500 282.525778 4598.049879 7.964015
10 4 3.500998 285.632849 4626.263142 7.676423
10 5 3.551309 281.586289 4683.004141 7.845938
10 6 3.545740 282.028556 4633.757591 7.518411
10 7 3.523758 283.787966 4607.816696 7.668674
10 8 3.550711 281.633735 4610.222101 7.816374
10 9 3.518227 284.234047 4659.436464 7.707119
4 2 True 1024 600 10 0 3.674564 272.141159 4589.384079 7.491559
10 1 3.675187 272.094965 4600.687265 7.456094
10 2 3.668294 272.606254 4670.823097 7.364601
10 3 3.659768 273.241341 4663.695574 7.741481
10 4 3.680945 271.669388 4619.314671 7.494420
10 5 3.701009 270.196557 4702.200174 7.385522
10 6 3.685486 271.334648 4649.031162 7.692933
10 7 3.695861 270.572960 4589.505911 7.417113
10 8 3.682430 271.559834 4730.802298 7.475913
10 9 3.688631 271.103263 4620.339155 7.459700
8 2 True 1024 600 10 0 3.728697 268.190205 4689.132690 6.411090
10 1 3.733332 267.857254 4633.130074 6.382719
10 2 3.721300 268.723279 4610.522270 6.602168
10 3 3.712861 269.334078 4680.472374 6.342486
10 4 3.729187 268.154949 4663.403511 6.322131
10 5 3.715667 269.130677 4674.349308 6.383047
10 6 3.735976 267.667681 4610.460997 6.335691
10 7 3.728257 268.221825 4592.166662 6.378040
10 8 3.714321 269.228220 4643.437147 6.308481
10 9 3.718817 268.902749 4688.884020 6.420180
16 2 True 1024 600 10 0 3.713056 269.319922 4676.514626 6.128848
10 1 3.710434 269.510269 4574.917793 6.112933
10 2 3.715925 269.112006 4717.499256 6.175146
10 3 3.717626 268.988878 4645.964146 6.201915
10 4 3.717039 269.031346 4708.994865 6.256603
10 5 3.714478 269.216806 4657.868862 6.165959
10 6 3.704128 269.969091 4651.146889 6.379701
10 7 3.693570 270.740747 4633.554697 6.231189
10 8 3.710271 269.522071 4608.161926 6.170891
10 9 3.704759 269.923061 4632.548571 6.360725
tensorrt 1 2 True 1024 600 10 0 3.376907 296.128988 71773.906469 9.302139
10 1 3.348232 298.665047 71333.559275 9.582281
10 2 3.366134 297.076702 71861.565590 9.280920
10 3 3.369728 296.759844 71676.164389 9.327054
10 4 3.351475 298.376083 71406.909466 9.361982
10 5 3.369701 296.762228 71786.504984 9.517074
10 6 3.382741 295.618296 72182.452917 9.416580
10 7 3.356584 297.921896 72012.970686 9.425163
10 8 3.342434 299.183130 72414.334059 9.431243
10 9 3.362265 297.418594 71405.121565 9.265065
2 2 True 1024 600 10 0 3.598338 277.906060 72460.325241 7.761598
10 1 3.619488 276.282191 73288.720846 7.611036
10 2 3.625512 275.823116 73017.344475 7.620037
10 3 3.600910 277.707577 71886.828661 7.649243
10 4 3.609189 277.070522 72389.738083 7.569075
10 5 3.614754 276.643991 73603.370428 7.697105
10 6 3.626780 275.726676 72315.528631 7.667959
10 7 3.610072 277.002811 71280.921936 7.760704
10 8 3.603602 277.500153 72457.260370 7.785320
10 9 3.616836 276.484728 72930.105448 7.605672
4 2 True 1024 600 10 0 3.749659 266.690910 63406.803846 7.282495
10 1 3.770994 265.182078 62365.354538 7.390857
10 2 3.764669 265.627623 63301.706314 7.337838
10 3 3.779923 264.555633 62900.637388 7.436097
10 4 3.758543 266.060531 62595.251083 7.462770
10 5 3.739821 267.392457 62498.919249 7.585466
10 6 3.760241 265.940428 63171.914101 7.459760
10 7 3.767618 265.419662 63504.255295 7.384747
10 8 3.765990 265.534461 62592.798471 7.361025
10 9 3.764941 265.608430 63832.051754 7.310838
8 2 True 1024 600 10 0 3.790856 263.792664 63456.116199 6.390765
10 1 3.792566 263.673693 63586.563826 6.568715
10 2 3.796260 263.417184 62492.844582 6.305456
10 3 3.786046 264.127791 63372.258663 6.388098
10 4 3.813703 262.212336 62674.717426 6.489903
10 5 3.796162 263.423920 63579.843760 6.310716
10 6 3.815625 262.080282 61867.995501 6.435558
10 7 3.782634 264.366031 64045.851231 6.348684
10 8 3.802486 262.985826 62815.172434 6.436229
10 9 3.805993 262.743503 62830.606461 6.401584
16 2 True 1024 600 10 0 3.809337 262.512863 63666.782379 6.077066
10 1 3.801270 263.069987 62935.933828 6.170094
10 2 3.789768 263.868377 64053.315401 6.218269
10 3 3.800880 263.096988 63089.780092 6.116427
10 4 3.798629 263.252839 63936.655283 6.117716
10 5 3.793507 263.608307 62832.702875 6.145746
10 6 3.674101 272.175416 62334.528685 6.276503
10 7 3.796823 263.378084 63668.829679 6.201357
10 8 3.811932 262.334198 63301.749706 6.155752
10 9 3.799858 263.167739 63519.116163 6.351136
rcnn-inception-v2 cpu 1 2 True 600 1024 10 0 5.014147 199.435711 1498.870850 9.715676
10 1 5.042867 198.299885 1505.791903 9.608984
10 2 4.946498 202.163219 1473.861456 10.332584
10 3 4.990290 200.389147 1584.379911 10.267496
10 4 4.931754 202.767611 1560.732603 9.686112
10 5 5.009362 199.626207 1587.606430 9.686351
10 6 4.903904 203.919172 1524.977446 9.994268
10 7 5.082360 196.758986 1501.475334 9.787917
10 8 5.036479 198.551416 1551.583290 9.751916
10 9 4.905659 203.846216 1501.851797 9.498477
2 2 True 600 1024 10 0 5.272674 189.657092 1466.687441 7.930219
10 1 5.335643 187.418818 1585.123062 8.521616
10 2 5.316342 188.099265 1564.975500 7.825553
10 3 5.268346 189.812899 1525.840521 7.982373
10 4 5.404975 185.014725 1541.822433 8.040488
10 5 5.325708 187.768459 1537.496328 8.111715
10 6 5.331035 187.580824 1607.749939 8.147657
10 7 5.166681 193.547845 1590.650558 8.406579
10 8 5.356324 186.695218 1587.227345 8.009434
10 9 5.272217 189.673543 1561.925173 8.123875
4 2 True 600 1024 10 0 5.448908 183.522999 1501.704931 7.514775
10 1 5.369132 186.249852 1519.568443 7.727951
10 2 5.485446 182.300568 1513.799906 7.464796
10 3 5.385897 185.670078 1560.333729 7.785231
10 4 5.409991 184.843183 1547.524691 7.684201
10 5 5.359472 186.585546 1524.288654 7.822454
10 6 5.348534 186.967134 1518.525362 7.608354
10 7 5.386990 185.632408 1504.456282 7.741630
10 8 5.377758 185.951114 1522.876978 7.642746
10 9 5.329975 187.618136 1515.112877 7.878870
8 2 True 600 1024 10 0 5.349165 186.945051 1576.133728 6.847322
10 1 5.433229 184.052616 1562.856674 6.624192
10 2 5.406175 184.973657 1570.407391 6.713390
10 3 5.411611 184.787869 1529.011250 6.627277
10 4 5.377932 185.945094 1595.106125 6.435454
10 5 5.338694 187.311739 1603.112698 6.564945
10 6 5.435720 183.968276 1525.926113 6.634235
10 7 5.293217 188.921034 1562.609196 6.491125
10 8 5.461332 183.105528 1506.470919 6.450713
10 9 5.408010 184.910893 1497.475624 6.719351
16 2 True 600 1024 10 0 5.954560 167.938516 1557.972193 6.492734
10 1 5.845675 171.066642 1513.023138 6.231360
10 2 5.926130 168.744192 1549.762249 6.216884
10 3 5.939444 168.365926 1494.459629 6.225139
10 4 5.939874 168.353751 1589.697361 6.574623
10 5 5.921630 168.872416 1563.881397 6.486021
10 6 5.886893 169.868901 1493.456364 6.590150
10 7 5.919163 168.942794 1541.707277 6.275326
10 8 5.886155 169.890180 1614.044666 6.533414
10 9 5.948207 168.117896 1519.454002 6.238572
32 2 True 600 1024 10 0 5.867390 170.433529 2000.246286 5.991984
10 1 5.872519 170.284688 1568.295002 6.092377
10 2 5.864556 170.515902 1531.263351 6.254148
10 3 5.855742 170.772560 1511.621475 6.313439
10 4 5.863773 170.538656 1523.082495 6.220240
10 5 5.846251 171.049796 1498.371840 6.043233
10 6 5.807043 172.204696 1503.453970 6.258313
10 7 5.836637 171.331540 1472.190380 6.224837
10 8 5.823156 171.728186 1582.651854 6.172523
10 9 5.840139 171.228804 1499.148607 5.944114
cpu-prebuilt 1 2 True 600 1024 10 0 4.731297 211.358547 1956.808805 9.831071
10 1 4.785440 208.967209 1418.408871 9.634018
10 2 4.714778 212.099075 1499.061346 9.669900
10 3 4.635287 215.736389 1422.900200 9.493232
10 4 4.725668 211.610317 1446.428299 9.874105
10 5 4.667941 214.227200 1424.752712 9.541869
10 6 4.722773 211.740017 1425.207853 9.795189
10 7 4.651227 214.997053 1425.072908 9.863138
10 8 4.666653 214.286327 1459.944725 10.440350
10 9 4.664204 214.398861 1451.465368 9.746313
2 2 True 600 1024 10 0 5.019488 199.223518 1448.516369 8.101106
10 1 4.988228 200.471997 1423.604965 7.747293
10 2 5.131576 194.871902 1449.007034 7.976055
10 3 5.084748 196.666598 1437.984705 7.960260
10 4 5.058507 197.686791 1450.055122 8.035362
10 5 5.166586 193.551421 1482.403040 7.938087
10 6 5.127191 195.038557 1474.911928 8.123815
10 7 5.093358 196.334124 1440.686941 8.092284
10 8 4.970026 201.206207 1430.893183 7.984698
10 9 5.016672 199.335337 1441.020012 7.787883
4 2 True 600 1024 10 0 5.132435 194.839299 1462.011337 7.741839
10 1 5.169760 193.432570 1556.849003 7.588327
10 2 5.116872 195.431888 1509.188175 7.510215
10 3 5.180694 193.024337 1464.471579 7.693589
10 4 5.169763 193.432450 1450.300455 7.868409
10 5 5.223467 191.443741 1451.604366 7.518381
10 6 5.092083 196.383297 1449.978352 7.638454
10 7 5.120138 195.307255 1467.447281 7.812858
10 8 5.224269 191.414356 1443.993807 7.774144
10 9 5.209573 191.954315 1465.530634 7.917315
8 2 True 600 1024 10 0 5.211387 191.887498 1460.234880 6.753191
10 1 5.159525 193.816274 1445.772171 6.473005
10 2 5.188005 192.752331 1429.254770 6.613567
10 3 5.072365 197.146684 1428.463459 6.731093
10 4 5.132520 194.836080 1440.501451 6.570667
10 5 5.213028 191.827089 1431.637764 6.464541
10 6 5.189467 192.698032 1463.212490 6.386876
10 7 5.161262 193.751067 1430.485964 6.712347
10 8 5.206055 192.084014 1523.363590 6.445169
10 9 5.188151 192.746907 1486.027479 6.785393
16 2 True 600 1024 10 0 5.687760 175.816134 1450.564861 6.334804
10 1 5.649171 177.017137 1505.919218 6.413735
10 2 5.678731 176.095694 1453.777075 6.276630
10 3 5.713200 175.033256 1452.397585 6.360650
10 4 5.671152 176.331028 1435.543537 6.207824
10 5 5.663570 176.567063 1430.095434 6.297491
10 6 5.634694 177.471921 1447.620392 6.262504
10 7 5.656261 176.795244 1424.173355 6.572500
10 8 5.683345 175.952703 1447.904348 6.459363
10 9 5.674009 176.242232 1467.178345 6.471470
32 2 True 600 1024 10 0 5.609742 178.261332 1435.495138 5.966764
10 1 5.630896 177.591622 1439.628124 6.052073
10 2 5.626136 177.741870 1482.169390 6.120678
10 3 5.601327 178.529128 1444.671154 6.279349
10 4 5.624987 177.778199 1466.797113 6.110840
10 5 5.621275 177.895591 1456.978559 6.175332
10 6 5.631302 177.578814 1439.316511 6.169483
10 7 5.643570 177.192807 1456.930637 6.114166
10 8 5.649665 177.001648 1444.205761 6.016966
10 9 5.616136 178.058371 1442.853689 6.239720
cuda 1 2 True 600 1024 10 0 18.356620 54.476261 1416.687250 9.582996
10 1 18.061380 55.366755 1434.233189 9.861350
10 2 17.978157 55.623055 1429.165602 10.132551
10 3 18.031176 55.459499 1437.551975 9.672880
10 4 18.085601 55.292606 1444.493055 9.456038
10 5 17.938175 55.747032 1448.664904 9.703994
10 6 18.289861 54.675102 1463.722229 9.909034
10 7 18.024202 55.480957 1454.885721 9.685397
10 8 17.665583 56.607246 1457.566261 10.692358
10 9 17.817093 56.125879 1427.454233 9.729981
2 2 True 600 1024 10 0 22.388726 44.665337 1432.064295 8.013248
10 1 22.363598 44.715524 1457.754374 8.409679
10 2 22.093308 45.262575 1428.236723 8.197665
10 3 22.525135 44.394851 1435.927868 7.975638
10 4 22.912057 43.645144 1431.225300 7.717252
10 5 22.225364 44.993639 1423.159599 7.735074
10 6 22.307221 44.828534 1454.227448 8.231044
10 7 22.606842 44.234395 1440.490723 8.176088
10 8 21.805697 45.859575 1473.582029 7.986307
10 9 22.413911 44.615149 1448.654175 8.064926
4 2 True 600 1024 10 0 24.050839 41.578591 1468.189001 7.651746
10 1 23.947507 41.758001 1457.059383 7.923931
10 2 25.217937 39.654315 1450.780630 7.563680
10 3 24.222864 41.283309 1444.934607 7.651985
10 4 24.246003 41.243911 1442.461967 7.494748
10 5 25.259550 39.588988 1447.603941 7.571995
10 6 24.393750 40.994108 1461.820126 7.508159
10 7 24.508888 40.801525 1444.098473 7.941574
10 8 25.118075 39.811969 1459.584475 7.815182
10 9 23.906694 41.829288 1459.226847 7.691115
8 2 True 600 1024 10 0 25.647059 38.990825 1424.114227 6.726891
10 1 25.092116 39.853156 1445.759535 6.453142
10 2 25.419004 39.340645 1446.643353 6.523430
10 3 25.541015 39.152712 1458.364010 6.682336
10 4 25.473944 39.255798 1430.957079 6.775945
10 5 25.806734 38.749576 1454.945564 6.586790
10 6 25.963829 38.515121 1448.742151 6.435499
10 7 25.178860 39.715856 1445.089817 6.762639
10 8 25.524986 39.177299 1455.610275 6.430954
10 9 25.785197 38.781941 1440.744162 6.651729
16 2 True 600 1024 10 0 25.987668 38.479790 1426.766396 6.466702
10 1 25.034530 39.944828 1419.351578 6.508283
10 2 25.371050 39.415002 1438.000917 6.564394
10 3 25.619527 39.032727 1467.727900 6.300062
10 4 25.525073 39.177164 1437.742710 6.188750
10 5 25.435363 39.315343 1448.985815 6.509133
10 6 25.863934 38.663879 1453.492165 6.320909
10 7 25.066211 39.894342 1441.774607 6.310403
10 8 25.542512 39.150417 1431.310892 6.462514
10 9 25.422470 39.335281 1459.959507 6.622389
32 2 True 600 1024 10 0 25.799874 38.759880 1449.286938 6.259121
10 1 25.739995 38.850047 1447.102070 6.001525
10 2 25.682806 38.936555 1433.722258 6.029718
10 3 25.064315 39.897360 1440.608501 6.020978
10 4 25.591951 39.074786 1452.372074 6.021012
10 5 25.341169 39.461479 1455.724001 5.995147
10 6 25.481194 39.244629 1432.259560 6.000403
10 7 25.517978 39.188057 1446.592093 6.137870
10 8 25.394582 39.378479 1452.353001 6.222215
10 9 25.738702 38.851999 1431.207895 6.265253
tensorrt 1 2 True 600 1024 10 0 18.262863 54.755926 17054.683208 9.464860
10 1 17.575053 56.898832 17040.428877 9.784460
10 2 18.187157 54.983854 16899.039745 9.646773
10 3 17.890735 55.894852 16674.649715 9.531498
10 4 17.856993 56.000471 16875.169754 9.568095
10 5 17.947232 55.718899 16942.353725 9.413719
10 6 17.661938 56.618929 17089.957237 9.646893
10 7 18.166047 55.047750 17048.625946 9.444356
10 8 17.906699 55.845022 16540.392637 9.705305
10 9 18.136433 55.137634 17052.166224 9.509802
2 2 True 600 1024 10 0 21.331977 46.877980 17464.916706 7.911503
10 1 22.205243 45.034409 17496.206284 7.734179
10 2 22.556570 44.332981 17304.871798 7.731199
10 3 21.681144 46.123028 17672.189474 7.677436
10 4 21.953500 45.550823 17278.563499 7.756829
10 5 21.282295 46.987414 17386.888504 7.874489
10 6 21.715999 46.048999 17380.331516 7.769227
10 7 21.653329 46.182275 17525.537491 7.760108
10 8 21.685740 46.113253 17375.267029 7.783175
10 9 21.975642 45.504928 17431.662321 7.817864
4 2 True 600 1024 10 0 23.671323 42.245209 13168.403149 7.465363
10 1 24.453910 40.893257 12983.771563 7.571787
10 2 24.499725 40.816784 12947.808743 7.530928
10 3 24.303007 41.147172 12901.455641 7.413596
10 4 24.247440 41.241467 13094.930172 7.633865
10 5 24.366717 41.039586 12983.961821 7.611096
10 6 24.121311 41.457117 13073.016167 7.443249
10 7 24.264378 41.212678 12953.689337 7.479340
10 8 24.217584 41.292310 13024.133205 7.487029
10 9 24.309592 41.136026 12940.512657 7.497519
8 2 True 600 1024 10 0 25.155812 39.752245 12938.249826 6.490529
10 1 25.961157 38.519084 12916.830063 6.556660
10 2 25.765259 38.811952 12981.657505 6.506383
10 3 25.533183 39.164722 12793.953180 6.464794
10 4 25.083525 39.866805 12827.604294 6.476507
10 5 26.215690 38.145095 12949.955940 6.514266
10 6 24.136615 41.430831 13063.522577 6.440580
10 7 25.839187 38.700908 13103.403330 6.555989
10 8 25.376557 39.406449 12828.648806 6.730393
10 9 25.413209 39.349616 13046.003580 6.464243
16 2 True 600 1024 10 0 24.577518 40.687591 12772.939444 6.251216
10 1 25.484130 39.240107 12845.211029 6.521888
10 2 25.402281 39.366543 12757.096291 6.329700
10 3 25.397292 39.374277 12854.661465 6.233133
10 4 25.350627 39.446756 12700.048208 6.255448
10 5 25.673659 38.950428 12906.770468 6.240763
10 6 25.323058 39.489701 12794.317722 6.543636
10 7 25.715884 38.886473 12839.040041 6.350070
10 8 25.261461 39.585993 12848.332882 6.533489
10 9 25.589750 39.078146 12715.066671 6.267227
32 2 True 600 1024 10 0 25.458072 39.280273 12844.294071 6.059401
10 1 25.299764 39.526060 12783.873081 6.105229
10 2 25.604283 39.055966 12953.673840 6.216351
10 3 25.667001 38.960531 12744.321346 6.073479
10 4 25.410101 39.354429 12983.851194 6.130971
10 5 25.206263 39.672680 12954.895973 6.046593
10 6 25.566489 39.113700 12904.195309 6.289415
10 7 25.532818 39.165281 12871.153831 6.005283
10 8 25.389638 39.386146 12927.368402 6.088052
10 9 25.302254 39.522171 12840.082169 6.215300
tensorrt-dynamic 1 2 True 600 1024 10 0 19.471711 51.356554 8882.416964 9.399652
10 1 19.488358 51.312685 8742.156982 9.500861
10 2 19.741153 50.655603 8809.425831 9.415388
10 3 20.174040 49.568653 8727.215528 9.346485
10 4 19.717859 50.715446 8771.088362 9.436131
10 5 19.789120 50.532818 8779.867887 9.456873
10 6 19.347849 51.685333 8846.498489 9.612441
10 7 19.634232 50.931454 8748.760462 9.413242
10 8 20.592616 48.561096 8759.370089 9.451628
10 9 19.604223 51.009417 8846.726894 9.759068
2 2 True 600 1024 10 0 24.902283 40.156960 8845.126629 7.715225
10 1 25.927977 38.568377 8732.381105 7.721901
10 2 26.110610 38.298607 8835.674286 7.717729
10 3 24.664977 40.543318 8861.158371 7.768750
10 4 25.611802 39.044499 8831.889391 7.829726
10 5 25.324558 39.487362 8809.125185 7.814646
10 6 25.166000 39.736152 8763.042688 7.672012
10 7 26.793816 37.322044 8699.295521 8.072317
10 8 25.743851 38.844228 8853.743315 7.687807
10 9 25.975426 38.497925 8777.847528 7.730305
4 2 True 600 1024 10 0 27.345878 36.568582 8896.489620 7.618845
10 1 26.674027 37.489653 8785.809994 7.472098
10 2 26.936552 37.124276 9066.257954 7.429898
10 3 27.555808 36.289990 8813.172817 7.455707
10 4 27.355153 36.556184 8797.575474 7.526100
10 5 27.546443 36.302328 8893.589020 7.406443
10 6 27.583625 36.253393 8820.943117 7.642150
10 7 27.066007 36.946714 8782.052517 7.625014
10 8 26.869900 37.216365 8864.279985 7.550091
10 9 27.119075 36.874413 8867.530107 7.411540
8 2 True 600 1024 10 0 28.623931 34.935802 8802.629948 6.562367
10 1 28.367541 35.251558 8795.866966 6.728917
10 2 27.900761 35.841316 8880.637407 6.499916
10 3 28.437188 35.165220 8802.983761 6.835207
10 4 28.814652 34.704566 8907.945871 6.456673
10 5 28.359725 35.261273 8850.032091 6.420612
10 6 28.682311 34.864694 8850.908995 6.408900
10 7 28.899303 34.602910 8853.540421 6.455719
10 8 28.408875 35.200268 8787.514210 6.467193
10 9 29.035769 34.440279 8714.293718 6.656498
16 2 True 600 1024 10 0 29.717633 33.650056 8770.996571 6.254293
10 1 29.310185 34.117833 8788.775206 6.225012
10 2 29.064026 34.406796 8714.112520 6.343611
10 3 29.366414 34.052506 8808.066130 6.363176
10 4 29.439665 33.967778 8803.171635 6.349988
10 5 29.161981 34.291223 8801.988602 6.538227
10 6 29.290625 34.140617 8902.085066 6.226599
10 7 29.781423 33.577979 8854.831934 6.227456
10 8 29.434706 33.973500 8847.275496 6.330125
10 9 29.232406 34.208611 8767.259359 6.279066
rcnn-nas-lowproposals cpu 1 2 True 1200 1200 10 0 0.422802 2365.170717 2737.598181 9.857059
10 1 0.424297 2356.838703 2750.021696 10.446787
10 2 0.425269 2351.453304 2692.065239 9.801030
10 3 0.424194 2357.411623 2690.779448 9.919405
10 4 0.422270 2368.151188 2773.290396 9.843230
10 5 0.422163 2368.752956 2624.770880 9.850144
10 6 0.425744 2348.830462 2733.825207 10.105371
10 7 0.425861 2348.184109 2787.533522 10.225415
10 8 0.424353 2356.526852 2670.963526 9.792089
10 9 0.418898 2387.213230 2661.247730 10.092616
2 2 True 1200 1200 10 0 0.426305 2345.739245 2773.810387 7.925332
10 1 0.428993 2331.039786 2669.348717 8.184373
10 2 0.424239 2357.163906 2642.301559 8.188963
10 3 0.424178 2357.503533 2726.520061 8.347154
10 4 0.425631 2349.451661 2632.941961 8.068979
10 5 0.425474 2350.321770 2602.002144 8.147836
10 6 0.426202 2346.306801 2581.935644 8.292913
10 7 0.422632 2366.124988 2635.398865 8.129060
10 8 0.423764 2359.805942 2767.369032 8.429885
10 9 0.426602 2344.103336 2670.548439 8.455515
4 2 True 1200 1200 10 0 0.413391 2419.015110 2719.013691 7.812947
10 1 0.383810 2605.458200 2780.741453 8.239448
10 2 0.410710 2434.805334 2718.132496 7.932514
10 3 0.408973 2445.148945 2635.082483 7.753611
10 4 0.404520 2472.068489 2686.142921 8.064926
10 5 0.407546 2453.712821 2728.694439 7.603019
10 6 0.407012 2456.929028 2756.821156 7.875770
10 7 0.414891 2410.270452 2732.528687 7.674634
10 8 0.403977 2475.385547 2655.443192 7.812589
10 9 0.410397 2436.666965 2581.789732 7.780552
8 2 True 1200 1200 10 0 0.409692 2440.859675 2723.955870 6.748095
10 1 0.408224 2449.632823 2662.474394 6.819546
10 2 0.409923 2439.483613 2782.727242 6.641164
10 3 0.410774 2434.425950 2610.941410 6.880209
10 4 0.410323 2437.106162 2621.670246 6.747261
10 5 0.408022 2450.850636 2767.273664 6.716713
10 6 0.407304 2455.170274 2680.396080 6.646156
10 7 0.408581 2447.495490 2784.536600 6.836891
10 8 0.406530 2459.843129 2648.112774 6.584793
10 9 0.410899 2433.690846 2724.629641 6.609336
16 2 True 1200 1200 10 0 0.347720 2875.881150 2602.493048 6.734543
10 1 0.345282 2896.181986 2752.411366 6.995350
10 2 0.341455 2928.640544 2760.817528 6.562606
10 3 0.344686 2901.188508 2695.853710 6.522126
10 4 0.337652 2961.632535 2750.416994 6.470360
10 5 0.338795 2951.634616 2767.542124 6.454885
10 6 0.340214 2939.327523 2680.156946 6.556422
10 7 0.337539 2962.623239 2645.760298 6.512657
10 8 0.336814 2968.998060 2791.553736 6.606594
10 9 0.338992 2949.924231 2613.782883 6.593280
32 2 True 1200 1200 10 0 0.342457 2920.075461 2630.548000 6.267168
10 1 0.345298 2896.047853 2688.624620 6.448139
10 2 0.347817 2875.072852 2706.780910 6.241683
10 3 0.343870 2908.073343 2681.253910 6.221186
10 4 0.346743 2883.977517 2751.889467 6.322131
10 5 0.348021 2873.386696 2747.138500 6.238777
10 6 0.352098 2840.122133 2750.693798 6.202526
10 7 0.351400 2845.761634 2763.027191 6.425463
10 8 0.349093 2864.565723 2688.318491 6.224327
10 9 0.347885 2874.511279 2760.948896 6.199230
cpu-prebuilt 1 2 True 1200 1200 10 0 0.412067 2426.792383 6361.849070 9.966373
10 1 0.415427 2407.164574 2569.520712 10.451794
10 2 0.412061 2426.827908 2573.592901 9.870052
10 3 0.412336 2425.205469 2561.060429 9.855032
10 4 0.411725 2428.805351 2551.715851 9.871840
10 5 0.410960 2433.329105 2559.540749 9.883046
10 6 0.411715 2428.862333 2562.439680 9.879351
10 7 0.412013 2427.105427 2617.561102 9.896278
10 8 0.410353 2436.924219 2543.494463 9.835482
10 9 0.417485 2395.296812 2572.293997 9.902477
2 2 True 1200 1200 10 0 0.411213 2431.826830 2644.977331 7.978022
10 1 0.414424 2412.988544 2561.379671 8.155406
10 2 0.412171 2426.176071 2534.359455 8.223176
10 3 0.413062 2420.944810 2574.661016 7.916808
10 4 0.415174 2408.629894 2594.500303 8.090079
10 5 0.408385 2448.668957 2567.672968 8.264244
10 6 0.410635 2435.251117 2579.747677 8.628190
10 7 0.412063 2426.814079 2595.364809 8.284450
10 8 0.413065 2420.928001 2566.778660 8.005679
10 9 0.412055 2426.860809 2530.684233 7.984638
4 2 True 1200 1200 10 0 0.408507 2447.936237 2578.824520 7.659644
10 1 0.407047 2456.720412 2606.344223 7.800460
10 2 0.406182 2461.948037 2548.595190 7.712364
10 3 0.406351 2460.924327 2597.253084 7.743061
10 4 0.406853 2457.892478 2595.927000 7.790834
10 5 0.407891 2451.637506 2562.724352 7.844478
10 6 0.407127 2456.234574 2530.983925 7.697970
10 7 0.407628 2453.219116 2589.909554 7.688671
10 8 0.403611 2477.634072 2560.188293 8.168221
10 9 0.406113 2462.365985 2535.008907 7.906646
8 2 True 1200 1200 10 0 0.400541 2496.621668 2534.344435 6.722003
10 1 0.401684 2489.517391 2566.886187 6.748512
10 2 0.401358 2491.544008 2518.408298 6.893903
10 3 0.401129 2492.961645 2573.572874 6.724015
10 4 0.402803 2482.601255 2652.946472 6.739601
10 5 0.401491 2490.716219 2643.571377 6.681859
10 6 0.402963 2481.614977 2521.253109 6.667063
10 7 0.402671 2483.415842 2550.433159 6.755248
10 8 0.401592 2490.090817 2552.996635 6.816626
10 9 0.402486 2484.557211 2571.765423 6.753311
16 2 True 1200 1200 10 0 0.387576 2580.137551 2575.905085 6.615490
10 1 0.387670 2579.514548 2548.897743 6.499007
10 2 0.388693 2572.727650 2547.498941 6.509326
10 3 0.387914 2577.890739 2567.530155 6.485753
10 4 0.387272 2582.166836 2572.117090 6.529614
10 5 0.384598 2600.119963 2588.905811 6.639086
10 6 0.386767 2585.533246 2571.714163 6.554551
10 7 0.384615 2600.002646 2643.561602 6.482288
10 8 0.383358 2608.531043 2667.171478 6.447896
10 9 0.384187 2602.899581 2691.616535 6.454766
32 2 True 1200 1200 10 0 0.354244 2822.915569 2539.497375 6.164365
10 1 0.355216 2815.188333 2579.700947 6.294392
10 2 0.358326 2790.754512 2586.631536 6.237816
10 3 0.356251 2807.010278 2562.347651 6.228499
10 4 0.356326 2806.418195 2578.271866 6.219782
10 5 0.359503 2781.621151 2526.288271 6.199598
10 6 0.357065 2800.613753 2606.225014 6.242082
10 7 0.356330 2806.383803 2601.206541 6.294459
10 8 0.357374 2798.189528 2694.718838 6.200224
10 9 0.356896 2801.935650 2544.285059 6.437883
cuda 1 2 True 1200 1200 10 0 2.771166 360.858917 2582.018852 9.901285
10 1 2.759923 362.329006 2621.037960 9.940386
10 2 2.752753 363.272667 2593.099356 9.884000
10 3 2.785760 358.968496 2569.315672 9.851217
10 4 2.765374 361.614704 2605.367422 9.914398
10 5 2.764355 361.747980 2599.323988 9.861708
10 6 2.773869 360.507250 2576.275110 10.036945
10 7 2.807295 356.214762 2588.441133 9.733438
10 8 2.758403 362.528563 2604.491472 9.893537
10 9 2.801866 356.904984 2585.087776 9.777427
2 2 True 1200 1200 10 0 2.901275 344.676018 2553.175926 8.119166
10 1 2.911759 343.435049 2586.572170 7.959247
10 2 2.884723 346.653700 2593.060255 8.024752
10 3 2.882487 346.922636 2594.633341 8.210301
10 4 2.905655 344.156504 2577.397823 8.071303
10 5 2.911228 343.497634 2599.563122 8.045614
10 6 2.887740 346.291542 2599.616051 8.206189
10 7 2.883306 346.824050 2616.064787 8.101344
10 8 2.882180 346.959591 2600.432396 8.085310
10 9 2.902249 344.560385 2605.264902 8.053899
4 2 True 1200 1200 10 0 2.939045 340.246618 2593.539000 7.703245
10 1 2.939430 340.201974 2572.237730 7.715553
10 2 2.935078 340.706408 2599.338770 7.794261
10 3 2.928367 341.487229 2598.786592 7.855088
10 4 2.941871 339.919746 2559.435368 7.808298
10 5 2.929240 341.385484 2584.117413 8.010060
10 6 2.920138 342.449546 2590.898752 7.594705
10 7 2.933626 340.875089 2647.840023 7.929742
10 8 2.939078 340.242803 2572.406054 7.688046
10 9 2.945699 339.477956 2580.096245 7.626444
8 2 True 1200 1200 10 0 2.921746 342.261106 2567.271948 6.669730
10 1 2.922858 342.130870 2594.122410 6.800830
10 2 2.920415 342.417121 2614.639044 6.576866
10 3 2.916166 342.916012 2592.293024 6.676361
10 4 2.910517 343.581617 2605.363369 6.958082
10 5 2.913586 343.219697 2568.465710 6.664857
10 6 2.917625 342.744559 2600.780010 6.701812
10 7 2.910468 343.587399 2606.662035 6.885782
10 8 2.913056 343.282133 2630.887985 6.758094
10 9 2.912372 343.362689 2594.960928 6.767809
tensorrt 1 2 True 1200 1200 10 0 2.767666 361.315250 110542.984009 10.462880
10 1 2.871997 348.189831 109657.408476 9.759068
10 2 2.852808 350.531816 110034.388065 9.696245
10 3 2.865664 348.959208 112439.590454 9.766340
10 4 2.841985 351.866722 111788.014889 9.869456
10 5 2.849151 350.981712 110608.415127 9.820223
10 6 2.840501 352.050543 107608.706713 9.712338
10 7 2.771097 360.867977 107305.213928 9.750366
10 8 2.866138 348.901510 106971.927166 9.839058
10 9 2.777872 359.987736 107022.459745 9.825230
2 2 True 1200 1200 10 0 2.948138 339.197159 107917.767048 8.176148
10 1 2.975112 336.121798 108731.031179 8.060157
10 2 2.972867 336.375594 108791.134596 7.930100
10 3 2.973682 336.283445 107481.444597 8.275807
10 4 2.860864 349.544764 109171.261549 8.073807
10 5 2.978943 335.689545 108843.255043 8.079827
10 6 2.943017 339.787364 108837.707996 7.902384
10 7 2.971424 336.539030 108641.510248 8.103907
10 8 2.960844 337.741494 108594.880342 8.282661
10 9 2.971701 336.507559 108941.448450 7.781625
4 2 True 1200 1200 10 0 3.017677 331.380725 73027.454138 7.832050
10 1 3.012794 331.917822 72546.567678 7.671565
10 2 3.015492 331.620872 72918.371916 7.654369
10 3 3.023033 330.793619 73165.400982 7.994235
10 4 3.023469 330.745935 73355.324745 7.640839
10 5 3.007373 332.516074 72925.657511 7.917792
10 6 3.020975 331.018925 72917.241812 7.816106
10 7 3.027055 330.354095 72621.436119 8.030981
10 8 3.018988 331.236780 72900.611401 7.661909
10 9 3.024568 330.625713 73549.364567 7.837266
8 2 True 1200 1200 10 0 2.998040 333.551198 71839.100599 6.672293
10 1 2.996960 333.671480 71953.596115 6.899774
10 2 3.003266 332.970887 72201.970339 6.658748
10 3 3.007223 332.532734 71876.399755 6.796494
10 4 3.001819 333.131313 72039.752007 6.663635
10 5 3.000991 333.223283 71869.973898 6.706268
10 6 2.998712 333.476454 71487.551451 6.839141
10 7 2.995689 333.813041 72090.383053 6.699994
10 8 2.997357 333.627254 71656.096697 6.747678
10 9 3.008561 332.384825 71390.659571 6.660312
tensorrt-dynamic 1 2 True 1200 1200 10 0 3.057757 327.037096 22028.067350 9.554386
10 1 3.068450 325.897455 22086.875200 9.461403
10 2 3.051777 327.677965 22220.805407 9.710312
10 3 3.059780 326.820850 22105.169535 9.832859
10 4 3.068549 325.886965 21711.289167 9.576082
10 5 3.036279 329.350471 21837.047577 9.606481
10 6 3.043676 328.550100 22272.382021 9.786844
10 7 3.084932 324.156284 22317.816019 9.663343
10 8 3.051861 327.668905 21800.987244 9.696484
10 9 3.078246 324.860334 21827.045918 9.767771
rcnn-resnet101-lowproposals cpu 1 2 True 600 1024 10 0 2.304830 433.871508 1827.384472 9.684801
10 1 2.335542 428.166151 1877.870560 9.973526
10 2 2.323291 430.423975 1907.244205 9.956360
10 3 2.341737 427.033424 1797.054529 9.700894
10 4 2.303445 434.132338 1904.339314 9.792686
10 5 2.316462 431.692839 1895.573139 9.884715
10 6 2.332635 428.699732 1874.328613 9.527802
10 7 2.316138 431.753159 1868.458986 10.022640
10 8 2.324732 430.157185 1855.124235 9.516835
10 9 2.319349 431.155443 1874.348402 9.882808
2 2 True 600 1024 10 0 2.392571 417.960405 1934.826851 8.248568
10 1 2.369553 422.020555 1864.797354 8.019984
10 2 2.366152 422.627091 1837.976933 7.781029
10 3 2.352325 425.111413 1921.272755 7.981777
10 4 2.344138 426.596045 1819.552898 8.148909
10 5 2.397723 417.062402 1812.511444 8.311629
10 6 2.392005 418.059349 1924.170017 8.090615
10 7 2.382561 419.716477 1938.081741 8.445263
10 8 2.381802 419.850230 1980.180025 8.017421
10 9 2.372068 421.573162 1875.776291 7.830441
4 2 True 600 1024 10 0 2.336892 427.918851 1868.940830 7.598341
10 1 2.344501 426.529944 1935.627699 7.593155
10 2 2.330032 429.178715 1898.720503 7.677764
10 3 2.305983 433.654487 1910.155773 7.690132
10 4 2.348180 425.861716 1922.926188 7.581472
10 5 2.359492 423.820019 1815.843582 7.585049
10 6 2.324034 430.286229 1827.770472 7.804841
10 7 2.340035 427.343965 1935.333729 7.580370
10 8 2.323268 430.428207 1933.232069 7.671297
10 9 2.336914 427.914739 1979.201555 7.660180
8 2 True 600 1024 10 0 2.339513 427.439332 1906.532288 6.749809
10 1 2.308595 433.163851 1906.987190 6.635755
10 2 2.324493 430.201411 1838.825464 6.627470
10 3 2.322419 430.585563 1893.810034 6.669953
10 4 2.316868 431.617111 1839.324236 6.772801
10 5 2.338430 427.637279 1803.636789 6.496117
10 6 2.321023 430.844545 1816.417933 6.503448
10 7 2.335022 428.261399 1866.621733 6.521270
10 8 2.337814 427.750051 1857.588291 6.576374
10 9 2.333129 428.608984 1889.589310 6.512001
16 2 True 600 1024 10 0 2.342415 426.909819 1828.043222 6.163120
10 1 2.329757 429.229394 1817.310810 6.488718
10 2 2.347167 426.045462 1913.735867 6.387718
10 3 2.340523 427.254871 1836.093187 6.356589
10 4 2.333166 428.602129 1832.617521 6.242298
10 5 2.330831 429.031491 1834.811449 6.567240
10 6 2.330327 429.124296 1868.567705 6.202891
10 7 2.342290 426.932633 1880.446196 6.326608
10 8 2.335471 428.179160 1822.031498 6.301835
10 9 2.338979 427.536935 1963.088036 6.282225
32 2 True 600 1024 10 0 2.346851 426.102959 1896.929979 6.133009
10 1 2.346771 426.117368 1909.787178 6.159011
10 2 2.345309 426.383093 1862.657070 6.230399
10 3 2.349178 425.680831 1807.484150 6.064776
10 4 2.347325 426.016867 1946.138382 6.048486
10 5 2.346566 426.154599 1935.488462 6.421760
10 6 2.345271 426.389873 1891.182423 6.187461
10 7 2.343750 426.666580 1946.294308 6.042723
10 8 2.343750 426.666588 1940.818548 6.043978
10 9 2.343544 426.704139 1812.900066 6.035201
cpu-prebuilt 1 2 True 600 1024 10 0 2.325807 429.958344 3526.874781 9.588480
10 1 2.270778 440.377712 1796.475649 9.777665
10 2 2.315289 431.911469 1790.876150 9.529471
10 3 2.325723 429.973841 1776.808739 9.592175
10 4 2.334878 428.287983 1768.345118 9.633541
10 5 2.328097 429.535389 1796.634674 9.695768
10 6 2.295947 435.550213 1761.363983 9.647131
10 7 2.317605 431.479931 1774.100780 9.778976
10 8 2.306634 433.532238 1779.556274 9.850621
10 9 2.310858 432.739735 1789.674282 9.897470
2 2 True 600 1024 10 0 2.330590 429.075837 1780.204296 8.139193
10 1 2.329304 429.312706 1774.967194 8.168697
10 2 2.340682 427.225828 1788.007498 7.784605
10 3 2.328938 429.380298 1796.847105 7.883370
10 4 2.329753 429.229975 1793.567181 8.091271
10 5 2.340722 427.218676 1793.089628 7.717967
10 6 2.347164 426.046133 1801.154137 7.815659
10 7 2.315344 431.901217 1765.961885 7.840872
10 8 2.345038 426.432371 1770.402670 8.082330
10 9 2.328573 429.447651 1768.368959 7.858813
4 2 True 600 1024 10 0 2.318044 431.398153 1775.799513 7.703036
10 1 2.317155 431.563675 1758.645058 7.533014
10 2 2.318512 431.311131 1766.268730 7.707924
10 3 2.347870 425.918043 1754.207850 7.762402
10 4 2.331571 428.895354 1769.369602 7.625192
10 5 2.334888 428.286076 1824.254274 8.000851
10 6 2.318316 431.347668 1796.149492 7.516295
10 7 2.293173 436.076939 1791.581631 7.714301
10 8 2.300733 434.644163 1780.581951 7.791013
10 9 2.307829 433.307648 1800.542831 7.868201
8 2 True 600 1024 10 0 2.308230 433.232456 1830.583811 6.714031
10 1 2.322982 430.481076 1754.957438 6.623983
10 2 2.313316 432.279974 1763.799429 6.718278
10 3 2.311795 432.564318 1818.261623 6.643608
10 4 2.318583 431.297898 1796.255350 6.703928
10 5 2.302579 434.295654 1855.784655 6.638959
10 6 2.323247 430.432141 1753.847599 6.945089
10 7 2.303336 434.152931 1804.566622 6.446645
10 8 2.321016 430.845886 1780.231476 6.791070
10 9 2.304421 433.948398 1857.248783 6.564587
16 2 True 600 1024 10 0 2.319178 431.187347 1768.025875 6.277628
10 1 2.329284 429.316476 1784.183502 6.618336
10 2 2.322164 430.632800 1818.145752 6.284736
10 3 2.320084 431.018889 1772.001266 6.400876
10 4 2.308925 433.101922 1768.416643 6.351672
10 5 2.321651 430.727914 1793.220520 6.333083
10 6 2.327525 429.640979 1785.733938 6.494991
10 7 2.325310 430.050239 1808.824539 6.321713
10 8 2.326018 429.919228 1767.354250 6.455906
10 9 2.316092 431.761786 1779.067755 6.235562
32 2 True 600 1024 10 0 2.324147 430.265404 1888.056993 6.092943
10 1 2.330996 429.001175 1773.266315 5.997755
10 2 2.324931 430.120274 1768.773794 6.219804
10 3 2.326243 429.877706 1755.516529 6.305285
10 4 2.330322 429.125234 1770.206451 6.067540
10 5 2.330357 429.118887 1774.690866 6.300285
10 6 2.328126 429.530039 1794.872761 6.287884
10 7 2.323454 430.393636 1837.211609 6.190877
10 8 2.327724 429.604284 1790.501833 6.103795
10 9 2.329613 429.255851 1901.494026 6.256886
cuda 1 2 True 600 1024 10 0 12.406357 80.603838 1780.699968 9.582043
10 1 12.251652 81.621647 1794.161797 9.646416
10 2 12.455653 80.284834 1797.565699 9.800553
10 3 12.595318 79.394579 1823.054552 9.614229
10 4 12.471096 80.185413 1764.760017 9.642482
10 5 12.331328 81.094265 1788.885593 9.572625
10 6 12.367471 80.857277 1807.747602 9.576082
10 7 12.322235 81.154108 1790.563583 9.756565
10 8 12.468353 80.203056 1789.198875 9.592295
10 9 12.288948 81.373930 1838.732004 9.908557
2 2 True 600 1024 10 0 13.884102 72.024822 1794.478178 7.758796
10 1 13.846106 72.222471 1793.740749 7.920980
10 2 13.656312 73.226213 1811.664581 7.816315
10 3 14.140685 70.717931 1795.699835 7.919908
10 4 13.914802 71.865916 1769.134283 7.877946
10 5 13.854018 72.181225 1774.894953 8.070529
10 6 13.765921 72.643161 1787.384510 7.844150
10 7 13.737043 72.795868 1797.870159 8.096159
10 8 13.927324 71.801305 1759.881735 7.981837
10 9 13.958052 71.643233 1791.018724 8.197069
4 2 True 600 1024 10 0 15.190190 65.831959 1776.934385 7.706136
10 1 15.243838 65.600276 1777.024984 7.771075
10 2 15.098110 66.233456 1792.327404 7.554978
10 3 15.040275 66.488147 1809.364080 7.516950
10 4 15.200375 65.787852 1815.351009 7.396162
10 5 15.293446 65.387487 1769.275188 7.549137
10 6 15.397154 64.947069 1799.986839 7.576406
10 7 15.295468 65.378845 1776.325464 7.525742
10 8 15.077933 66.322088 1804.390907 7.936239
10 9 15.237096 65.629303 1808.810949 7.494003
8 2 True 600 1024 10 0 15.461981 64.674765 1766.000986 6.446227
10 1 15.051272 66.439569 1797.045469 6.708249
10 2 15.583301 64.171255 1795.336723 6.666273
10 3 15.675751 63.792795 1806.338787 6.792292
10 4 15.192460 65.822124 1789.094448 6.449252
10 5 15.368740 65.067142 1798.160791 6.783634
10 6 15.472476 64.630896 1788.929939 6.678030
10 7 15.579365 64.187467 1796.284676 6.702468
10 8 15.561165 64.262539 1804.042816 6.751165
10 9 15.482578 64.588726 1787.196159 6.467342
16 2 True 600 1024 10 0 15.844891 63.111827 1794.874191 6.226681
10 1 15.785117 63.350812 1797.908306 6.222032
10 2 15.863543 63.037619 1770.837784 6.245166
10 3 15.797339 63.301802 1781.039000 6.200060
10 4 15.756247 63.466892 1792.331934 6.264597
10 5 15.942631 62.724903 1819.593191 6.288953
10 6 15.947454 62.705934 1806.744814 6.427646
10 7 15.969829 62.618077 1803.507328 6.459050
10 8 15.943665 62.720835 1801.426888 6.716251
10 9 15.819410 63.213482 1814.743996 6.266527
32 2 True 600 1024 10 0 15.984870 62.559158 3650.776625 6.016515
10 1 15.851304 63.086294 1801.421165 6.203383
10 2 15.823961 63.195303 1777.224064 6.303597
10 3 15.931699 62.767945 1807.895422 6.169319
10 4 15.918470 62.820107 1787.414074 6.007489
10 5 15.942908 62.723815 1788.667917 6.297566
10 6 15.861777 63.044637 1799.606562 5.995713
10 7 15.835060 63.151009 1788.318634 6.021909
10 8 15.901913 62.885515 1781.450510 6.085332
10 9 15.919637 62.815502 1806.703806 6.254602
tensorrt 1 2 True 600 1024 10 0 12.214332 81.871033 21520.837307 9.573817
10 1 12.332597 81.085920 21626.183748 9.447336
10 2 12.442351 80.370665 21467.853069 9.792447
10 3 12.087889 82.727432 21474.493980 9.541512
10 4 12.321185 81.161022 21446.175575 9.764910
10 5 12.515192 79.902887 21555.088282 9.565473
10 6 12.355776 80.933809 21599.481106 9.463191
10 7 12.368492 80.850601 21490.350962 9.759068
10 8 12.350864 80.965996 21481.666803 9.582162
10 9 12.578623 79.499960 21465.908527 9.615660
2 2 True 600 1024 10 0 13.813024 72.395444 22227.952480 7.780313
10 1 13.716603 72.904348 22319.144011 7.857382
10 2 13.694680 73.021054 22134.814024 7.850707
10 3 13.688133 73.055983 22103.310585 7.707417
10 4 13.961142 71.627378 22314.942598 7.925808
10 5 13.772498 72.608471 21941.455126 7.641792
10 6 13.697386 73.006630 22218.677521 7.755816
10 7 13.799504 72.466373 22129.404545 7.634580
10 8 13.901105 71.936727 22288.933516 7.660806
10 9 13.729849 72.834015 22086.379051 7.763565
4 2 True 600 1024 10 0 14.887289 67.171395 17102.623701 7.688612
10 1 14.901266 67.108393 17197.448254 7.571965
10 2 14.920641 67.021251 17331.134081 7.504016
10 3 15.179828 65.876901 17033.634424 7.494926
10 4 14.912948 67.055821 16926.265717 7.471025
10 5 15.140375 66.048563 16950.997591 7.679582
10 6 14.804567 67.546725 17085.191727 7.456332
10 7 14.981155 66.750526 17058.808327 7.495344
10 8 15.026346 66.549778 17133.375883 7.729977
10 9 15.227070 65.672517 17176.436424 7.496089
8 2 True 600 1024 10 0 15.221861 65.694988 17159.111500 6.547987
10 1 15.405163 64.913303 16983.058214 6.352276
10 2 15.410370 64.891368 17195.908546 6.562412
10 3 15.475930 64.616472 17181.518078 6.568372
10 4 15.287969 65.410912 17114.690065 6.539419
10 5 15.276213 65.461248 17141.021967 6.698266
10 6 15.247918 65.582722 17249.283552 6.569609
10 7 15.357240 65.115869 17039.963484 6.565511
10 8 15.370338 65.060377 17096.708536 6.546274
10 9 15.376417 65.034658 17224.649191 6.451920
16 2 True 600 1024 10 0 15.976646 62.591359 16602.551460 6.260730
10 1 15.857681 63.060924 16584.681034 6.315164
10 2 15.944358 62.718108 16473.892927 6.287269
10 3 15.798477 63.297242 16396.528482 6.403334
10 4 15.984916 62.558979 16479.218006 6.443962
10 5 15.886793 62.945366 16550.421476 6.713882
10 6 15.834240 63.154280 16433.331013 6.300107
10 7 15.821920 63.203454 16525.912762 6.309062
10 8 15.796469 63.305289 16298.932791 6.288730
10 9 15.686025 63.751012 16429.342270 6.249793
32 2 True 600 1024 10 0 15.851167 63.086838 17299.829245 6.002590
10 1 15.771455 63.405693 17187.531471 6.064180
10 2 15.847697 63.100651 16918.631315 6.084248
10 3 15.869457 63.014127 16942.778826 6.116644
10 4 15.902235 62.884241 17201.121569 6.206512
10 5 15.848666 63.096792 17203.660488 6.110922
10 6 15.744702 63.513428 16960.286617 6.126728
10 7 15.995883 62.516086 16781.747580 6.115861
10 8 16.003934 62.484637 17277.103662 6.359812
10 9 15.886509 62.946491 17069.191456 6.053250
tensorrt-dynamic 1 2 True 600 1024 10 0 16.734109 59.758186 12546.623468 9.447694
10 1 15.540503 64.347982 12751.722574 9.544373
10 2 17.016744 58.765650 12663.028955 9.646416
10 3 15.905952 62.869549 12665.186644 9.731531
10 4 16.617291 60.178280 12725.333691 9.541631
10 5 15.995180 62.518835 12693.892956 9.388208
10 6 16.785542 59.575081 12597.485781 9.372473
10 7 16.232141 61.606169 12776.030064 9.385705
10 8 16.406110 60.952902 12575.339556 9.355068
10 9 16.690824 59.913158 12786.731243 9.495974
2 2 True 600 1024 10 0 18.098985 55.251718 12683.964491 7.887363
10 1 17.265093 57.920337 12633.701324 7.705867
10 2 18.116848 55.197239 12655.244827 7.636547
10 3 17.753818 56.325912 12547.592402 7.811964
10 4 17.941206 55.737615 12566.540241 7.757425
10 5 18.475157 54.126740 12551.458359 7.632315
10 6 18.155195 55.080652 12679.618835 7.662952
10 7 17.944161 55.728436 12613.080740 7.677913
10 8 17.781290 56.238890 12662.493467 7.966459
10 9 17.730702 56.399345 12775.243044 7.885098
4 2 True 600 1024 10 0 18.789180 53.222120 12475.938320 7.554084
10 1 19.372804 51.618755 12353.553295 7.454753
10 2 18.962770 52.734911 12520.078421 7.705092
10 3 19.544867 51.164329 12576.491117 7.529080
10 4 19.132111 52.268147 12633.758307 7.454246
10 5 19.420457 51.492095 12593.590736 7.604241
10 6 19.240831 51.972806 12665.605545 7.520586
10 7 19.024893 52.562714 12597.863674 7.429093
10 8 19.546916 51.158965 12497.672319 7.483214
10 9 19.574055 51.088035 12649.907827 7.510900
8 2 True 600 1024 10 0 19.121426 52.297354 12547.012329 6.488472
10 1 19.530715 51.201403 12815.947533 6.437615
10 2 19.642082 50.911099 12591.820240 6.410971
10 3 19.629661 50.943315 12722.246408 6.491035
10 4 19.466289 51.370859 12605.246305 6.557107
10 5 19.570128 51.098287 12560.843945 6.665528
10 6 19.819616 50.455064 12569.098711 6.636485
10 7 19.353596 51.669985 12711.484432 6.569028
10 8 19.770418 50.580621 12687.051296 6.467119
10 9 19.236000 51.985860 12635.541916 6.484702
16 2 True 600 1024 10 0 19.764188 50.596565 12726.392508 6.491296
10 1 19.618740 50.971672 12463.183403 6.296717
10 2 19.746944 50.640747 12824.097633 6.244048
10 3 19.659316 50.866470 12399.031162 6.438285
10 4 19.586755 51.054910 12678.436041 6.400786
10 5 19.455510 51.399320 12383.874655 6.279379
10 6 19.421693 51.488817 12594.056368 6.321602
10 7 19.557552 51.131144 12426.792145 6.236024
10 8 19.560932 51.122308 12633.033276 6.164126
10 9 19.611006 50.991774 12655.103922 6.318189
rcnn-resnet50-lowproposals cpu 1 2 True 600 1024 10 0 3.776142 264.820576 1686.880589 9.643912
10 1 3.688364 271.122932 1704.359293 9.935141
10 2 3.657459 273.413897 1718.741655 9.537816
10 3 3.681368 271.638155 1619.982958 9.682417
10 4 3.737363 267.568350 1606.738567 9.875298
10 5 3.723914 268.534660 1680.978775 9.723186
10 6 3.679088 271.806479 1595.991850 10.011077
10 7 3.744985 267.023802 1645.612001 9.914756
10 8 3.670180 272.466183 1597.892761 9.897351
10 9 3.638788 274.816751 1685.933113 9.681463
2 2 True 600 1024 10 0 3.829103 261.157751 1675.863743 8.010089
10 1 3.719642 268.843055 1708.093643 8.337080
10 2 3.827712 261.252642 1583.373070 8.207142
10 3 3.872635 258.222103 1676.360369 8.103073
10 4 3.844418 260.117412 1701.800108 8.178890
10 5 3.838114 260.544658 1634.593487 7.981360
10 6 3.844719 260.097027 1629.212379 7.709503
10 7 3.889469 257.104516 1712.654352 8.213580
10 8 3.845763 260.026455 1666.113377 8.351564
10 9 3.876064 257.993698 1699.373484 7.974684
4 2 True 600 1024 10 0 3.947951 253.295958 1684.618473 7.599264
10 1 3.891619 256.962478 1657.998562 7.850498
10 2 3.894700 256.759167 1633.961201 7.827699
10 3 3.903964 256.149888 1669.556618 7.757664
10 4 3.934027 254.192472 1699.060202 7.414043
10 5 3.935468 254.099369 1680.139542 7.916749
10 6 3.906316 255.995691 1682.508945 7.896006
10 7 3.908063 255.881250 1696.475744 7.962078
10 8 3.870694 258.351624 1627.237558 7.742405
10 9 3.939330 253.850281 1670.836210 7.573932
8 2 True 600 1024 10 0 3.859469 259.102970 1613.138199 6.570980
10 1 3.840609 260.375351 1678.300142 6.501794
10 2 3.866028 258.663416 1619.373083 6.715238
10 3 3.824838 261.448950 1683.642387 6.480053
10 4 3.865737 258.682907 1648.265123 6.416053
10 5 3.853725 259.489179 1659.857988 6.652877
10 6 3.878279 257.846326 1640.050888 6.461129
10 7 3.836441 260.658264 1623.372793 6.792918
10 8 3.868346 258.508444 1711.113930 6.581336
10 9 3.881968 257.601291 1625.286818 6.494537
16 2 True 600 1024 10 0 3.866765 258.614108 1654.815435 6.322645
10 1 3.869823 258.409739 1629.233122 6.453760
10 2 3.884349 257.443383 1612.328529 6.364495
10 3 3.876037 257.995442 1632.277966 6.321780
10 4 3.850778 259.687766 1690.767527 6.561197
10 5 3.880620 257.690772 1673.472166 6.332882
10 6 3.877476 257.899716 1623.745441 6.240845
10 7 3.866697 258.618683 1625.013351 6.257720
10 8 3.868957 258.467570 1662.688732 6.280772
10 9 3.876812 257.943884 1590.418100 6.240308
32 2 True 600 1024 10 0 3.892849 256.881282 1600.942612 6.155942
10 1 3.903304 256.193206 1638.194084 6.021615
10 2 3.878720 257.816985 1664.097309 6.247904
10 3 3.887547 257.231601 1596.452713 6.269380
10 4 3.901603 256.304920 1619.798660 5.978562
10 5 3.885975 257.335678 1698.755503 6.111003
10 6 3.880982 257.666714 1659.761429 6.098967
10 7 3.888605 257.161640 1715.822935 6.176483
10 8 3.894709 256.758608 1653.138161 6.132122
10 9 3.861509 258.966148 1622.913361 5.929604
cpu-prebuilt 1 2 True 600 1024 10 0 3.571742 279.975414 2762.686491 17.721534
10 1 3.650224 273.955822 1638.625622 9.535074
10 2 3.613571 276.734591 1577.911139 9.868741
10 3 3.677456 271.927118 1553.699970 9.736419
10 4 3.490724 286.473513 1564.384937 9.866118
10 5 3.660782 273.165703 1538.223743 9.918094
10 6 3.549237 281.750679 1550.678253 9.536862
10 7 3.651088 273.890972 1556.126833 9.942293
10 8 3.597361 277.981520 1622.909307 9.845018
10 9 3.706443 269.800425 1549.013138 9.619832
2 2 True 600 1024 10 0 3.783168 264.328718 1579.739332 8.128703
10 1 3.735273 267.718077 1639.370680 8.020580
10 2 3.812983 262.261868 1627.659321 8.489132
10 3 3.779005 264.619946 1547.396183 7.694066
10 4 3.778412 264.661431 1540.544748 8.074224
10 5 3.742069 267.231822 1542.426109 7.966459
10 6 3.686432 271.265030 1562.769413 8.267522
10 7 3.810065 262.462735 1558.845758 7.917762
10 8 3.810388 262.440443 1551.403522 8.073688
10 9 3.842094 260.274768 1539.196014 7.852554
4 2 True 600 1024 10 0 3.818147 261.907160 1545.026302 7.532388
10 1 3.838781 260.499358 1588.864565 7.528603
10 2 3.873019 258.196533 1605.333567 7.553667
10 3 3.875109 258.057237 1602.406979 7.640928
10 4 3.880915 257.671177 1651.915789 8.132309
10 5 3.847162 259.931862 1609.186649 7.605106
10 6 3.842808 260.226369 1652.390003 7.553011
10 7 3.842139 260.271668 1558.289051 7.787108
10 8 3.834420 260.795653 1633.130312 7.855207
10 9 3.835751 260.705113 1630.680799 7.746011
8 2 True 600 1024 10 0 3.845960 260.013074 1665.308952 6.696820
10 1 3.823265 261.556566 1555.235147 6.600454
10 2 3.834839 260.767132 1551.889658 6.581768
10 3 3.820514 261.744857 1559.449196 6.481290
10 4 3.827530 261.265099 1616.168261 6.699994
10 5 3.720410 268.787593 1541.977167 6.636932
10 6 3.846771 259.958267 1547.911406 6.595314
10 7 3.814516 262.156427 1532.830238 6.751522
10 8 3.834496 260.790437 1566.138983 6.781057
10 9 3.811435 262.368351 1609.718800 6.552622
16 2 True 600 1024 10 0 3.854812 259.416029 1636.818171 6.427616
10 1 3.837959 260.555148 1606.189728 6.298289
10 2 3.841213 260.334447 1573.059797 6.187394
10 3 3.826297 261.349320 1541.368961 6.510794
10 4 3.810786 262.413070 1613.836050 6.600820
10 5 3.844609 260.104463 1562.781334 6.305188
10 6 3.832436 260.930672 1536.884785 6.461143
10 7 3.813556 262.222484 1559.064865 6.540872
10 8 3.830241 261.080176 1580.272436 6.225564
10 9 3.813589 262.220159 1595.091581 6.294802
32 2 True 600 1024 10 0 3.864463 258.768164 1577.320099 6.226402
10 1 3.848046 259.872176 1571.588993 6.021578
10 2 3.843316 260.191955 1545.690060 6.018501
10 3 3.847767 259.890988 1636.627197 6.211761
10 4 3.839370 260.459386 1544.546366 6.139554
10 5 3.838207 260.538295 1589.231253 6.220452
10 6 3.844068 260.141060 1549.546003 6.105881
10 7 3.852871 259.546742 1572.925806 6.310277
10 8 3.843005 260.213032 1527.139902 6.078742
10 9 3.830920 261.033930 1548.498392 6.016731
cuda 1 2 True 600 1024 10 0 16.084366 62.172174 1567.276478 9.552002
10 1 15.767586 63.421249 1556.388855 9.823442
10 2 16.132559 61.986446 1579.791784 9.697795
10 3 15.805613 63.268661 1537.657022 9.535074
10 4 16.114708 62.055111 1572.025537 9.755731
10 5 16.276486 61.438322 1556.050539 9.635806
10 6 15.995302 62.518358 1579.099894 9.872913
10 7 15.939197 62.738419 1558.127880 9.831071
10 8 16.222410 61.643124 1542.162418 9.619713
10 9 16.040508 62.342167 1581.772089 9.413123
2 2 True 600 1024 10 0 18.869419 52.995801 1550.209761 7.856190
10 1 18.601131 53.760171 1540.921688 8.131802
10 2 18.757439 53.312182 1539.927483 8.144259
10 3 19.118986 52.304029 1575.437307 8.070350
10 4 18.425405 54.272890 1562.089443 7.939517
10 5 18.737412 53.369164 1564.453125 8.044004
10 6 19.089531 52.384734 1557.888746 8.139849
10 7 18.854364 53.038120 1559.280396 8.034647
10 8 19.140580 52.245021 1544.927359 7.880390
10 9 18.818411 53.139448 1578.546524 8.000910
4 2 True 600 1024 10 0 20.235260 49.418688 1598.390818 7.522523
10 1 20.737907 48.220873 1590.816736 7.533103
10 2 20.664268 48.392713 1582.442522 7.434607
10 3 20.443440 48.915446 1579.503536 7.438391
10 4 20.673919 48.370123 1575.470686 7.680893
10 5 20.722769 48.256099 1590.139151 7.592112
10 6 20.617720 48.501968 1567.997932 7.628381
10 7 20.751708 48.188806 1584.429502 7.561356
10 8 20.672900 48.372507 1578.259468 7.628292
10 9 20.251308 49.379528 1586.111307 7.510394
8 2 True 600 1024 10 0 21.168346 47.240347 1552.038193 6.471038
10 1 20.942423 47.749966 1568.918228 6.541625
10 2 20.961485 47.706544 1569.382429 6.726533
10 3 21.078944 47.440708 1556.970835 6.642163
10 4 21.030300 47.550440 1575.931072 6.492779
10 5 21.013560 47.588319 1558.716536 6.599799
10 6 20.729477 48.240483 1554.875374 6.788582
10 7 20.900406 47.845960 1577.084303 6.423414
10 8 21.157708 47.264099 1579.204321 6.443471
10 9 21.093294 47.408432 1558.570147 6.477118
16 2 True 600 1024 10 0 21.285771 46.979740 1574.208736 6.273493
10 1 21.814649 45.840755 1547.834158 6.276205
10 2 21.272560 47.008917 1569.916487 6.227680
10 3 21.355689 46.825930 1588.491201 6.268911
10 4 21.516113 46.476796 1551.367760 6.255686
10 5 21.321133 46.901822 1562.980175 6.638221
10 6 21.605507 46.284497 1574.791670 6.450735
10 7 21.552654 46.397999 1586.225033 6.326117
10 8 21.325659 46.891868 1552.012682 6.230108
10 9 21.816472 45.836926 1549.138546 6.490417
32 2 True 600 1024 10 0 21.648104 46.193421 2622.985125 5.982913
10 1 21.612583 46.269342 1578.860044 5.948056
10 2 21.579898 46.339422 1557.170868 5.979646
10 3 21.113880 47.362208 1564.578056 6.043833
10 4 21.264138 47.027536 1575.670719 5.968492
10 5 21.639975 46.210773 1553.828239 6.170273
10 6 21.754156 45.968227 1556.079626 6.273590
10 7 21.594380 46.308346 1561.932802 6.225429
10 8 21.437587 46.647042 1545.925856 5.935386
10 9 21.671702 46.143122 1582.964659 5.895533
tensorrt 1 2 True 600 1024 10 0 15.691373 63.729286 19319.557190 9.539604
10 1 16.066745 62.240362 19135.199308 9.448171
10 2 16.237734 61.584949 18963.383913 9.652853
10 3 15.890946 62.928915 19239.841938 9.773254
10 4 16.168193 61.849833 18947.120667 9.727597
10 5 16.050697 62.302589 19103.104830 9.832144
10 6 15.662310 63.847542 19037.372112 9.446740
10 7 15.738181 63.539743 19217.607260 9.669065
10 8 15.890404 62.931061 19089.416265 9.625077
10 9 16.007144 62.472105 19097.249746 9.390116
2 2 True 600 1024 10 0 18.722816 53.410769 19902.643919 7.814705
10 1 18.369201 54.438949 20071.136236 7.931173
10 2 18.643340 53.638458 19664.289951 7.718384
10 3 18.347425 54.503560 19818.207979 7.790208
10 4 18.914263 52.870154 19927.687168 7.919431
10 5 18.418932 54.291964 19979.107618 7.784545
10 6 18.046460 55.412531 19901.773214 7.896602
10 7 18.879697 52.966952 19892.263412 7.928789
10 8 19.006444 52.613735 19943.155289 7.798970
10 9 18.303110 54.635525 19775.532484 7.949591
4 2 True 600 1024 10 0 20.535098 48.697114 14887.097836 7.770628
10 1 20.871364 47.912538 14837.127686 7.451624
10 2 19.893728 50.267100 14736.799479 7.715344
10 3 20.228770 49.434543 14832.970142 7.595092
10 4 20.390196 49.043179 14797.318459 7.516712
10 5 19.897644 50.257206 14650.147915 7.577091
10 6 20.963004 47.703087 14841.446400 7.591426
10 7 20.416796 48.979282 14787.937403 7.556558
10 8 20.272229 49.328566 14814.521074 7.412940
10 9 20.624741 48.485458 14747.805834 7.531047
8 2 True 600 1024 10 0 20.523932 48.723608 14709.718943 6.500468
10 1 20.800436 48.075914 14736.230612 6.597221
10 2 20.801120 48.074335 14834.714651 6.659746
10 3 20.381575 49.063921 14885.342598 6.665811
10 4 20.680889 48.353821 14964.942932 6.466866
10 5 20.674594 48.368543 14761.517286 6.718606
10 6 20.751785 48.188627 14838.866711 6.756231
10 7 20.668252 48.383385 14749.555826 6.562561
10 8 20.842673 47.978491 14885.761738 6.570294
10 9 20.686052 48.341751 14799.415827 6.419659
16 2 True 600 1024 10 0 21.451991 46.615720 14167.942286 6.249599
10 1 21.320795 46.902567 14100.311995 6.188668
10 2 21.426163 46.671912 14353.028536 6.249785
10 3 21.325700 46.891779 14178.853989 6.322034
10 4 21.600778 46.294630 14232.103586 6.441049
10 5 21.403223 46.721935 14159.862041 6.345078
10 6 21.221657 47.121674 14205.360174 6.275162
10 7 21.808517 45.853645 14251.514673 6.265521
10 8 21.472631 46.570912 14136.046886 6.623082
10 9 21.514844 46.479538 14198.043108 6.200984
32 2 True 600 1024 10 0 21.399831 46.729341 14813.133955 5.983353
10 1 21.077047 47.444977 15146.105051 6.208703
10 2 21.434564 46.653621 14987.927675 6.029747
10 3 21.054068 47.496758 14913.025141 5.958252
10 4 21.407391 46.712838 14824.986458 6.372876
10 5 21.667721 46.151601 14543.232679 6.207205
10 6 21.553353 46.396494 15270.527840 6.166164
10 7 21.426570 46.671025 15092.374563 6.257627
10 8 21.373267 46.787418 15003.621578 6.111909
10 9 21.835027 45.797974 14296.097994 5.982772
tensorrt-dynamic 1 2 True 600 1024 10 0 19.547941 51.156282 10180.581331 9.346962
10 1 19.277509 51.873922 10163.849592 9.350657
10 2 19.510115 51.255465 10287.260056 9.528279
10 3 19.058867 52.469015 10135.560513 9.253860
10 4 19.540564 51.175594 10073.168755 9.530902
10 5 19.310525 51.785231 10132.394791 9.380341
10 6 19.149625 52.220345 10165.155172 9.348631
10 7 19.451754 51.409245 10150.344610 9.406328
10 8 20.320059 49.212456 10031.230688 9.447813
10 9 19.664979 50.851822 10164.824486 9.219527
2 2 True 600 1024 10 0 22.949729 43.573499 10234.578848 7.745802
10 1 23.307969 42.903781 10127.582073 7.786870
10 2 22.395421 44.651985 10190.723181 7.788539
10 3 22.153697 45.139194 10205.787420 7.702053
10 4 22.028213 45.396328 10160.840750 7.712722
10 5 22.668666 44.113755 10264.372826 7.872105
10 6 22.913997 43.641448 10351.198196 7.881701
10 7 23.252020 43.007016 10173.133373 7.907987
10 8 23.347671 42.830825 10346.702337 7.804811
10 9 22.761730 43.933392 10099.149704 7.834911
4 2 True 600 1024 10 0 24.980928 40.030539 10210.936308 7.672369
10 1 24.315863 41.125417 10129.615307 7.373780
10 2 24.481564 40.847063 10110.827208 7.653028
10 3 24.314594 41.127563 10235.378981 7.516921
10 4 24.006926 41.654646 10211.080074 7.368803
10 5 23.840958 41.944623 10235.380888 7.643521
10 6 24.729655 40.437281 10239.523649 7.675588
10 7 23.922443 41.801751 10143.581629 7.406592
10 8 25.000061 39.999902 10203.043222 7.486582
10 9 24.427135 40.938079 10122.092724 7.486641
8 2 True 600 1024 10 0 25.554476 39.132088 10071.565628 6.368309
10 1 25.588033 39.080769 10232.257605 6.553963
10 2 25.195574 39.689511 10183.977604 6.563455
10 3 25.130810 39.791793 10264.825344 6.493405
10 4 25.078032 39.875537 10099.519730 6.400272
10 5 25.040546 39.935231 10256.351471 6.537244
10 6 25.259359 39.589286 10054.270744 6.733045
10 7 24.842162 40.254146 10206.881762 6.629705
10 8 25.477580 39.250195 10128.998518 6.429940
10 9 25.006116 39.990216 10079.476833 6.499588
16 2 True 600 1024 10 0 25.089114 39.857924 10100.908995 6.341994
10 1 25.180514 39.713249 10158.185482 6.254509
10 2 25.322160 39.491102 10162.925482 6.388113
10 3 25.077367 39.876595 10212.424040 6.413274
10 4 25.038762 39.938077 10214.348316 6.333299
10 5 25.090333 39.855987 10069.103003 6.450847
10 6 25.516960 39.189622 10053.850174 6.228276
10 7 25.400204 39.369762 10129.296064 6.264724
10 8 25.372432 39.412856 10157.729149 6.278612
10 9 25.307674 39.513707 10180.256844 6.326571
ssd-inception-v2 cpu 1 2 True 300 300 10 0 26.572632 37.632704 1036.832571 9.979129
10 1 25.716150 38.886070 1052.604437 10.076046
10 2 26.590658 37.607193 1026.300430 9.832025
10 3 26.577347 37.626028 1056.766033 10.233045
10 4 23.907342 41.828156 1039.777040 9.956598
10 5 26.816429 37.290573 1021.740913 9.971499
10 6 26.684888 37.474394 1044.924021 10.188818
10 7 25.423568 39.333582 1032.768726 10.091543
10 8 25.973013 38.501501 1049.998283 9.798884
10 9 26.749388 37.384033 1030.857325 9.883761
2 2 True 300 300 10 0 32.043027 31.208038 1066.937923 8.204281
10 1 31.326843 31.921506 1013.860941 8.305132
10 2 32.727602 30.555248 1017.535210 8.301854
10 3 31.137768 32.115340 1122.016907 8.570492
10 4 31.417301 31.829596 1069.966793 8.527339
10 5 31.583139 31.662464 1033.203602 8.397460
10 6 32.600540 30.674338 1074.125528 8.381069
10 7 31.778762 31.467557 1052.721739 8.592367
10 8 32.721602 30.560851 1072.947264 8.622944
10 9 31.767570 31.478643 1021.064281 8.429468
4 2 True 300 300 10 0 34.712152 28.808355 1079.628944 8.134544
10 1 35.080577 28.505802 1052.744389 8.479595
10 2 36.187273 27.634025 1064.787149 8.079141
10 3 34.302220 29.152632 1054.917336 8.094251
10 4 34.690117 28.826654 1076.272964 8.246601
10 5 33.377930 29.959917 1068.475246 8.141667
10 6 34.564902 28.931081 1044.254780 8.358598
10 7 35.557907 28.123140 1029.908180 8.054137
10 8 35.066060 28.517604 1081.802130 8.091629
10 9 34.662022 28.850019 1085.449696 8.209705
8 2 True 300 300 10 0 37.562291 26.622444 1047.247410 7.124349
10 1 37.033181 27.002811 1053.466082 7.265568
10 2 37.049783 26.990712 1032.723665 6.791547
10 3 37.758629 26.484013 1050.156116 7.198974
10 4 37.212164 26.872933 1031.953812 6.740212
10 5 37.178190 26.897490 1072.241068 6.956279
10 6 37.115846 26.942670 1005.733490 6.915882
10 7 37.672318 26.544690 1009.104013 7.019505
10 8 37.186307 26.891619 1058.635473 6.974444
10 9 37.773038 26.473910 1056.904554 6.848559
16 2 True 300 300 10 0 38.760825 25.799245 1037.404537 6.621554
10 1 38.587141 25.915369 1084.344625 6.622039
10 2 37.776908 26.471198 1084.999084 6.839000
10 3 38.882073 25.718793 1034.811497 6.646998
10 4 38.437735 26.016101 1054.725647 6.825827
10 5 37.957524 26.345238 1047.569513 6.969713
10 6 38.986408 25.649965 1011.844397 6.576128
10 7 37.696145 26.527911 1029.245853 6.486863
10 8 37.715064 26.514605 1066.723347 6.803140
10 9 37.829549 26.434362 1069.365263 6.891549
32 2 True 300 300 10 0 41.369771 24.172239 1877.101421 6.227642
10 1 41.823907 23.909770 1041.133165 6.384440
10 2 41.438539 24.132125 1031.316280 6.295051
10 3 41.317048 24.203084 1022.511482 6.299019
10 4 41.612971 24.030969 1039.436340 6.318260
10 5 41.997414 23.810990 1026.664257 6.244041
10 6 41.327073 24.197213 1021.815777 6.243180
10 7 40.745813 24.542399 1082.146883 6.241050
10 8 41.205146 24.268813 1055.332422 6.234813
10 9 40.993315 24.394222 1066.293716 6.245423
cpu-prebuilt 1 2 True 300 300 10 0 25.310193 39.509773 1867.937565 9.948254
10 1 26.275655 38.058043 981.299639 9.884000
10 2 26.514176 37.715673 962.152719 10.094285
10 3 26.516020 37.713051 978.623629 9.932637
10 4 26.118577 38.286924 965.626478 9.927273
10 5 27.215064 36.744356 957.661629 10.138273
10 6 24.788738 40.340900 966.713190 10.067463
10 7 26.974057 37.072659 968.365669 9.965897
10 8 25.825564 38.721323 1029.874802 10.140657
10 9 27.613726 36.213875 979.713202 9.949684
2 2 True 300 300 10 0 32.721091 30.561328 963.088512 8.421898
10 1 31.530668 31.715155 959.016800 8.090496
10 2 32.665919 30.612946 967.540503 8.118570
10 3 31.888452 31.359315 1007.323503 8.356035
10 4 31.761195 31.484962 992.185116 8.307159
10 5 32.031525 31.219244 964.762449 8.395672
10 6 33.484917 29.864192 1035.646677 8.118153
10 7 31.882877 31.364799 949.239254 8.268595
10 8 32.539461 30.731916 972.521782 8.156657
10 9 32.463151 30.804157 958.978176 8.524895
4 2 True 300 300 10 0 36.524136 27.379155 1057.915449 7.965326
10 1 34.525851 28.963804 977.621317 8.055300
10 2 34.211428 29.229999 973.495722 8.031219
10 3 35.736275 27.982771 972.449064 8.033156
10 4 35.382728 28.262377 975.290775 7.904619
10 5 35.067306 28.516591 971.166134 8.013546
10 6 35.571552 28.112352 972.326040 8.023113
10 7 34.884363 28.666139 969.641447 8.077502
10 8 36.474825 27.416170 970.388651 7.910192
10 9 34.980591 28.587282 964.927912 8.209705
8 2 True 300 300 10 0 38.185275 26.188105 968.335152 6.790340
10 1 37.877078 26.401192 959.783554 6.948248
10 2 37.318362 26.796460 992.677689 7.083684
10 3 38.653840 25.870651 987.122536 6.998539
10 4 37.473358 26.685625 961.175203 6.757125
10 5 37.138070 26.926547 959.964514 6.811559
10 6 37.649619 26.560694 978.461504 6.938145
10 7 37.893846 26.389509 975.080729 6.824479
10 8 37.623319 26.579261 975.236893 6.731853
10 9 37.729631 26.504368 1045.089245 6.872997
16 2 True 300 300 10 0 38.612606 25.898278 978.707552 6.551683
10 1 38.210822 26.170596 1036.368132 6.597474
10 2 37.996466 26.318237 972.175360 6.533176
10 3 38.168986 26.199281 978.079319 6.587684
10 4 38.205188 26.174456 964.130640 6.586052
10 5 39.245455 25.480658 1013.383627 6.861612
10 6 38.218634 26.165247 960.180998 6.540664
10 7 37.764514 26.479885 967.053413 6.576061
10 8 38.970583 25.660381 968.597889 6.532528
10 9 37.874491 26.402995 983.211279 6.573610
32 2 True 300 300 10 0 41.838131 23.901641 975.125790 6.271232
10 1 41.999227 23.809962 982.212782 6.267916
10 2 41.301829 24.212003 1006.886482 6.328121
10 3 41.721473 23.968473 969.808578 6.299771
10 4 41.430263 24.136946 978.005648 6.249879
10 5 41.507189 24.092212 972.927809 6.207235
10 6 41.930330 23.849085 975.277662 6.333619
10 7 41.525321 24.081692 1004.874945 6.319590
10 8 41.925196 23.852006 970.216036 6.385561
10 9 41.331082 24.194866 982.164860 6.246779
cuda 1 2 True 300 300 10 0 28.109696 35.574913 965.090752 10.004997
10 1 29.774500 33.585787 976.880312 9.919405
10 2 27.870426 35.880327 962.543249 9.828210
10 3 27.885435 35.861015 951.909065 9.944797
10 4 27.480387 36.389589 978.668928 9.979367
10 5 27.832513 35.929203 980.034351 10.072112
10 6 27.539389 36.311626 986.590862 10.156035
10 7 28.420545 35.185814 962.435722 9.937167
10 8 27.704925 36.094666 979.434967 9.976506
10 9 28.175408 35.491943 972.663403 9.947777
2 2 True 300 300 10 0 38.258201 26.138186 972.539425 8.261263
10 1 38.212889 26.169181 978.279114 8.065999
10 2 35.699091 28.011918 973.811865 8.113444
10 3 36.653724 27.282357 994.790077 8.419037
10 4 37.787203 26.463985 978.210688 8.262396
10 5 37.740282 26.496887 973.612070 8.270741
10 6 37.583874 26.607156 989.429951 8.133411
10 7 37.682979 26.537180 969.769478 8.420289
10 8 37.053792 26.987791 960.559368 8.078039
10 9 36.546732 27.362227 962.187052 8.148491
4 2 True 300 300 10 0 45.336475 22.057295 994.799137 7.867455
10 1 44.779843 22.331476 976.380587 7.975012
10 2 45.576637 21.941066 965.592623 8.256733
10 3 44.983955 22.230148 973.847628 7.987708
10 4 43.230665 23.131728 975.527048 7.878542
10 5 42.922093 23.298025 987.102747 7.803261
10 6 46.138532 21.673858 965.362549 8.048773
10 7 44.891515 22.275925 976.725578 7.851899
10 8 46.558759 21.478236 963.508606 7.830173
10 9 43.252844 23.119867 970.390081 7.907033
8 2 True 300 300 10 0 48.758858 20.509094 974.528313 6.979302
10 1 47.543417 21.033406 972.609997 6.738648
10 2 48.377209 20.670891 977.948189 6.830394
10 3 46.740548 21.394700 966.236353 6.759718
10 4 48.735205 20.519048 974.811792 6.884545
10 5 48.517108 20.611286 986.879349 6.732196
10 6 47.390588 21.101236 967.341661 6.750762
10 7 49.090206 20.370662 952.601671 6.834149
10 8 48.310482 20.699441 970.391750 6.702393
10 9 47.784858 20.927131 965.574026 6.894395
16 2 True 300 300 10 0 49.848775 20.060673 979.045868 6.553441
10 1 48.719283 20.525753 983.616829 6.514989
10 2 50.066969 19.973248 971.480370 6.541945
10 3 48.540656 20.601287 975.140095 6.616287
10 4 51.361248 19.469932 988.603592 6.547920
10 5 49.087728 20.371690 977.543354 6.680205
10 6 50.312001 19.875973 977.187634 6.563298
10 7 48.251675 20.724669 963.807583 6.742224
10 8 48.084928 20.796537 976.600885 6.880574
10 9 51.159407 19.546747 965.466738 6.520070
32 2 True 300 300 10 0 49.659819 20.137005 976.863623 6.238621
10 1 53.331901 18.750504 999.082327 6.282549
10 2 53.744168 18.606670 986.496449 6.261077
10 3 50.400741 19.840978 969.493151 6.621391
10 4 52.747946 18.958084 964.337826 6.287921
10 5 51.363214 19.469187 975.760937 6.225426
10 6 53.496259 18.692896 978.708506 6.241269
10 7 52.518142 19.041039 988.959074 6.210350
10 8 51.895814 19.269377 983.508110 6.235622
10 9 53.257749 18.776610 966.600418 6.271277
tensorrt 1 2 True 300 300 10 0 27.321608 36.601067 7839.582682 9.767532
10 1 28.862141 34.647465 7773.597240 9.782195
10 2 28.424782 35.180569 7721.834660 9.779096
10 3 27.456282 36.421537 7757.497311 9.789348
10 4 27.313602 36.611795 7707.455635 9.891152
10 5 28.126850 35.553217 7734.989643 9.801149
10 6 27.824573 35.939455 7794.776917 9.782791
10 7 28.169921 35.498857 7746.086597 9.799957
10 8 26.829809 37.271976 7796.165466 9.813070
10 9 28.702160 34.840584 7715.471506 10.026932
2 2 True 300 300 10 0 36.985829 27.037382 7632.844210 8.126676
10 1 36.670869 27.269602 7687.655926 8.127272
10 2 38.436832 26.016712 7789.947987 8.062303
10 3 35.066060 28.517604 7786.698580 8.104444
10 4 37.074426 26.972771 7666.020632 8.024037
10 5 37.581685 26.608706 7781.909943 8.216202
10 6 37.255051 26.841998 7782.733679 8.075118
10 7 35.951383 27.815342 7691.067696 8.031487
10 8 37.446134 26.705027 7735.203028 8.328497
10 9 36.652603 27.283192 7770.588875 8.040667
4 2 True 300 300 10 0 44.146386 22.651911 7740.692139 7.919252
10 1 44.761683 22.340536 7769.182444 7.846475
10 2 43.392681 23.045361 7905.809402 7.800639
10 3 44.693715 22.374511 7671.643257 7.870644
10 4 42.553577 23.499787 7732.587576 7.852584
10 5 42.821742 23.352623 7811.178684 7.852226
10 6 46.288487 21.603644 7808.376551 7.758766
10 7 44.280263 22.583425 7781.347036 8.029252
10 8 44.176260 22.636592 7809.327841 7.926226
10 9 44.379356 22.533000 7739.988327 8.020282
8 2 True 300 300 10 0 48.026221 20.821959 7748.511791 6.949395
10 1 46.435372 21.535307 7780.887365 6.781697
10 2 48.929357 20.437628 7824.104071 6.719023
10 3 47.240668 21.168202 7722.230673 6.973475
10 4 48.485700 20.624638 7709.615469 6.775752
10 5 47.835064 20.905167 7706.537724 6.921723
10 6 48.050636 20.811379 7825.880289 6.969839
10 7 47.129924 21.217942 7810.398340 7.025376
10 8 47.452246 21.073818 7759.536266 6.922394
10 9 47.338171 21.124601 7723.293781 6.927058
16 2 True 300 300 10 0 49.736907 20.105794 7798.283100 6.811634
10 1 49.476083 20.211786 7794.642210 6.920174
10 2 50.439437 19.825757 7814.896822 6.792612
10 3 47.691438 20.968124 7675.437212 6.487630
10 4 49.773685 20.090938 7709.287167 6.701380
10 5 51.694456 19.344434 7704.929352 6.803341
10 6 48.661065 20.550311 7803.060770 6.765634
10 7 47.635446 20.992771 7775.330544 6.620087
10 8 50.336757 19.866198 7829.347372 6.883018
10 9 49.976701 20.009324 7808.050156 6.663896
32 2 True 300 300 10 0 51.145702 19.551985 7739.619017 6.216977
10 1 51.043740 19.591041 7721.035242 6.571375
10 2 51.182974 19.537747 7775.461912 6.222706
10 3 50.527391 19.791245 7787.240505 6.206919
10 4 51.528362 19.406788 7768.876076 6.189309
10 5 51.621468 19.371785 7786.898851 6.566465
10 6 52.143376 19.177891 7719.694853 6.319970
10 7 51.260305 19.508272 7847.770214 6.463062
10 8 53.134703 18.820092 7692.381144 6.631248
10 9 52.980898 18.874727 7763.093472 6.218217
tensorrt-dynamic 1 2 True 300 300 10 0 31.003467 32.254457 6521.617889 9.732723
10 1 31.812144 31.434536 6554.062366 9.750843
10 2 31.259486 31.990290 6575.267792 9.656787
10 3 32.676353 30.603170 6674.555779 9.833217
10 4 32.167128 31.087637 6644.479513 9.810805
10 5 32.307368 30.952692 6647.858381 9.827614
10 6 32.150361 31.103849 6569.484949 9.874940
10 7 31.519294 31.726599 6587.929964 9.768128
10 8 34.118895 29.309273 6669.947863 9.757519
10 9 33.977382 29.431343 6591.243029 9.645939
2 2 True 300 300 10 0 38.949390 25.674343 6567.115545 8.064389
10 1 39.330880 25.425315 6570.331097 8.101523
10 2 38.165079 26.201963 6622.971058 8.036017
10 3 37.756759 26.485324 6618.402481 8.016825
10 4 38.691949 25.845170 6611.620665 8.111596
10 5 37.549050 26.631832 6552.416801 7.992029
10 6 39.299740 25.445461 6609.288931 8.073509
10 7 38.597949 25.908113 6576.433897 8.067608
10 8 40.288589 24.820924 6521.219254 7.982850
10 9 37.713304 26.515841 6615.052223 8.041561
4 2 True 300 300 10 0 46.616974 21.451414 6590.298176 7.730275
10 1 45.601661 21.929026 6675.766706 7.896870
10 2 45.557207 21.950424 6571.399927 7.940561
10 3 44.636282 22.403300 6588.617086 8.052975
10 4 43.675881 22.895932 6583.994389 7.861435
10 5 45.446635 22.003829 6663.607359 7.858515
10 6 45.990554 21.743596 6600.759268 7.935882
10 7 46.355248 21.572530 6503.491163 7.827282
10 8 44.395327 22.524893 6576.805353 7.761031
10 9 45.889164 21.791637 6550.295115 8.018255
8 2 True 300 300 10 0 49.487904 20.206958 6548.888445 6.936729
10 1 48.825906 20.480931 6566.396713 6.986946
10 2 50.300009 19.880712 6565.106630 6.874979
10 3 50.178603 19.928813 6559.686184 6.937101
10 4 46.904872 21.319747 6659.745932 6.944045
10 5 49.049518 20.387560 6500.623941 6.987199
10 6 50.392320 19.844294 6513.159513 6.693140
10 7 47.459293 21.070689 6589.456081 6.958291
10 8 49.403747 20.241380 6576.894045 6.721020
10 9 47.675954 20.974934 6514.142036 6.808162
16 2 True 300 300 10 0 52.676255 18.983886 6599.140167 6.817013
10 1 52.344231 19.104302 6621.922255 6.884761
10 2 52.629326 19.000813 6682.018757 6.860964
10 3 52.352235 19.101381 6650.681973 6.515272
10 4 52.005322 19.228801 6671.128035 6.897494
10 5 50.304760 19.878834 6507.030725 6.698079
10 6 51.827640 19.294724 6568.404913 6.520748
10 7 49.720029 20.112619 6560.986042 6.504893
10 8 51.525256 19.407958 6544.950724 6.568745
10 9 51.635946 19.366354 6619.815350 6.623700
32 2 True 300 300 10 0 51.582584 19.386388 6629.342318 6.319094
10 1 52.750102 18.957309 6636.018515 6.235145
10 2 53.658631 18.636331 6637.416601 6.539639
10 3 52.206575 19.154675 6632.087469 6.554540
10 4 53.958676 18.532701 6691.053629 6.191995
10 5 52.312486 19.115895 6623.284817 6.639510
10 6 53.036880 18.854804 6683.665991 6.238684
10 7 52.403929 19.082539 6585.723400 6.232236
10 8 53.319994 18.754691 6604.930401 6.212588
10 9 50.975810 19.617148 6513.706923 6.610345
ssd-mobilenet-v1-fpn cpu 1 2 True 640 640 10 0 4.166774 239.993811 949.252129 9.547591
10 1 4.111962 243.192911 990.202427 9.488583
10 2 4.142085 241.424322 1005.813122 9.365082
10 3 4.182889 239.069223 1031.843662 9.510636
10 4 4.218243 237.065554 941.507101 9.429336
10 5 4.183807 239.016771 1013.438463 9.552360
10 6 4.162032 240.267277 1008.560419 9.738922
10 7 4.249527 235.320330 1016.711235 9.618044
10 8 4.177964 239.351034 951.983213 9.445429
10 9 4.193914 238.440752 976.160049 9.563684
2 2 True 640 640 10 0 4.154407 240.708232 1017.332315 8.106768
10 1 4.097705 244.039059 1011.907101 7.809579
10 2 4.065540 245.969772 1007.675886 8.003175
10 3 4.187782 238.789916 1001.776695 7.879436
10 4 4.133901 241.902232 968.447208 7.845998
10 5 4.162819 240.221858 1012.756824 8.244276
10 6 4.194730 238.394380 1018.407106 8.022308
10 7 4.155825 240.626097 1002.189875 8.193851
10 8 4.182516 239.090562 996.820211 7.999599
10 9 4.185278 238.932729 927.052975 7.790089
4 2 True 640 640 10 0 4.094902 244.206071 965.651274 7.668853
10 1 4.117809 242.847621 978.833675 7.734925
10 2 4.127013 242.305994 990.135908 7.600337
10 3 4.094468 244.231999 993.541956 7.612795
10 4 4.099731 243.918419 965.957642 7.927179
10 5 4.118635 242.798865 983.108044 7.523805
10 6 4.048905 246.980369 1011.893272 7.682055
10 7 4.189854 238.671780 1001.241684 7.517993
10 8 4.120459 242.691398 988.546610 7.618070
10 9 4.099926 243.906856 970.993280 7.643878
8 2 True 640 640 10 0 3.939119 253.863901 984.187841 6.399423
10 1 3.949129 253.220409 1007.894754 6.475702
10 2 3.952765 252.987474 980.275393 6.470412
10 3 3.982086 251.124680 967.080355 6.411701
10 4 3.979674 251.276881 991.317987 6.421462
10 5 3.939259 253.854841 985.423326 6.594971
10 6 3.953561 252.936542 1023.940563 6.901354
10 7 3.965071 252.202302 985.110283 6.592333
10 8 3.946019 253.419966 994.375944 6.497964
10 9 3.960012 252.524465 982.768536 6.395921
16 2 True 640 640 10 0 3.885555 257.363498 1022.043705 6.149754
10 1 3.900976 256.346077 1015.346289 6.295212
10 2 3.914565 255.456194 987.673759 6.167620
10 3 3.918962 255.169600 943.874836 6.149560
10 4 3.882112 257.591739 990.517616 6.132275
10 5 3.910913 255.694762 1012.816429 6.346256
10 6 3.920508 255.069003 974.900007 6.129012
10 7 3.923843 254.852176 995.025635 6.178208
10 8 3.905189 256.069526 963.919640 6.140187
10 9 3.905582 256.043792 981.600523 6.344199
32 2 True 640 640 10 0 3.897739 256.559022 1403.848410 5.851276
10 1 3.890113 257.061958 974.174738 6.015714
10 2 3.886433 257.305361 985.171795 5.890772
10 3 3.887184 257.255651 982.946157 5.989950
10 4 3.891006 257.002912 1027.502298 6.099410
10 5 3.889437 257.106610 1023.852587 6.042928
10 6 3.881790 257.613115 982.138157 5.939063
10 7 3.886426 257.305816 984.569788 5.910162
10 8 3.886651 257.290915 985.730886 5.911052
10 9 3.895844 256.683797 1014.714718 5.932350
cpu-prebuilt 1 2 True 640 640 10 0 4.172486 239.665270 1395.193100 9.908676
10 1 4.087677 244.637728 909.227133 9.484172
10 2 4.130065 242.126942 932.307005 9.700894
10 3 4.108261 243.412018 904.148817 9.459496
10 4 4.124635 242.445707 911.671877 9.546399
10 5 4.091377 244.416475 906.635523 9.561539
10 6 4.072824 245.529890 909.437656 9.397030
10 7 4.075099 245.392799 901.979446 9.731174
10 8 4.056446 246.521235 901.348352 9.513497
10 9 4.157077 240.553617 922.986269 9.658337
2 2 True 640 640 10 0 4.089441 244.532228 907.879114 7.840633
10 1 4.068966 245.762706 967.983723 8.200467
10 2 4.014391 249.103785 923.645258 7.966101
10 3 4.159423 240.417957 915.006638 8.030295
10 4 4.048118 247.028351 917.805433 7.984757
10 5 4.084958 244.800568 942.643404 8.256495
10 6 4.095344 244.179726 897.751331 7.796288
10 7 4.120690 242.677808 893.104553 8.310497
10 8 4.147510 241.108537 908.740520 7.971227
10 9 4.118610 242.800355 906.265259 7.853031
4 2 True 640 640 10 0 4.101861 243.791759 916.483641 7.702053
10 1 4.047837 247.045517 927.080154 7.462740
10 2 4.013608 249.152362 910.526037 7.573068
10 3 4.050227 246.899724 905.446291 7.483482
10 4 4.122979 242.543101 920.054197 7.383615
10 5 4.090524 244.467437 927.403688 7.703304
10 6 4.065780 245.955288 907.986641 7.302433
10 7 4.017075 248.937368 897.381544 7.494330
10 8 4.034593 247.856498 915.532827 7.501543
10 9 4.059147 246.357203 917.310238 7.793844
8 2 True 640 640 10 0 3.922392 254.946470 923.279047 6.436124
10 1 3.914174 255.481750 906.167507 6.409690
10 2 3.930945 254.391760 951.775789 6.742597
10 3 3.898218 256.527483 985.730171 6.353289
10 4 3.938811 253.883749 967.151880 6.462514
10 5 3.928303 254.562825 902.591705 6.386936
10 6 3.951163 253.090024 907.949686 6.553531
10 7 3.932486 254.292041 918.996096 6.339043
10 8 3.892489 256.905049 916.304827 6.409019
10 9 3.867488 258.565754 907.797575 6.496862
16 2 True 640 640 10 0 3.854445 259.440720 921.696901 6.182805
10 1 3.855517 259.368613 908.030510 6.175518
10 2 3.869401 258.437961 929.420471 6.223805
10 3 3.873543 258.161590 901.366949 6.210037
10 4 3.879507 257.764712 906.081915 6.112203
10 5 3.869912 258.403823 907.177687 6.196164
10 6 3.873590 258.158445 909.431458 6.212786
10 7 3.872074 258.259520 913.040638 6.348215
10 8 3.856121 259.327963 935.850382 6.444044
10 9 3.878208 257.851034 918.668747 6.283529
32 2 True 640 640 10 0 3.864554 258.762047 922.457933 6.037913
10 1 3.855414 259.375550 920.516729 5.923584
10 2 3.850972 259.674720 915.154219 6.017704
10 3 3.867050 258.595072 1005.593061 6.051864
10 4 3.855459 259.372473 920.384884 5.902320
10 5 3.839192 260.471471 953.571320 6.019197
10 6 3.864190 258.786455 967.902422 6.094638
10 7 3.854052 259.467207 906.492949 5.923774
10 8 3.845433 260.048725 901.516914 6.269075
10 9 3.859637 259.091735 904.022455 6.338898
cuda 1 2 True 640 640 10 0 19.573027 51.090717 908.805370 9.612441
10 1 19.350080 51.679373 902.155399 9.599686
10 2 19.189576 52.111626 919.715643 9.597778
10 3 19.237015 51.983118 907.780409 9.479403
10 4 19.720825 50.707817 900.603056 9.435534
10 5 18.962877 52.734613 913.372993 9.566069
10 6 18.795895 53.203106 927.624464 9.580970
10 7 19.962610 50.093651 904.953718 9.366274
10 8 18.800192 53.190947 904.262066 9.561539
10 9 19.563624 51.115274 909.379244 9.578466
2 2 True 640 640 10 0 21.948273 45.561671 925.212383 7.884979
10 1 22.139722 45.167685 904.875040 7.715762
10 2 22.356327 44.730067 940.769911 7.929206
10 3 22.583168 44.280767 930.185318 7.603943
10 4 22.491797 44.460654 929.051876 7.873476
10 5 22.584931 44.277310 921.633244 7.705450
10 6 22.455852 44.531822 924.786568 7.706344
10 7 21.845675 45.775652 936.403036 7.775545
10 8 22.369383 44.703960 920.533895 7.923305
10 9 22.693871 44.064760 945.333481 7.853866
4 2 True 640 640 10 0 24.841187 40.255725 909.008265 7.411778
10 1 24.960672 40.063024 920.446873 7.487774
10 2 24.869909 40.209234 909.488440 7.314205
10 3 24.831076 40.272117 921.611786 7.309943
10 4 25.481373 39.244354 910.260439 7.448256
10 5 24.773730 40.365338 923.399687 7.429630
10 6 24.993618 40.010214 906.753302 7.403255
10 7 25.702873 38.906157 928.272963 7.490605
10 8 25.486986 39.235711 920.787573 7.463127
10 9 24.908494 40.146947 914.773226 7.400990
8 2 True 640 640 10 0 25.328554 39.481133 927.286863 6.389275
10 1 24.961025 40.062457 915.333271 6.609529
10 2 25.891824 38.622230 904.504776 6.402537
10 3 24.953284 40.074885 920.495033 6.543800
10 4 26.049211 38.388878 914.555073 6.447956
10 5 25.514098 39.194018 915.846586 6.402671
10 6 25.208824 39.668649 923.800230 6.348789
10 7 25.849120 38.686037 913.218498 6.303921
10 8 25.196898 39.687425 910.559893 6.518662
10 9 25.084481 39.865285 912.072182 6.638706
16 2 True 640 640 10 0 26.573337 37.631705 907.567739 6.137401
10 1 26.623159 37.561283 924.509764 6.364942
10 2 27.052281 36.965460 919.187307 6.238095
10 3 27.121224 36.871493 923.502922 6.373614
10 4 26.488891 37.751675 913.130999 6.189905
10 5 26.545493 37.671179 917.901754 6.296627
10 6 26.711612 37.436903 928.848267 6.250955
10 7 26.795325 37.319943 956.294537 6.157666
10 8 26.468143 37.781268 915.115356 6.229423
10 9 26.731327 37.409291 904.491425 6.205052
32 2 True 640 640 10 0 26.939688 37.119955 918.742418 6.182730
10 1 27.054648 36.962226 914.777994 5.869318
10 2 27.190304 36.777817 933.827639 5.975340
10 3 27.137109 36.849909 926.579714 6.142229
10 4 26.749159 37.384354 918.555498 5.963236
10 5 26.429376 37.836686 930.832863 5.858600
10 6 26.825701 37.277684 923.490524 5.952094
10 7 26.777069 37.345387 920.230150 5.977575
10 8 26.844947 37.250958 925.060749 5.984571
10 9 26.654104 37.517674 928.603172 5.865470
tensorrt 1 2 True 640 640 10 0 19.431928 51.461697 19697.484970 9.247065
10 1 20.250110 49.382448 19534.241676 9.296894
10 2 19.502223 51.276207 19808.818340 9.350300
10 3 20.191522 49.525738 19857.414722 9.371519
10 4 19.747938 50.638199 19568.608046 9.309173
10 5 20.043123 49.892426 19725.984812 9.344578
10 6 19.708779 50.738811 19778.983593 9.471059
10 7 19.915217 50.212860 19713.012457 9.250283
10 8 19.488449 51.312447 19635.494947 9.330153
10 9 19.188347 52.114964 19832.072496 9.377003
2 2 True 640 640 10 0 22.928527 43.613791 21607.429504 7.562697
10 1 22.794449 43.870330 21589.494705 7.522821
10 2 22.722702 44.008851 21722.892046 7.638097
10 3 23.202242 43.099284 21588.389874 7.699311
10 4 22.835838 43.790817 21966.397762 7.536530
10 5 22.663828 44.123173 21663.414240 7.575333
10 6 22.792096 43.874860 21542.892933 7.686675
10 7 22.051724 45.347929 21649.904966 7.707298
10 8 22.281979 44.879317 21570.734739 7.615447
10 9 22.871202 43.723106 21798.259735 7.703841
4 2 True 640 640 10 0 25.176272 39.719939 11973.683357 7.282495
10 1 24.944937 40.088296 11978.130341 7.464558
10 2 25.543018 39.149642 12047.371149 7.466644
10 3 25.446628 39.297938 12042.906523 7.524014
10 4 25.667953 38.959086 12023.584127 7.380068
10 5 25.583838 39.087176 12001.087666 7.340968
10 6 25.606877 39.052010 11880.704880 7.398009
10 7 25.328611 39.481044 12004.886150 7.490426
10 8 25.458714 39.279282 12014.826536 7.544011
10 9 25.397551 39.373875 12006.185532 7.362038
8 2 True 640 640 10 0 25.097521 39.844573 13302.614450 6.381348
10 1 25.150250 39.761037 13281.148195 6.316394
10 2 25.261242 39.586335 13336.726189 6.398126
10 3 25.413016 39.349914 13257.743835 6.481037
10 4 25.249704 39.604425 13518.876791 6.493971
10 5 25.795009 38.767189 13310.051680 6.457239
10 6 25.376385 39.406717 13284.626007 6.421357
10 7 25.492059 39.227903 13395.771027 6.501481
10 8 25.256127 39.594352 13550.631762 6.353378
10 9 25.049351 39.921194 13315.902948 6.421164
16 2 True 640 640 10 0 25.640180 39.001286 16542.674780 6.298348
10 1 25.066698 39.893568 16451.378345 6.477304
10 2 25.613591 39.041772 16377.570391 6.301396
10 3 25.727941 38.868248 16550.896883 6.395690
10 4 25.279684 39.557457 16486.351967 6.212004
10 5 26.107898 38.302585 16543.775082 6.223150
10 6 25.412680 39.350435 16419.026852 6.312534
10 7 25.515922 39.191216 16384.063244 6.471224
10 8 25.295892 39.532110 16443.823814 6.333791
10 9 25.614960 39.039686 16459.635496 6.241322
32 2 True 640 640 10 0 25.498979 39.217256 22580.085278 5.934775
10 1 25.200389 39.681926 22651.859522 5.900986
10 2 25.942776 38.546376 22777.650595 5.904812
10 3 25.432211 39.320216 22732.622623 5.968172
10 4 25.543134 39.149463 22682.816267 6.100923
10 5 25.276280 39.562784 22743.459463 6.045006
10 6 25.224363 39.644212 22797.020435 5.873557
10 7 25.824476 38.722955 22711.929560 6.002318
10 8 25.430418 39.322987 22762.054443 6.018624
10 9 25.681106 38.939133 22828.942776 5.862597
tensorrt-dynamic 1 2 True 640 640 10 0 23.651334 42.280912 6379.939556 9.138942
10 1 22.586938 44.273376 6337.580919 9.437680
10 2 22.947532 43.577671 6281.933784 9.322882
10 3 22.223597 44.997215 6375.229359 9.170175
10 4 23.297677 42.922735 6335.123301 9.294748
10 5 23.255179 43.001175 6332.339287 9.318829
10 6 23.215533 43.074608 6294.281244 9.399056
10 7 23.157341 43.182850 6345.356703 9.421587
10 8 23.110259 43.270826 6355.256796 9.445548
10 9 22.905425 43.657780 6455.956936 9.447217
2 2 True 640 640 10 0 25.801892 38.756847 6311.048508 7.634759
10 1 26.239656 38.110256 6313.688993 7.738590
10 2 25.808005 38.747668 6355.959177 7.593751
10 3 26.471672 37.776232 6325.766563 7.632911
10 4 26.796898 37.317753 6292.751551 7.551193
10 5 26.432468 37.832260 6369.322538 7.606745
10 6 26.197535 38.171530 6398.904562 8.122444
10 7 26.126794 38.274884 6403.906345 7.721722
10 8 26.200808 38.166761 6384.660721 7.561564
10 9 27.021237 37.007928 6344.237089 7.621288
4 2 True 640 640 10 0 29.037478 34.438252 6343.791246 7.474542
10 1 28.185254 35.479546 6421.319008 7.432252
10 2 29.108413 34.354329 6373.960495 7.496566
10 3 28.662588 34.888685 6463.953733 7.437229
10 4 28.786274 34.738779 6337.746382 7.398278
10 5 28.889127 34.615099 6300.127268 7.326245
10 6 27.752679 36.032557 6315.848827 7.420301
10 7 28.809556 34.710705 6358.988285 7.201135
10 8 28.539864 35.038710 6364.130735 7.515818
10 9 29.355791 34.064829 6352.153301 7.515222
8 2 True 640 640 10 0 29.055934 34.416378 6499.320030 6.336465
10 1 28.686921 34.859091 6326.475620 6.397486
10 2 29.407272 34.005195 6237.746716 6.408036
10 3 28.364759 35.255015 6369.799614 6.522298
10 4 28.367253 35.251915 6358.793736 6.287098
10 5 28.584551 34.983933 6367.268085 6.484717
10 6 28.227340 35.426646 6397.254944 6.266743
10 7 28.760909 34.769416 6262.703419 6.326750
10 8 28.986106 34.499288 6388.038158 6.303608
10 9 28.658377 34.893811 6389.622927 6.511480
16 2 True 640 640 10 0 29.197202 34.249857 6348.584414 6.167918
10 1 28.978346 34.508526 6252.276421 6.100789
10 2 29.110219 34.352198 6358.824015 6.137222
10 3 28.091987 35.597339 6363.001823 6.313227
10 4 28.525913 35.055846 6329.951763 6.257400
10 5 28.889649 34.614474 6311.621428 6.251894
10 6 28.824578 34.692615 6385.085344 6.299280
10 7 28.331457 35.296455 6438.660860 6.476067
10 8 28.771859 34.756184 6343.760490 6.229334
10 9 28.315033 35.316929 6278.670788 6.128475
32 2 True 640 640 10 0 28.655912 34.896813 6336.538076 5.954769
10 1 28.375175 35.242073 6304.973125 5.921368
10 2 28.241969 35.408296 6316.627264 5.957868
10 3 28.419750 35.186797 6391.951561 5.892187
10 4 28.534015 35.045892 6357.141733 5.938355
10 5 28.553780 35.021633 6352.425098 5.932610
10 6 28.431785 35.171904 6358.306885 6.077591
10 7 28.640527 34.915559 6339.262486 5.921669
10 8 28.113447 35.570167 6322.747946 5.909916
10 9 28.453580 35.144962 6266.034603 5.850889
ssd-mobilenet-v1-non-quantized-mlperf cpu 1 2 True 300 300 10 0 44.508983 22.467375 855.396509 10.349393
10 1 45.153936 22.146463 860.897064 10.306597
10 2 44.406255 22.519350 849.796295 10.158300
10 3 42.634573 23.455143 790.455103 10.026693
10 4 45.146160 22.150278 862.575769 10.165095
10 5 44.147315 22.651434 855.742931 10.263205
10 6 42.617678 23.464441 794.686079 9.904742
10 7 42.135600 23.732901 842.722654 9.894967
10 8 44.101361 22.675037 891.747236 11.799097
10 9 43.114903 23.193836 847.278595 10.382295
2 2 True 300 300 10 0 54.922271 18.207550 844.219685 8.467555
10 1 53.679877 18.628955 861.808300 8.341193
10 2 54.998971 18.182158 838.324308 8.305073
10 3 58.247346 17.168164 824.517250 8.350492
10 4 57.007965 17.541409 823.447227 8.308351
10 5 56.391729 17.733097 875.511646 8.492947
10 6 56.694724 17.638326 816.074848 8.246958
10 7 56.834928 17.594814 866.644382 8.589089
10 8 55.320324 18.076539 815.828323 8.362293
10 9 53.852872 18.569112 844.886780 8.561969
4 2 True 300 300 10 0 63.342040 15.787303 855.436325 7.863492
10 1 62.332221 16.043067 844.418526 8.021593
10 2 63.359024 15.783072 848.405600 7.992387
10 3 65.518852 15.262783 854.447126 7.989645
10 4 62.017936 16.124368 849.282026 8.039773
10 5 62.335232 16.042292 843.133688 7.888258
10 6 67.212371 14.878213 864.814758 8.324951
10 7 63.548616 15.735984 859.345913 8.412153
10 8 65.048391 15.373170 838.207245 8.437037
10 9 61.140344 16.355813 828.130722 8.056521
8 2 True 300 300 10 0 70.252965 14.234275 811.311960 6.807566
10 1 69.835522 14.319360 837.511301 6.794080
10 2 70.439083 14.196664 808.301926 6.772205
10 3 66.213525 15.102655 833.750248 6.830335
10 4 71.619763 13.962626 853.405714 6.823197
10 5 68.943308 14.504671 809.007645 7.050350
10 6 70.880411 14.108270 868.093491 7.027805
10 7 70.659058 14.152467 850.801706 6.897926
10 8 68.717542 14.552325 820.596933 6.741554
10 9 70.115999 14.262080 829.351425 6.841555
16 2 True 300 300 10 0 68.719301 14.551952 781.180143 6.951325
10 1 69.414681 14.406174 863.272190 6.765753
10 2 67.408630 14.834896 849.275351 6.513953
10 3 66.561198 15.023768 852.967024 6.759889
10 4 68.436393 14.612108 866.723061 6.660096
10 5 68.497095 14.599159 839.836121 6.589592
10 6 70.165043 14.252111 851.131439 6.539963
10 7 68.609000 14.575347 826.849222 7.227622
10 8 66.593960 15.016377 862.487555 6.580189
10 9 69.750526 14.336810 842.926502 7.111885
32 2 True 300 300 10 0 66.393080 15.061811 1078.144073 6.210353
10 1 66.788014 14.972746 863.686085 6.306849
10 2 66.606914 15.013456 870.333433 6.739758
10 3 64.867184 15.416116 871.635437 6.371699
10 4 66.392686 15.061900 851.468325 6.762244
10 5 66.775919 14.975458 860.826015 6.418712
10 6 67.615711 14.789462 853.632450 6.495692
10 7 66.320938 15.078194 867.979050 6.417651
10 8 66.948122 14.936939 868.599415 6.417613
10 9 66.927491 14.941543 862.969875 6.263901
cpu-prebuilt 1 2 True 300 300 10 0 40.713888 24.561644 1037.388802 9.943485
10 1 42.277029 23.653507 771.442175 10.058641
10 2 42.541600 23.506403 780.729771 10.006428
10 3 44.113884 22.668600 790.134907 10.029793
10 4 43.597568 22.937059 765.999556 9.960651
10 5 42.141950 23.729324 774.374962 10.036230
10 6 42.938351 23.289204 772.964001 10.028243
10 7 44.268462 22.589445 785.434484 9.983778
10 8 44.407195 22.518873 779.783487 9.913445
10 9 43.708879 22.878647 792.213917 9.917259
2 2 True 300 300 10 0 55.870046 17.898679 777.980089 8.077025
10 1 56.983181 17.549038 773.367167 8.310854
10 2 55.902438 17.888308 772.530079 8.320510
10 3 53.071630 18.842459 771.639109 8.308947
10 4 55.269296 18.093228 764.454842 8.248508
10 5 53.726978 18.612623 780.973673 7.923067
10 6 57.128707 17.504334 774.303675 8.177876
10 7 55.807601 17.918706 771.489859 8.307040
10 8 50.296541 19.882083 766.900301 8.293748
10 9 55.417903 18.044710 764.200687 8.201241
4 2 True 300 300 10 0 62.462410 16.009629 774.709702 8.008301
10 1 63.380805 15.777647 774.232388 8.020818
10 2 65.092304 15.362799 768.320560 7.854253
10 3 61.739730 16.197026 795.464516 8.208185
10 4 61.302533 16.312540 909.101248 8.582771
10 5 63.020355 15.867889 763.962030 7.897943
10 6 63.541877 15.737653 767.083645 7.900655
10 7 63.157479 15.833437 760.612011 7.863700
10 8 63.071292 15.855074 794.938564 7.943928
10 9 62.629362 15.966952 766.959667 8.274227
8 2 True 300 300 10 0 68.078713 14.688879 777.205944 6.846160
10 1 67.215602 14.877498 785.662174 6.964460
10 2 69.126814 14.466166 762.176514 6.857082
10 3 70.328355 14.219016 797.222376 7.128149
10 4 67.411000 14.834374 767.273903 6.741390
10 5 67.367691 14.843911 763.527870 6.913438
10 6 68.967114 14.499664 763.197184 6.917983
10 7 66.953933 14.935642 763.759375 6.836981
10 8 68.047235 14.695674 804.507256 6.786123
10 9 62.645965 15.962720 763.679981 6.918207
16 2 True 300 300 10 0 66.415092 15.056819 788.475513 6.619506
10 1 67.211429 14.878422 829.724312 6.628186
10 2 65.238921 15.328273 779.192448 6.518036
10 3 65.852529 15.185446 767.430544 6.868519
10 4 66.074143 15.134513 788.130522 6.531455
10 5 67.154865 14.890954 771.184921 6.687127
10 6 67.007950 14.923602 773.524284 6.542481
10 7 67.325045 14.853314 769.742727 7.126704
10 8 64.996478 15.385449 767.860651 6.565034
10 9 66.780704 14.974385 764.333487 6.552696
32 2 True 300 300 10 0 65.663865 15.229076 788.587332 6.262325
10 1 64.805796 15.430719 771.837950 6.273981
10 2 66.753900 14.980398 787.343979 6.266356
10 3 65.936383 15.166134 763.461351 6.200690
10 4 64.695433 15.457042 775.409937 6.465737
10 5 66.763397 14.978267 781.033039 6.326370
10 6 64.756425 15.442483 768.030882 6.328046
10 7 64.535759 15.495285 779.752254 6.213501
10 8 65.751136 15.208863 764.954567 6.244205
10 9 65.092904 15.362658 775.216103 6.226167
cuda 1 2 True 300 300 10 0 34.838148 28.704166 781.345367 10.071874
10 1 35.993341 27.782917 783.009768 9.917855
10 2 35.707752 28.005123 779.697180 9.979963
10 3 34.500292 28.985262 784.879208 10.042787
10 4 35.513950 28.157949 774.436474 9.806156
10 5 33.792875 29.592037 778.587103 9.847999
10 6 35.608320 28.083324 777.881622 9.941220
10 7 32.425756 30.839682 786.830187 10.071516
10 8 33.300020 30.030012 774.903297 9.952307
10 9 35.324322 28.309107 770.522356 9.963036
2 2 True 300 300 10 0 44.507330 22.468209 778.796196 8.328557
10 1 43.788049 22.837281 774.612427 8.106649
10 2 44.823391 22.309780 772.593975 8.289516
10 3 44.332097 22.557020 776.311636 8.288622
10 4 45.798845 21.834612 771.892548 8.101702
10 5 42.399256 23.585320 779.415607 8.286357
10 6 45.019444 22.212625 780.917883 8.302510
10 7 43.681111 22.893190 786.888838 8.253455
10 8 45.360991 22.045374 776.494265 8.383989
10 9 41.860157 23.889065 766.506195 8.248746
4 2 True 300 300 10 0 48.156008 20.765841 777.790785 8.004785
10 1 45.749264 21.858275 765.287399 7.831812
10 2 51.807275 19.302309 790.815353 7.882565
10 3 49.348527 20.264030 774.976254 8.005470
10 4 48.305475 20.701587 773.345232 7.986993
10 5 46.588107 21.464705 773.250103 7.926434
10 6 48.916446 20.443022 775.005817 7.983893
10 7 47.273613 21.153450 782.892227 8.123457
10 8 47.400630 21.096766 774.888039 8.026481
10 9 47.832268 20.906389 785.664797 7.982075
8 2 True 300 300 10 0 51.164556 19.544780 776.380301 6.983578
10 1 51.929225 19.256979 781.573534 6.825358
10 2 53.323468 18.753469 766.869307 6.804869
10 3 50.549466 19.782603 774.079561 6.917417
10 4 50.507084 19.799203 778.929710 6.971538
10 5 51.109379 19.565880 777.666569 6.716475
10 6 51.273465 19.503266 781.795979 6.735116
10 7 53.648636 18.639803 786.628246 6.750897
10 8 52.151825 19.174784 781.433582 7.006764
10 9 51.116153 19.563287 769.179583 6.954193
16 2 True 300 300 10 0 50.197783 19.921198 792.678595 6.564252
10 1 55.402394 18.049762 781.792879 6.630279
10 2 54.092962 18.486694 793.374062 6.697290
10 3 52.851113 18.921077 785.227776 6.510057
10 4 52.790747 18.942714 785.464764 6.560571
10 5 50.150667 19.939914 790.877581 6.897539
10 6 52.605480 19.009426 783.695459 6.516472
10 7 52.157418 19.172728 778.192520 6.834872
10 8 51.597119 19.380927 774.550915 6.571859
10 9 53.542313 18.676817 777.358055 6.890178
32 2 True 300 300 10 0 53.818149 18.581092 779.099703 6.274331
10 1 54.704909 18.279895 789.219856 6.233845
10 2 52.738370 18.961526 783.346176 6.209411
10 3 53.099324 18.832631 777.828693 6.456271
10 4 54.019571 18.511809 770.405769 6.422836
10 5 53.729559 18.611729 801.283598 6.235909
10 6 52.587157 19.016050 772.557497 6.303951
10 7 54.049134 18.501684 784.798145 6.272491
10 8 55.155167 18.130668 781.807423 6.262083
10 9 52.850239 18.921390 781.973362 6.223582
tensorrt 1 2 True 300 300 10 0 34.491213 28.992891 5693.123102 9.831667
10 1 34.909769 28.645277 5614.737511 9.802699
10 2 34.405768 29.064894 5548.023462 9.771824
10 3 33.113885 30.198812 5601.720810 9.880662
10 4 33.933400 29.469490 5567.697525 9.871960
10 5 36.429127 27.450562 5629.716873 9.792566
10 6 35.355885 28.283834 5636.938572 9.807348
10 7 34.066797 29.354095 5734.992504 10.001659
10 8 35.254547 28.365135 5749.200821 9.958982
10 9 32.603454 30.671597 5569.962025 9.765744
2 2 True 300 300 10 0 41.345392 24.186492 5669.808388 8.022785
10 1 42.247433 23.670077 5584.904432 7.995546
10 2 43.110915 23.195982 5624.918938 8.073151
10 3 43.516149 22.979975 5633.650541 8.129537
10 4 44.190573 22.629261 5563.676834 8.092761
10 5 42.802514 23.363113 5689.352512 8.023202
10 6 40.245099 24.847746 5572.940826 8.210301
10 7 41.530641 24.078608 5571.259499 8.071780
10 8 41.993012 23.813486 5589.100838 8.057117
10 9 43.195271 23.150682 5695.473194 8.090854
4 2 True 300 300 10 0 48.249906 20.725429 5556.540966 7.749975
10 1 51.402517 19.454300 5524.202108 7.788837
10 2 50.275139 19.890547 5638.597012 7.917553
10 3 48.851497 20.470202 5544.986963 7.811934
10 4 48.008012 20.829856 5682.790756 7.802725
10 5 47.691675 20.968020 5664.843082 8.069426
10 6 47.879909 20.885587 5650.647640 7.854670
10 7 47.997299 20.834506 5630.479336 7.815361
10 8 47.534796 21.037221 5514.477968 7.812589
10 9 47.777782 20.930231 5590.559244 7.780701
8 2 True 300 300 10 0 50.022484 19.991010 5636.264801 6.643206
10 1 52.279151 19.128084 5630.309820 6.805852
10 2 49.403165 20.241618 5643.610477 6.730497
10 3 51.554468 19.396961 5637.037992 6.989181
10 4 49.197883 20.326078 5590.353727 6.730586
10 5 52.063461 19.207329 5637.848139 6.794155
10 6 50.623136 19.753814 5649.809122 6.753027
10 7 51.294236 19.495368 5627.628326 6.769717
10 8 51.352248 19.473344 5567.685366 6.736740
10 9 51.455487 19.434273 5618.886232 6.746203
16 2 True 300 300 10 0 52.854860 18.919736 5635.564804 6.518625
10 1 50.472707 19.812688 5616.070032 6.881125
10 2 51.503744 19.416064 5620.897532 6.502539
10 3 52.060472 19.208431 5690.553904 6.879076
10 4 52.010119 19.227028 5623.275042 6.887466
10 5 51.807155 19.302353 5606.832981 6.520502
10 6 55.841826 17.907724 5690.791845 6.719068
10 7 50.728944 19.712612 5658.821583 6.842412
10 8 53.442538 18.711686 5633.854151 6.828517
10 9 53.552225 18.673360 5628.093243 6.515622
32 2 True 300 300 10 0 53.113382 18.827647 5677.363634 6.352689
10 1 53.698669 18.622436 5608.144283 6.485902
10 2 54.095621 18.485785 5602.419615 6.409224
10 3 55.063932 18.160708 5697.304487 6.222568
10 4 51.639145 19.365154 5669.396877 6.242018
10 5 53.686082 18.626802 5628.892422 6.337665
10 6 56.188279 17.797306 5633.605242 6.216008
10 7 52.798887 18.939793 5645.969152 6.189067
10 8 52.907468 18.900923 5606.514692 6.473586
10 9 54.338853 18.403038 5652.473211 6.511830
tensorrt-dynamic 1 2 True 300 300 10 0 33.075499 30.233860 4428.390741 9.896278
10 1 35.898765 27.856112 4479.261875 9.874344
10 2 35.232633 28.382778 4560.572863 9.808779
10 3 35.927978 27.833462 4453.529119 9.793162
10 4 34.725084 28.797626 4468.748808 9.745121
10 5 33.368370 29.968500 4467.795610 9.806395
10 6 34.366865 29.097795 4409.515142 9.976268
10 7 35.376760 28.267145 4497.458935 9.778738
10 8 32.995618 30.307055 4504.912615 9.788394
10 9 32.630341 30.646324 4538.682461 9.847879
2 2 True 300 300 10 0 44.159865 22.644997 4527.956724 8.072674
10 1 43.264556 23.113608 4500.693560 8.028150
10 2 41.775520 23.937464 4431.074381 8.014083
10 3 41.853265 23.892999 4470.486164 8.065104
10 4 42.518528 23.519158 4494.201183 8.137047
10 5 43.462714 23.008227 4473.366261 8.062184
10 6 45.637386 21.911860 4491.267443 8.057714
10 7 42.913969 23.302436 4471.668482 8.096933
10 8 41.951220 23.837209 4529.442310 8.108199
10 9 42.023094 23.796439 4510.761976 8.094668
4 2 True 300 300 10 0 49.638937 20.145476 4398.817539 8.087516
10 1 48.622142 20.566761 4474.259615 7.867664
10 2 48.535916 20.603299 4536.252737 8.033156
10 3 49.745497 20.102322 4426.193714 8.049458
10 4 48.561344 20.592511 4510.492563 7.939309
10 5 48.524265 20.608246 4523.043633 7.985383
10 6 48.355597 20.680130 4416.669130 7.866770
10 7 47.774381 20.931721 4529.404402 7.852346
10 8 48.585533 20.582259 4480.943680 8.084804
10 9 47.964230 20.848870 4440.499544 7.843494
8 2 True 300 300 10 0 49.644224 20.143330 4441.440582 6.721780
10 1 50.144783 19.942254 4443.773985 6.800517
10 2 51.281614 19.500166 4456.780910 6.947145
10 3 51.593945 19.382119 4501.610041 6.735057
10 4 50.013910 19.994438 4496.972561 6.897643
10 5 52.018020 19.224107 4485.256433 6.950125
10 6 52.346926 19.103318 4493.339300 6.978855
10 7 51.063412 19.583493 4480.364561 6.962791
10 8 50.863554 19.660443 4510.897398 6.955460
10 9 51.222275 19.522756 4498.049259 6.953925
16 2 True 300 300 10 0 55.135931 18.136993 4443.065166 6.560914
10 1 53.695210 18.623635 4502.415180 6.883219
10 2 52.671252 18.985689 4430.941820 6.867565
10 3 53.916820 18.547088 4419.278145 6.855160
10 4 52.181184 19.163996 4415.038824 6.553091
10 5 51.752221 19.322842 4370.236158 6.860144
10 6 51.836888 19.291282 4543.978930 6.480657
10 7 53.332155 18.750414 4506.866932 6.521039
10 8 53.612891 18.652231 4417.213440 6.573483
10 9 52.644684 18.995270 4505.507708 6.823301
32 2 True 300 300 10 0 53.851619 18.569544 4444.882870 6.220557
10 1 55.655008 17.967835 4402.207851 6.254584
10 2 56.931938 17.564833 4446.827650 6.302960
10 3 57.242784 17.469451 4511.808872 6.349295
10 4 54.578289 18.322304 4474.316120 6.366946
10 5 54.404888 18.380702 4485.224247 6.514076
10 6 53.523417 18.683411 4479.867935 6.208107
10 7 53.068734 18.843487 4491.706133 6.267328
10 8 55.089674 18.152222 4475.549459 6.558314
10 9 54.115993 18.478826 4445.625305 6.254699
ssd-mobilenet-v1-quantized-mlperf cpu 1 2 True 300 300 10 0 41.348449 24.184704 1141.518593 9.508848
10 1 40.496114 24.693727 1171.593904 10.394216
10 2 41.953528 23.835897 1141.031742 10.167122
10 3 41.385168 24.163246 1137.880325 9.997845
10 4 39.455007 25.345325 1147.413731 9.865284
10 5 43.608900 22.931099 1123.368263 10.007501
10 6 36.653564 27.282476 1281.663418 10.389447
10 7 39.008073 25.635719 1127.808571 10.025382
10 8 42.125443 23.738623 1143.161297 10.257363
10 9 40.091226 24.943113 1132.368088 10.350823
2 2 True 300 300 10 0 54.527070 18.339515 1154.313564 8.379519
10 1 52.303272 19.119263 1100.859404 8.277714
10 2 48.142051 20.771861 1136.225939 8.205116
10 3 50.676042 19.733191 1108.897209 8.291483
10 4 46.849337 21.345019 1142.849684 8.236468
10 5 51.675628 19.351482 1120.688438 8.327782
10 6 52.557237 19.026875 1108.834267 8.068800
10 7 50.874273 19.656301 1104.010820 8.074701
10 8 52.412421 19.079447 1131.302118 8.441091
10 9 51.449018 19.436717 1149.085522 8.410871
4 2 True 300 300 10 0 55.355917 18.064916 1158.612013 8.156538
10 1 55.221452 18.108904 1160.176277 8.203596
10 2 55.130903 18.138647 1082.293510 8.134514
10 3 59.414665 16.830862 1095.991850 8.108407
10 4 54.393422 18.384576 1091.183186 8.103549
10 5 53.077003 18.840551 1098.875284 7.998437
10 6 54.653363 18.297136 1150.421858 8.282721
10 7 53.999337 18.518746 1106.023312 7.997632
10 8 55.504991 18.016398 1161.789656 8.129805
10 9 54.145552 18.468738 1119.019747 8.028477
8 2 True 300 300 10 0 63.781389 15.678555 1130.697250 6.893709
10 1 63.636597 15.714228 1124.075413 6.753534
10 2 61.982765 16.133517 1108.465195 6.904945
10 3 62.164084 16.086459 1108.081102 6.685421
10 4 63.903466 15.648603 1153.816938 6.953776
10 5 63.158787 15.833110 1165.574312 6.874070
10 6 65.668748 15.227944 1097.320318 6.773993
10 7 65.437460 15.281767 1131.476164 6.856203
10 8 64.929819 15.401244 1157.756805 6.803423
10 9 64.890266 15.410632 1127.169847 6.915033
16 2 True 300 300 10 0 61.278967 16.318813 1116.379499 6.518774
10 1 60.619049 16.496465 1111.896038 6.598473
10 2 60.792907 16.449288 1105.592966 6.524555
10 3 61.690240 16.210020 1125.183582 6.771743
10 4 58.035130 17.230943 1161.416769 6.706670
10 5 61.566240 16.242668 1137.651682 6.813996
10 6 61.949522 16.142175 1147.143364 6.436564
10 7 59.478540 16.812786 1132.051229 6.585948
10 8 61.019874 16.388103 1106.467009 6.517507
10 9 60.501496 16.528517 1114.739418 6.584004
32 2 True 300 300 10 0 60.571968 16.509287 1335.273027 6.240573
10 1 58.888475 16.981252 1139.098406 6.271660
10 2 59.194109 16.893573 1164.017200 6.336439
10 3 60.497487 16.529612 1140.981436 6.424516
10 4 57.294612 17.453648 1146.106958 6.223690
10 5 57.372592 17.429926 1160.487175 6.275326
10 6 57.879513 17.277271 1129.624128 6.222263
10 7 57.862246 17.282426 1111.538410 6.232411
10 8 60.131693 16.630165 1117.774963 6.223902
10 9 57.830558 17.291896 1154.215574 6.232198
cpu-prebuilt 1 2 True 300 300 10 0 40.441059 24.727345 1309.064388 9.849906
10 1 37.045610 26.993752 1047.098160 9.984732
10 2 43.521567 22.977114 1052.574873 9.961367
10 3 37.864298 26.410103 1045.958281 9.989738
10 4 37.574952 26.613474 1031.730890 10.116339
10 5 39.878149 25.076389 1095.279455 9.931684
10 6 40.063654 24.960279 1048.730612 10.042787
10 7 36.900576 27.099848 1040.490389 10.020614
10 8 40.199969 24.875641 1122.545481 10.402083
10 9 39.426450 25.363684 1047.638655 10.037899
2 2 True 300 300 10 0 50.685534 19.729495 1040.905714 8.306503
10 1 48.573857 20.587206 1037.618876 8.160651
10 2 51.053234 19.587398 1043.580294 8.207619
10 3 46.482008 21.513700 1048.915386 8.179367
10 4 48.895490 20.451784 1034.586668 8.265018
10 5 48.553896 20.595670 1055.195570 8.268952
10 6 49.706146 20.118237 1033.941031 8.346558
10 7 52.070490 19.204736 1167.417288 8.097112
10 8 49.188219 20.330071 1048.541069 8.374274
10 9 52.884933 18.908978 1043.299913 8.308649
4 2 True 300 300 10 0 55.742704 17.939568 1098.007917 8.038461
10 1 64.986156 15.387893 1042.261124 7.875025
10 2 57.551021 17.375886 1044.894695 8.018285
10 3 64.205000 15.575111 1042.172432 7.973731
10 4 58.329767 17.143905 1032.604456 7.848501
10 5 57.594683 17.362714 1057.824612 7.987052
10 6 60.255630 16.595960 1063.431978 7.877618
10 7 54.803683 18.246949 1043.725729 8.001059
10 8 55.601933 17.984986 1052.542448 8.195430
10 9 57.626731 17.353058 1188.806295 7.968754
8 2 True 300 300 10 0 61.420131 16.281307 1038.946867 6.855279
10 1 59.209176 16.889274 1118.232012 6.844476
10 2 60.896120 16.421407 1035.385132 6.851062
10 3 61.316760 16.308755 1037.714481 6.793469
10 4 63.736926 15.689492 1068.039417 6.991088
10 5 63.553912 15.734673 1049.790144 6.952107
10 6 61.231386 16.331494 1031.262636 6.818563
10 7 62.301089 16.051084 1056.199789 6.892756
10 8 62.549739 15.987277 1022.531748 6.714419
10 9 67.521012 14.810205 1041.496992 6.728142
16 2 True 300 300 10 0 60.322466 16.577572 1165.717125 6.635241
10 1 61.596246 16.234756 1067.545891 6.781265
10 2 58.108648 17.209142 1034.183741 6.792784
10 3 60.298455 16.584173 1027.941942 6.518237
10 4 59.369461 16.843677 1031.779289 6.543800
10 5 59.765905 16.731948 1041.755676 6.752186
10 6 59.762233 16.732976 1048.759222 6.551087
10 7 60.554341 16.514093 1062.961578 6.571285
10 8 59.123575 16.913727 1081.460476 6.551549
10 9 55.715399 17.948359 1028.668642 6.500758
32 2 True 300 300 10 0 58.722987 17.029107 1089.289427 6.313432
10 1 56.852863 17.589264 1098.215103 6.244589
10 2 56.595440 17.669268 1055.323839 6.296054
10 3 59.791731 16.724721 1037.086487 6.419033
10 4 56.710342 17.633468 1034.259796 6.256852
10 5 57.984183 17.246082 1064.899206 6.237723
10 6 58.078550 17.218061 1122.566223 6.265502
10 7 58.812017 17.003328 1052.809000 6.317627
10 8 57.999869 17.241418 1058.693171 6.457172
10 9 57.860250 17.283022 1046.101570 6.279267
cuda 1 2 True 300 300 10 0 47.736863 20.948172 1072.395086 9.973407
10 1 44.426951 22.508860 1061.632156 9.851217
10 2 45.339906 22.055626 1052.780151 9.912848
10 3 45.633911 21.913528 1051.055908 9.842157
10 4 45.379149 22.036552 1045.479774 10.001063
10 5 47.179492 21.195650 1213.608742 9.879351
10 6 46.484584 21.512508 1034.584761 9.995222
10 7 46.701451 21.412611 1059.213400 10.006666
10 8 45.686600 21.888256 1056.025982 10.016084
10 9 46.721740 21.403313 1036.118746 10.092020
2 2 True 300 300 10 0 58.822431 17.000318 1072.357416 8.326054
10 1 57.381152 17.427325 1038.028002 8.314252
10 2 57.955590 17.254591 1023.281574 8.316636
10 3 59.619255 16.773105 1030.957222 8.168757
10 4 56.107337 17.822981 1056.241989 8.212805
10 5 57.715941 17.326236 1067.446709 8.144021
10 6 57.721898 17.324448 1051.792383 8.059025
10 7 54.800282 18.248081 1052.458763 8.316636
10 8 57.631878 17.351508 1046.841145 8.296788
10 9 56.964994 17.554641 1033.020496 8.300185
4 2 True 300 300 10 0 65.204375 15.336394 1040.844440 8.035362
10 1 65.846973 15.186727 1054.573298 7.947505
10 2 69.567952 14.374435 1053.631783 7.832497
10 3 64.892650 15.410066 1065.173388 7.954657
10 4 65.823980 15.192032 1044.612885 7.860094
10 5 63.247240 15.810966 1047.475338 7.912278
10 6 65.543681 15.257001 1056.888580 7.834315
10 7 65.533952 15.259266 1036.722898 7.966399
10 8 67.347273 14.848411 1054.852724 7.903993
10 9 64.580909 15.484452 1056.848288 7.932007
8 2 True 300 300 10 0 74.732140 13.381124 1029.964924 6.796435
10 1 74.683903 13.389766 1054.652691 6.828427
10 2 72.663400 13.762087 1051.391840 6.789342
10 3 73.634178 13.580650 1047.311306 6.884679
10 4 72.830264 13.730556 1052.465200 6.779656
10 5 74.047729 13.504803 1062.226772 6.733328
10 6 72.906696 13.716161 1042.716980 6.745428
10 7 72.332110 13.825119 1049.687862 6.696597
10 8 72.947272 13.708532 1049.215794 6.844774
10 9 71.459915 13.993859 1044.232845 6.917447
16 2 True 300 300 10 0 76.220230 13.119876 1058.016539 6.763577
10 1 77.116415 12.967408 1043.073893 6.593823
10 2 76.519544 13.068557 1061.495543 6.925061
10 3 78.035741 12.814641 1043.457985 6.579340
10 4 77.564478 12.892500 1060.048103 6.749853
10 5 76.995467 12.987778 1046.889782 6.552026
10 6 77.696217 12.870640 1057.307720 6.630041
10 7 75.357326 13.270110 1063.073635 6.586321
10 8 77.163411 12.959510 1055.134773 6.653994
10 9 78.046994 12.812793 1045.463085 6.904095
32 2 True 300 300 10 0 78.084545 12.806632 1036.835432 6.313942
10 1 77.134630 12.964346 1049.352884 6.631158
10 2 77.850885 12.845069 1039.554834 6.222758
10 3 78.302340 12.771010 1035.851955 6.272547
10 4 77.800478 12.853391 1066.196203 6.254006
10 5 77.417553 12.916967 1058.874130 6.411519
10 6 77.410989 12.918063 1062.852144 6.332852
10 7 76.002140 13.157524 1041.868925 6.322779
10 8 77.588153 12.888566 1063.034534 6.214421
10 9 77.352635 12.927808 1079.278946 6.207768
tensorrt 1 2 True 300 300 10 0 44.879506 22.281885 7123.430490 9.938359
10 1 45.896571 21.788120 7108.290195 9.763122
10 2 45.750886 21.857500 7096.955061 9.778380
10 3 46.714454 21.406651 7063.511372 9.863853
10 4 44.961773 22.241116 7079.815626 9.698868
10 5 46.786364 21.373749 7096.222401 9.900928
10 6 45.375222 22.038460 7078.917027 9.767294
10 7 46.308035 21.594524 7104.745865 9.731770
10 8 45.647817 21.906853 7133.631945 9.848475
10 9 45.444050 22.005081 7150.356531 9.665608
2 2 True 300 300 10 0 58.572711 17.072797 7159.089088 8.055568
10 1 56.268416 17.771959 7057.126999 8.049965
10 2 56.662555 17.648339 7124.001980 8.127034
10 3 57.482204 17.396688 7202.754974 8.050621
10 4 57.387825 17.425299 7141.631842 8.012891
10 5 56.594128 17.669678 7013.514280 8.051515
10 6 59.028978 16.940832 7109.473228 8.028746
10 7 57.955590 17.254591 7135.635853 8.022904
10 8 57.148556 17.498255 7042.576551 8.270383
10 9 58.484913 17.098427 7034.452677 8.007348
4 2 True 300 300 10 0 64.725415 15.449882 7074.600697 8.012056
10 1 63.373144 15.779555 7085.893393 8.088231
10 2 66.429424 15.053570 7048.755884 8.031487
10 3 65.250275 15.325606 7034.267187 8.069247
10 4 65.214766 15.333951 7130.547285 7.950902
10 5 65.505550 15.265882 7089.531422 7.904530
10 6 64.029799 15.617728 7119.791985 7.780850
10 7 65.942214 15.164793 7195.689201 7.865220
10 8 65.534464 15.259147 7132.315159 7.791936
10 9 63.680558 15.703380 7093.014956 7.820636
8 2 True 300 300 10 0 71.298507 14.025539 7074.872017 6.767228
10 1 72.376888 13.816565 7089.612961 6.696120
10 2 73.121817 13.675809 7110.588074 6.759971
10 3 73.200939 13.661027 7105.098248 6.812528
10 4 72.154638 13.859123 7090.573788 6.809860
10 5 72.213958 13.847739 7094.809532 6.766707
10 6 74.139514 13.488084 7119.920015 6.968752
10 7 72.254699 13.839930 7242.331505 6.968305
10 8 74.052304 13.503969 7073.332071 6.843194
10 9 71.986527 13.891488 7078.927040 6.784365
16 2 True 300 300 10 0 76.237202 13.116956 7137.548923 6.830409
10 1 75.098016 13.315931 7022.910595 6.522901
10 2 75.640508 13.220429 7039.772511 6.748743
10 3 74.806781 13.367772 7034.528971 6.886221
10 4 74.952158 13.341844 7142.646551 6.512262
10 5 75.163708 13.304293 7138.361692 6.879665
10 6 76.975154 12.991205 7137.981892 6.521046
10 7 74.334390 13.452724 7127.883911 6.899901
10 8 76.398632 13.089240 7101.144314 6.532736
10 9 78.012699 12.818426 7104.007959 6.573454
32 2 True 300 300 10 0 75.877238 13.179183 7028.620958 6.213941
10 1 78.366163 12.760609 7215.569258 6.314900
10 2 77.502625 12.902789 7181.191206 6.204851
10 3 77.052797 12.978114 7204.288960 6.472256
10 4 74.507041 13.421550 7076.518536 6.345276
10 5 74.613696 13.402365 7155.494452 6.593291
10 6 77.437162 12.913696 7074.496746 6.642245
10 7 78.017823 12.817584 7032.968521 6.642971
10 8 75.998827 13.158098 7065.843344 6.629419
10 9 77.519635 12.899958 7109.196186 6.290577
tensorrt-dynamic 1 2 True 300 300 10 0 45.938293 21.768332 5793.458462 9.775043
10 1 45.675654 21.893501 5750.547647 9.920835
10 2 51.432930 19.442797 5830.008745 9.742856
10 3 49.596821 20.162582 5755.321026 9.704709
10 4 49.436064 20.228148 5926.057339 9.736776
10 5 45.546200 21.955729 5836.916924 9.829998
10 6 47.782545 20.928144 5818.340778 9.796858
10 7 49.294299 20.286322 5859.798670 9.711623
10 8 45.307092 22.071600 5864.037991 9.728909
10 9 45.384059 22.034168 5768.066883 9.842157
2 2 True 300 300 10 0 57.416500 17.416596 5758.180380 8.009374
10 1 54.676339 18.289447 5801.112652 8.045614
10 2 59.107588 16.918302 5800.835371 8.006454
10 3 59.636632 16.768217 5811.677456 8.012474
10 4 57.085943 17.517447 5873.568058 7.994771
10 5 58.615276 17.060399 5802.568436 8.155286
10 6 57.741367 17.318606 5855.651617 7.979155
10 7 60.370980 16.564250 5841.228485 7.967949
10 8 57.798656 17.301440 5840.348482 8.026183
10 9 54.218695 18.443823 5810.206175 7.998705
4 2 True 300 300 10 0 68.002691 14.705300 5865.043879 8.005500
10 1 69.428613 14.403284 5854.304552 7.781118
10 2 66.962350 14.933765 5865.014315 7.751703
10 3 69.325334 14.424741 5867.359638 8.062124
10 4 69.243788 14.441729 5873.419046 7.969081
10 5 69.559876 14.376104 5812.558889 8.033216
10 6 69.630895 14.361441 5865.339279 8.025318
10 7 67.982025 14.709771 5872.568369 7.962793
10 8 69.782656 14.330208 5845.269680 7.953584
10 9 67.747041 14.760792 5910.014391 8.042395
8 2 True 300 300 10 0 73.276231 13.646990 5830.368996 6.824210
10 1 71.899684 13.908267 5899.592161 6.959260
10 2 73.132814 13.673753 5873.061419 6.992325
10 3 73.893467 13.532996 5847.330570 6.635457
10 4 72.348174 13.822049 5839.567661 6.717265
10 5 73.382318 13.627261 5853.819370 6.948799
10 6 74.256497 13.466835 5942.756414 6.785452
10 7 74.786273 13.371438 5841.147900 6.892428
10 8 73.390504 13.625741 5794.231892 6.856382
10 9 76.317831 13.103098 5854.991436 6.855786
16 2 True 300 300 10 0 79.187677 12.628227 5828.539610 6.874055
10 1 76.099831 13.140634 5960.506916 6.546430
10 2 78.597822 12.722999 5811.300993 6.724693
10 3 77.659263 12.876764 5794.836998 6.541140
10 4 78.604635 12.721896 5913.775682 6.476842
10 5 76.692777 13.039038 5850.826740 6.695062
10 6 76.448153 13.080761 5859.509468 6.677456
10 7 76.065414 13.146579 5859.508038 6.517343
10 8 77.981605 12.823537 5865.531921 6.869994
10 9 76.892161 13.005227 5846.389294 6.884292
32 2 True 300 300 10 0 80.189638 12.470439 5762.305021 6.599370
10 1 78.977063 12.661904 5833.571672 6.236974
10 2 82.863083 12.068100 5827.735186 6.217435
10 3 80.388719 12.439556 5816.005707 6.597616
10 4 78.810997 12.688585 5843.477011 6.239973
10 5 79.709880 12.545496 5795.105457 6.575037
10 6 80.851082 12.368418 5842.574835 6.621655
10 7 81.112587 12.328543 5779.558420 6.600361
10 8 80.312524 12.451358 5818.472385 6.293070
10 9 79.353741 12.601800 5791.924715 6.193805
ssd-resnet50-fpn cpu 1 2 True 640 640 10 0 2.885392 346.573353 1215.217590 9.813547
10 1 2.902233 344.562292 1213.778973 9.861469
10 2 2.901994 344.590664 1163.634300 9.424686
10 3 2.908665 343.800306 1198.755503 9.889960
10 4 2.922375 342.187405 1183.545589 9.415150
10 5 2.865341 348.998547 1217.129230 9.387851
10 6 2.888207 346.235514 1232.977867 9.844065
10 7 2.838141 352.343321 1231.852293 9.802461
10 8 2.925300 341.845274 1222.834587 9.759426
10 9 2.910623 343.569040 1165.804625 9.410620
2 2 True 640 640 10 0 2.873998 347.947359 1159.969807 7.540405
10 1 2.908272 343.846798 1234.817982 7.984102
10 2 2.894643 345.465779 1150.921106 7.940948
10 3 2.877366 347.540021 1126.492739 7.705986
10 4 2.903768 344.380140 1224.944115 7.951438
10 5 2.895594 345.352292 1193.471193 7.978678
10 6 2.923425 342.064500 1203.737497 7.766485
10 7 2.900185 344.805598 1162.850142 7.783830
10 8 2.831976 353.110313 1156.932592 7.961750
10 9 2.923105 342.101932 1166.455507 7.627249
4 2 True 640 640 10 0 2.866955 348.802149 1184.247494 7.521212
10 1 2.856502 350.078583 1208.733320 7.707179
10 2 2.894666 345.463037 1201.236486 7.814735
10 3 2.854054 350.378752 1137.424946 7.486254
10 4 2.889178 346.119225 1205.458641 7.488489
10 5 2.849271 350.966930 1258.634806 7.556528
10 6 2.862804 349.307895 1127.643108 7.340252
10 7 2.872507 348.127961 1195.940971 7.396042
10 8 2.848825 351.021886 1144.015312 7.435232
10 9 2.856191 350.116611 1192.873001 7.433087
8 2 True 640 640 10 0 2.837015 352.483124 1195.983171 6.284982
10 1 2.832456 353.050530 1168.994665 6.403863
10 2 2.823737 354.140669 1218.814135 6.399363
10 3 2.810910 355.756640 1235.333920 6.492764
10 4 2.829624 353.403866 1225.069523 6.618798
10 5 2.794465 357.850224 1211.092949 6.317213
10 6 2.812321 355.578184 1205.430984 6.580472
10 7 2.820940 354.491740 1201.059580 6.551847
10 8 2.806979 356.254905 1186.581135 6.576344
10 9 2.821171 354.462683 1192.502022 6.348789
16 2 True 640 640 10 0 2.813172 355.470613 1164.383173 6.226808
10 1 2.818235 354.831964 1223.844290 6.216079
10 2 2.810229 355.842948 1199.043274 6.146997
10 3 2.802453 356.830269 1204.602718 6.347194
10 4 2.784725 359.101936 1191.951036 6.114006
10 5 2.806251 356.347278 1238.991499 6.278954
10 6 2.802616 356.809527 1217.591047 6.227031
10 7 2.801952 356.894091 1185.748577 6.277829
10 8 2.807754 356.156498 1172.047138 6.207116
10 9 2.806592 356.304035 1185.555696 6.203011
32 2 True 640 640 10 0 2.798054 357.391268 2438.719749 5.896248
10 1 2.807195 356.227458 1187.539339 5.959757
10 2 2.796749 357.558042 1222.377062 5.879760
10 3 2.799535 357.202135 1218.173742 5.875945
10 4 2.804346 356.589369 1235.699415 5.910259
10 5 2.789166 358.530104 1222.545624 5.984668
10 6 2.789829 358.444877 1208.798170 5.984273
10 7 2.796627 357.573621 1180.090189 5.946502
10 8 2.794275 357.874520 1225.750208 5.948119
10 9 2.797504 357.461505 1233.389139 6.108604
cpu-prebuilt 1 2 True 640 640 10 0 2.855453 350.207090 2278.793573 9.891272
10 1 2.954680 338.446140 1129.206181 9.652257
10 2 2.889051 346.134424 1142.500162 9.572744
10 3 2.922671 342.152834 1112.239361 9.393334
10 4 2.902564 344.522953 1125.231028 9.726524
10 5 2.850115 350.862980 1139.945984 9.559512
10 6 2.867265 348.764420 1113.990545 9.485006
10 7 2.847977 351.126432 1118.859053 9.380341
10 8 2.846104 351.357460 1108.821630 9.745359
10 9 2.880664 347.142220 1126.899481 9.636045
2 2 True 640 640 10 0 2.878353 347.420931 1111.602306 7.814884
10 1 2.882036 346.976876 1120.844841 7.746935
10 2 2.843334 351.699829 1121.196508 7.841945
10 3 2.875354 347.783208 1135.895967 8.069575
10 4 2.888179 346.238971 1108.161211 7.781208
10 5 2.850747 350.785255 1180.108547 7.913172
10 6 2.910487 343.585134 1111.355066 7.823527
10 7 2.870341 348.390698 1114.764690 7.639170
10 8 2.879936 347.229958 1117.313623 7.627785
10 9 2.857527 349.952936 1119.010687 7.763743
4 2 True 640 640 10 0 2.815198 355.214775 1127.735138 7.499129
10 1 2.859160 349.753082 1123.138428 7.436484
10 2 2.852198 350.606740 1103.125572 7.412642
10 3 2.856863 350.034297 1115.490675 7.509857
10 4 2.804357 356.588006 1113.413334 7.386327
10 5 2.821205 354.458451 1123.439074 7.423639
10 6 2.833225 352.954626 1111.227512 7.583082
10 7 2.857581 349.946320 1113.034725 7.508188
10 8 2.855387 350.215197 1134.685278 8.161485
10 9 2.814018 355.363786 1095.266342 7.439435
8 2 True 640 640 10 0 2.789838 358.443707 1118.631601 6.431282
10 1 2.820658 354.527265 1109.792948 6.465271
10 2 2.806657 356.295764 1105.580330 6.383821
10 3 2.811476 355.684996 1129.949808 6.376565
10 4 2.792511 358.100593 1101.967335 6.407395
10 5 2.801499 356.951714 1119.222879 6.572589
10 6 2.804273 356.598675 1107.131481 6.362662
10 7 2.794106 357.896268 1120.184898 6.467849
10 8 2.787627 358.728081 1100.586653 6.664187
10 9 2.806029 356.375545 1097.465277 6.390840
16 2 True 640 640 10 0 2.793312 357.997969 1117.201328 6.260753
10 1 2.787269 358.774066 1137.892246 6.353207
10 2 2.775893 360.244483 1107.661009 6.194562
10 3 2.795524 357.714742 1106.746435 6.221645
10 4 2.789867 358.440027 1113.170147 6.227575
10 5 2.785558 358.994484 1116.304398 6.189026
10 6 2.782097 359.441057 1177.857637 6.233983
10 7 2.777166 360.079348 1109.665394 6.439984
10 8 2.774922 360.370502 1112.155676 6.156623
10 9 2.779182 359.818116 1106.245518 6.240182
32 2 True 640 640 10 0 2.793304 357.999019 1118.134022 5.916126
10 1 2.784916 359.077275 1106.137037 5.865730
10 2 2.783606 359.246209 1111.650705 5.980566
10 3 2.787308 358.769089 1121.110439 5.964067
10 4 2.775340 360.316262 1107.404470 5.879041
10 5 2.783919 359.205842 1092.256308 5.947400
10 6 2.781490 359.519571 1114.127398 6.119121
10 7 2.769306 361.101329 1105.468512 5.958006
10 8 2.776802 360.126570 1110.611677 5.901009
10 9 2.777269 360.065982 1107.602119 5.951378
cuda 1 2 True 640 640 10 0 15.044780 66.468239 1098.026752 9.421945
10 1 15.477536 64.609766 1129.821777 9.449124
10 2 14.993473 66.695690 1105.736256 9.390235
10 3 14.742047 67.833185 1108.495235 9.438276
10 4 15.425056 64.829588 1129.258633 9.686470
10 5 15.192460 65.822124 1117.984056 9.646535
10 6 14.797803 67.577600 1102.870703 9.317398
10 7 15.038253 66.497087 1133.430004 9.444475
10 8 15.015265 66.598892 1104.692459 9.422898
10 9 15.172181 65.910101 1112.380028 9.325027
2 2 True 640 640 10 0 17.788794 56.215167 1117.373943 7.849932
10 1 17.869584 55.961013 1113.932133 7.669449
10 2 17.461570 57.268620 1129.554510 7.846773
10 3 17.505369 57.125330 1125.203371 7.897556
10 4 17.872325 55.952430 1120.651484 7.834733
10 5 17.840814 56.051254 1126.897812 7.941365
10 6 17.768145 56.280494 1131.600618 7.722080
10 7 17.404943 57.454944 1110.110521 7.553518
10 8 17.549720 56.980968 1103.447437 7.722795
10 9 17.620128 56.753278 1129.095316 7.849455
4 2 True 640 640 10 0 19.009631 52.604914 1105.010509 7.283628
10 1 19.646119 50.900638 1117.938995 7.447034
10 2 19.069981 52.438438 1107.816696 7.761061
10 3 19.608669 50.997853 1134.235382 7.297724
10 4 19.501497 51.278114 1118.249416 7.535100
10 5 19.580863 51.070273 1114.309549 7.481039
10 6 19.669613 50.839841 1120.993614 7.478744
10 7 19.541293 51.173687 1113.941669 7.593006
10 8 19.117287 52.308679 1117.832422 7.227778
10 9 19.412457 51.513314 1126.186371 7.310241
8 2 True 640 640 10 0 19.876099 50.311685 1117.755175 6.379679
10 1 19.936954 50.158113 1116.433382 6.557599
10 2 20.212125 49.475253 1117.033720 6.511047
10 3 20.008964 49.977601 1131.582737 6.359369
10 4 19.981511 50.046265 1133.501291 6.507501
10 5 19.702229 50.755680 1121.912003 6.450221
10 6 19.764490 50.595790 1111.566782 6.495044
10 7 19.619205 50.970465 1114.754200 6.510496
10 8 19.858724 50.355703 1116.500139 6.394997
10 9 19.596427 51.029712 1126.966000 6.479204
16 2 True 640 640 10 0 21.172854 47.230288 1117.800713 6.400287
10 1 21.047138 47.512397 1107.041359 6.163068
10 2 20.762853 48.162937 1124.565840 6.408639
10 3 20.749295 48.194408 1126.641273 6.126396
10 4 21.492133 46.528652 1130.404949 6.149955
10 5 21.321513 46.900988 1110.189438 6.229348
10 6 21.329177 46.884134 1120.916843 6.235205
10 7 21.451538 46.616703 1117.917299 6.182045
10 8 21.151159 47.278732 1129.104376 6.296031
10 9 20.991265 47.638863 1125.458479 6.268002
32 2 True 640 640 10 0 21.131578 47.322541 1135.979176 5.965155
10 1 21.439374 46.643153 1124.232292 6.127946
10 2 21.342394 46.855099 1130.904198 5.932517
10 3 21.092555 47.410093 1113.237143 5.863123
10 4 20.837156 47.991194 1138.520241 5.897202
10 5 21.044528 47.518291 1127.521992 5.988065
10 6 21.254780 47.048241 1127.163410 5.989894
10 7 21.061998 47.478877 1136.384249 5.897298
10 8 21.253932 47.050118 1131.122589 6.049484
10 9 21.122978 47.341809 1120.296955 5.881034
tensorrt 1 2 True 640 640 10 0 15.570907 64.222336 21729.543209 9.394884
10 1 16.074873 62.208891 21746.314526 9.488940
10 2 16.338304 61.205864 21647.392273 9.246588
10 3 16.071054 62.223673 21447.484493 9.172678
10 4 16.122947 62.023401 21772.876501 9.240508
10 5 16.205862 61.706066 21609.282255 9.258628
10 6 16.713970 59.830189 21890.000582 9.244323
10 7 15.914824 62.834501 21275.634527 9.418249
10 8 16.356207 61.138868 21520.533800 9.368420
10 9 16.302551 61.340094 21316.932917 9.224415
2 2 True 640 640 10 0 18.955122 52.756190 23919.415712 7.594585
10 1 18.540696 53.935409 23624.637127 7.618606
10 2 18.921430 52.850127 23499.101639 7.565439
10 3 18.671433 53.557754 23391.526222 7.526278
10 4 18.196231 54.956436 23598.426342 7.588744
10 5 18.777005 53.256631 23855.793238 7.571578
10 6 18.793537 53.209782 23437.061310 7.582605
10 7 18.557061 53.887844 23561.732769 7.421434
10 8 18.949341 52.772284 23810.215235 7.553399
10 9 18.621654 53.700924 23574.730396 7.589579
4 2 True 640 640 10 0 20.415578 48.982203 13964.055777 7.495016
10 1 20.558223 48.642337 13821.491480 7.330626
10 2 20.337696 49.169779 13901.971102 7.315248
10 3 20.208668 49.483716 13633.214474 7.559001
10 4 20.480075 48.827946 13880.959511 7.485181
10 5 20.495311 48.791647 13754.168272 7.535517
10 6 20.423531 48.963130 13678.659678 7.557243
10 7 20.295060 49.273074 13856.650591 7.384658
10 8 20.356598 49.124122 13791.505575 7.390469
10 9 20.404430 49.008965 13819.857597 7.358462
8 2 True 640 640 10 0 21.386117 46.759307 15457.005501 6.325573
10 1 20.925588 47.788382 15310.769081 6.405577
10 2 21.052784 47.499657 15355.959654 6.333739
10 3 20.773098 48.139185 15436.644316 6.615311
10 4 20.757895 48.174441 15398.816586 6.397918
10 5 20.607792 48.525333 15496.545076 6.420434
10 6 20.722935 48.255712 15283.169031 6.394431
10 7 20.734371 48.229098 15375.441313 6.582066
10 8 21.183180 47.207266 15395.080090 6.349027
10 9 21.018878 47.576278 15299.387217 6.549969
16 2 True 640 640 10 0 21.361222 46.813801 18176.128626 6.257810
10 1 21.380230 46.772182 18112.614870 6.449446
10 2 21.488789 46.535894 18246.995211 6.438613
10 3 21.359951 46.816587 18380.831480 6.129384
10 4 20.977000 47.671258 18351.099253 6.451964
10 5 21.244209 47.071651 18284.997940 6.177515
10 6 21.471016 46.574414 18289.912701 6.147586
10 7 21.308548 46.929523 18348.347902 6.401733
10 8 20.823155 48.023462 18243.337631 6.366849
10 9 21.180365 47.213539 18333.791256 6.314740
32 2 True 640 640 10 0 21.191782 47.188103 24676.125288 6.090987
10 1 21.083311 47.430880 24752.432585 5.904481
10 2 21.125678 47.335759 24673.513651 5.980283
10 3 21.127640 47.331363 24676.829815 5.923975
10 4 21.269877 47.014847 24702.081442 5.911347
10 5 21.000507 47.617897 24648.655176 5.978446
10 6 21.364272 46.807118 24671.283007 5.951062
10 7 21.282035 46.987988 24687.311172 5.950600
10 8 20.938467 47.758989 24765.106678 6.044965
10 9 21.084040 47.429241 24839.820147 5.940046
tensorrt-dynamic 1 2 True 640 640 10 0 19.510659 51.254034 8519.782305 9.186983
10 1 19.327788 51.738977 8500.771523 9.692311
10 2 19.465386 51.373243 8400.408030 9.241462
10 3 20.335329 49.175501 8444.699049 9.147525
10 4 19.936516 50.159216 8430.749416 9.275198
10 5 18.953537 52.760601 8339.345455 9.375453
10 6 19.509933 51.255941 8486.658812 9.344339
10 7 19.839855 50.403595 8455.882311 9.330869
10 8 19.872755 50.320148 8421.129942 9.277582
10 9 20.354079 49.130201 8518.504620 9.328961
2 2 True 640 640 10 0 22.375349 44.692039 8460.820913 7.580161
10 1 22.628919 44.191241 8456.921101 7.472217
10 2 21.955511 45.546651 8441.268921 7.547975
10 3 22.382514 44.677734 8476.423502 7.572532
10 4 22.365625 44.711471 8486.182928 7.580638
10 5 21.967873 45.521021 8448.971510 7.503867
10 6 22.394703 44.653416 8465.628386 7.618845
10 7 22.409241 44.624448 8421.254396 7.518291
10 8 21.862299 45.740843 8426.948309 7.569313
10 9 22.333352 44.776082 8466.913939 7.509351
4 2 True 640 640 10 0 23.920294 41.805506 8466.913700 7.536948
10 1 23.513210 42.529285 8473.692417 7.440627
10 2 23.655570 42.273343 8480.237722 7.608294
10 3 23.764469 42.079628 8435.468674 7.287860
10 4 23.972588 41.714311 8399.927855 7.281214
10 5 24.284995 41.177690 8280.506611 7.427931
10 6 23.890456 41.857719 8474.812031 7.368922
10 7 23.802365 42.012632 8443.987608 7.532328
10 8 23.798989 42.018592 8468.756676 7.303774
10 9 24.087061 41.516066 8756.255865 7.281870
8 2 True 640 640 10 0 24.149400 41.408896 8371.308565 6.414413
10 1 24.079766 41.528642 8380.914211 6.348655
10 2 23.011062 43.457359 8537.108660 6.313503
10 3 23.971218 41.716695 8319.250345 6.473899
10 4 23.989573 41.684777 8341.042280 6.289378
10 5 24.223966 41.281432 8475.624800 6.320685
10 6 24.313308 41.129738 8340.878010 6.365463
10 7 24.019264 41.633248 8514.063358 6.458774
10 8 24.014709 41.641146 8560.509443 6.431967
10 9 24.146463 41.413933 8319.505215 6.476253
16 2 True 640 640 10 0 24.324447 41.110903 8523.086071 6.373674
10 1 23.775641 42.059854 8404.950380 6.176323
10 2 24.280584 41.185170 8522.124052 6.459951
10 3 24.073271 41.539848 8431.672335 6.225787
10 4 24.300199 41.151926 8449.722290 6.536342
10 5 24.594776 40.659040 8488.881826 6.295867
10 6 24.248246 41.240096 8375.000477 6.217360
10 7 24.075231 41.536465 8429.190874 6.083399
10 8 23.958176 41.739404 8284.771442 6.156728
10 9 23.548903 42.464823 8354.423761 6.275095
32 2 True 640 640 10 0 24.111050 41.474760 8494.910240 5.925834
10 1 23.958574 41.738711 8489.223480 5.846027
10 2 23.979064 41.703045 8476.033926 6.055284
10 3 24.214622 41.297361 8539.590120 5.996846
10 4 23.922038 41.802458 8418.959856 6.071333
10 5 24.738744 40.422425 8410.470009 5.948696
10 6 23.969677 41.719377 8425.133944 6.118607
10 7 23.967135 41.723803 8452.291250 6.080065
10 8 24.044863 41.588925 8353.692293 5.913790
10 9 24.059470 41.563675 8362.168312 5.903147
ssdlite-mobilenet-v2 cpu 1 2 True 300 300 10 0 39.685341 25.198221 862.735510 10.295510
10 1 38.059108 26.274920 875.326633 10.284185
10 2 38.016678 26.304245 850.307703 9.809732
10 3 37.161473 26.909590 848.699808 9.995341
10 4 36.292639 27.553797 815.796137 9.835124
10 5 36.992680 27.032375 855.764389 10.437012
10 6 36.585465 27.333260 823.440313 9.884477
10 7 36.976048 27.044535 863.843441 10.589600
10 8 37.688058 26.533604 842.252016 10.356188
10 9 36.079414 27.716637 836.422205 10.013103
2 2 True 300 300 10 0 50.425035 19.831419 848.634005 8.233964
10 1 50.965145 19.621253 856.149912 8.628905
10 2 52.640348 18.996835 875.950813 8.672178
10 3 51.905527 19.265771 846.791983 8.576155
10 4 49.655538 20.138741 880.769253 8.548200
10 5 51.085569 19.575000 873.834133 8.720458
10 6 53.114307 18.827319 850.406885 8.596361
10 7 52.699544 18.975496 877.855301 8.659899
10 8 50.205031 19.918323 814.801931 8.282781
10 9 51.838489 19.290686 855.423450 8.524060
4 2 True 300 300 10 0 58.231374 17.172873 860.884905 8.060366
10 1 54.661197 18.294513 867.341518 8.239925
10 2 56.616855 17.662585 859.322309 8.227885
10 3 55.873581 17.897546 854.336262 8.497715
10 4 55.702728 17.952442 860.478878 8.361101
10 5 57.902385 17.270446 838.958502 8.259237
10 6 55.431452 18.040299 866.113901 8.402050
10 7 55.038484 18.169105 873.219967 8.160889
10 8 59.594901 16.779959 866.595745 8.166939
10 9 56.829537 17.596483 864.920139 8.304119
8 2 True 300 300 10 0 60.394994 16.557664 847.701073 6.887943
10 1 65.184488 15.341073 856.268167 7.110670
10 2 64.526389 15.497535 883.671045 6.978080
10 3 63.105098 15.846580 853.388071 7.361040
10 4 61.999715 16.129106 854.346037 6.790102
10 5 60.889048 16.423315 873.489618 6.929606
10 6 63.257614 15.808374 860.797882 6.799281
10 7 63.593540 15.724868 873.976231 6.884992
10 8 63.919900 15.644580 800.388098 6.886363
10 9 65.018141 15.380323 857.316971 6.874159
16 2 True 300 300 10 0 60.372718 16.563773 876.538992 6.611742
10 1 62.656143 15.960127 845.313787 7.245757
10 2 60.175827 16.617969 880.301714 6.795049
10 3 62.156023 16.088545 841.965914 6.593056
10 4 61.231944 16.331345 851.051569 6.575733
10 5 60.472601 16.536415 874.032497 6.778210
10 6 62.592794 15.976280 877.879620 6.638654
10 7 61.218594 16.334906 871.866226 6.617002
10 8 61.320850 16.307667 857.220888 6.511718
10 9 61.836840 16.171589 838.464499 6.742433
32 2 True 300 300 10 0 60.648523 16.488448 1003.379822 6.257281
10 1 62.038518 16.119018 858.625889 6.314501
10 2 61.297690 16.313829 870.803595 6.263062
10 3 61.261961 16.323343 878.454447 6.259497
10 4 58.678264 17.042086 857.612133 6.264176
10 5 60.631180 16.493164 824.439287 6.288312
10 6 60.984496 16.397610 827.073097 6.329279
10 7 60.128515 16.631044 834.861994 6.248549
10 8 59.084872 16.924806 796.676874 6.223783
10 9 60.305743 16.582169 817.299366 6.313596
cpu-prebuilt 1 2 True 300 300 10 0 37.004102 27.024031 949.477911 9.892106
10 1 37.368067 26.760817 782.538652 10.296226
10 2 37.597519 26.597500 761.238337 10.153770
10 3 37.601901 26.594400 777.010679 10.021806
10 4 38.062907 26.272297 765.479803 9.991169
10 5 38.547045 25.942326 772.315025 10.000944
10 6 36.258755 27.579546 783.654451 10.087013
10 7 37.603586 26.593208 770.739794 9.885430
10 8 38.028052 26.296377 782.424688 9.866834
10 9 36.295151 27.551889 792.186737 10.169029
2 2 True 300 300 10 0 48.991724 20.411611 785.945415 8.294463
10 1 49.559901 20.177603 859.849215 8.650362
10 2 48.770410 20.504236 766.938686 8.291662
10 3 49.186777 20.330667 786.423683 8.145630
10 4 48.470012 20.631313 862.405300 8.572221
10 5 49.019494 20.400047 780.265808 8.450449
10 6 50.361462 19.856453 768.947363 8.481145
10 7 49.089775 20.370841 774.491072 8.264840
10 8 47.876630 20.887017 786.745787 8.256376
10 9 50.821876 19.676566 801.594973 8.289516
4 2 True 300 300 10 0 55.883445 17.894387 794.907331 7.973313
10 1 52.787923 18.943727 785.968304 8.051932
10 2 57.212673 17.478645 784.177780 8.001566
10 3 53.389477 18.730283 799.047709 8.024305
10 4 55.693113 17.955542 782.022476 8.064538
10 5 53.976752 18.526495 795.651436 8.258611
10 6 57.645739 17.347336 777.827501 7.873803
10 7 52.706497 18.972993 782.297850 8.054614
10 8 53.816768 18.581569 781.893015 7.929951
10 9 53.702729 18.621027 791.301727 7.836729
8 2 True 300 300 10 0 58.745708 17.022520 770.885468 6.994680
10 1 58.582426 17.069966 782.234907 6.892636
10 2 60.420768 16.550601 804.603100 7.039666
10 3 63.338453 15.788198 776.143074 6.763458
10 4 61.200115 16.339839 773.752689 6.763175
10 5 60.864749 16.429871 779.057026 6.774992
10 6 62.106324 16.101420 789.723873 6.877095
10 7 61.079354 16.372144 768.090725 6.792203
10 8 63.146901 15.836090 771.646738 7.009849
10 9 61.256425 16.324818 787.130833 6.843984
16 2 True 300 300 10 0 58.729256 17.027289 786.702633 6.566666
10 1 58.600279 17.064765 775.289297 6.510623
10 2 57.869381 17.280295 801.863432 6.936930
10 3 58.835426 16.996562 784.094810 6.589137
10 4 57.795620 17.302349 775.352240 6.557748
10 5 57.727956 17.322630 767.080784 6.591357
10 6 59.615441 16.774178 829.609156 6.648198
10 7 57.891795 17.273605 776.070118 6.593898
10 8 59.204161 16.890705 766.680241 6.543301
10 9 57.923675 17.264098 837.984800 6.583177
32 2 True 300 300 10 0 57.782930 17.306149 784.720898 6.350499
10 1 57.193657 17.484456 784.260035 6.251678
10 2 58.120374 17.205670 778.075695 6.204356
10 3 57.936727 17.260209 780.213118 6.225489
10 4 57.048188 17.529041 794.155359 6.290961
10 5 58.113252 17.207779 773.253918 6.246470
10 6 58.781315 17.012209 774.410248 6.380927
10 7 58.123168 17.204843 770.066977 6.393492
10 8 57.623787 17.353944 794.302940 6.217804
10 9 58.012679 17.237611 833.894014 6.635834
cuda 1 2 True 300 300 10 0 32.349978 30.911922 789.393902 10.004997
10 1 32.277533 30.981302 777.021646 9.960294
10 2 33.160485 30.156374 778.235674 9.951949
10 3 32.484773 30.783653 773.295403 9.849787
10 4 32.289460 30.969858 782.374382 9.889722
10 5 33.100035 30.211449 780.789375 9.934664
10 6 33.402650 29.937744 792.579651 9.874344
10 7 33.941364 29.462576 780.071735 10.011554
10 8 33.921874 29.479504 784.799099 9.832978
10 9 32.897534 30.397415 792.848825 9.998083
2 2 True 300 300 10 0 44.786537 22.328138 788.602829 8.304298
10 1 40.870994 24.467230 782.656431 8.443475
10 2 41.782387 23.933530 781.589985 8.288801
10 3 43.433234 23.023844 793.455124 8.127034
10 4 44.176609 22.636414 791.908979 8.200824
10 5 42.058491 23.776412 789.214134 7.984221
10 6 43.240916 23.126245 797.036409 8.341908
10 7 42.246582 23.670554 784.744978 8.341849
10 8 44.305638 22.570491 792.612791 8.268595
10 9 43.159046 23.170114 790.400267 8.097827
4 2 True 300 300 10 0 50.584978 19.768715 797.727823 8.020848
10 1 48.003205 20.831943 775.827408 7.975131
10 2 46.668974 21.427512 791.930914 7.946432
10 3 48.008424 20.829678 791.920185 7.830560
10 4 47.840998 20.902574 776.671886 8.027762
10 5 47.704558 20.962358 784.771681 8.086592
10 6 48.114714 20.783663 792.357206 7.941931
10 7 47.777646 20.930290 785.501957 7.936209
10 8 47.000006 21.276593 773.928642 8.145422
10 9 47.861606 20.893574 789.598465 8.004040
8 2 True 300 300 10 0 49.444806 20.224571 791.300058 6.731212
10 1 49.426816 20.231932 788.105965 6.797835
10 2 51.412283 19.450605 798.650742 7.010430
10 3 49.625942 20.150751 798.736334 6.831899
10 4 49.323283 20.274401 786.125898 6.808758
10 5 51.141161 19.553721 793.793678 6.724164
10 6 50.413292 19.836038 795.562744 6.924197
10 7 50.309436 19.876987 791.213989 6.806701
10 8 51.698558 19.342899 789.509296 6.778181
10 9 52.680183 18.982470 798.557997 6.741926
16 2 True 300 300 10 0 50.071265 19.971535 775.088072 6.526105
10 1 51.439750 19.440219 788.074493 6.713055
10 2 50.958334 19.623876 798.418999 6.875217
10 3 50.748355 19.705072 796.075106 6.591968
10 4 51.126668 19.559264 794.337988 6.549180
10 5 53.981311 18.524930 781.834364 6.682500
10 6 51.376701 19.464076 796.562195 6.605364
10 7 52.089203 19.197837 796.659946 6.516390
10 8 51.525138 19.408002 801.054716 6.619275
10 9 50.990199 19.611612 793.004751 6.594062
32 2 True 300 300 10 0 52.525603 19.038334 794.289351 6.252967
10 1 53.810231 18.583827 794.852257 6.505683
10 2 53.541330 18.677160 785.206079 6.416973
10 3 52.971238 18.878169 779.656410 6.322965
10 4 53.455948 18.706992 792.399883 6.244738
10 5 51.737299 19.328415 787.132978 6.224751
10 6 53.231684 18.785805 793.002367 6.631620
10 7 51.691490 19.345544 779.975653 6.219286
10 8 52.593215 19.013859 780.918360 6.271467
10 9 53.496430 18.692836 782.269716 6.598189
tensorrt 1 2 True 300 300 10 0 35.050382 28.530359 35672.811985 9.877920
10 1 37.707710 26.519775 35734.363079 9.876132
10 2 35.758896 27.965069 35356.230259 9.820700
10 3 35.530797 28.144598 35558.438778 9.880185
10 4 36.383938 27.484655 35570.314646 9.799600
10 5 35.633429 28.063536 35888.130188 9.905934
10 6 35.771705 27.955055 35514.756680 9.971023
10 7 35.799795 27.933121 35461.050272 9.771466
10 8 37.590780 26.602268 35439.099073 9.826541
10 9 37.125620 26.935577 35536.747932 9.871244
2 2 True 300 300 10 0 44.218991 22.614717 36102.570772 8.064449
10 1 43.894594 22.781849 35689.090729 8.099318
10 2 46.790018 21.372080 35965.830326 8.110583
10 3 46.000011 21.739125 36059.750080 8.101583
10 4 45.685605 21.888733 35974.194527 8.059144
10 5 45.064670 22.190332 35960.492134 8.215189
10 6 46.366907 21.567106 36009.477854 8.117616
10 7 42.498711 23.530126 35665.980339 8.049250
10 8 45.227458 22.110462 36006.569862 8.111775
10 9 45.155152 22.145867 36033.411980 8.205235
4 2 True 300 300 10 0 51.897339 19.268811 20436.912537 7.952839
10 1 47.530217 21.039248 20523.124695 8.060575
10 2 50.317358 19.873857 20642.505884 7.989317
10 3 49.739303 20.104825 20552.595377 8.039922
10 4 47.691540 20.968080 20634.422541 7.922381
10 5 50.226524 19.909799 20610.222816 8.024842
10 6 49.207839 20.321965 20518.425941 8.024156
10 7 48.164994 20.761967 20581.721306 7.939219
10 8 50.559520 19.778669 20608.603477 7.805288
10 9 49.909909 20.036101 20471.409798 8.016050
8 2 True 300 300 10 0 51.884579 19.273549 21162.240028 6.769925
10 1 52.411521 19.079775 21199.075937 7.179871
10 2 51.620535 19.372135 21240.265608 6.921858
10 3 51.347533 19.475132 21119.003057 6.866679
10 4 50.172901 19.931078 21245.890617 7.002726
10 5 53.531379 18.680632 21303.807735 6.739497
10 6 52.406200 19.081712 21211.510420 6.978109
10 7 50.439589 19.825697 21201.290131 6.951377
10 8 51.978053 19.238889 21105.926037 6.746009
10 9 51.141551 19.553572 21163.401842 6.907374
16 2 True 300 300 10 0 54.150795 18.466949 22976.331234 6.852485
10 1 54.501297 18.348187 23107.044697 6.870307
10 2 52.729979 18.964544 23003.034115 6.869771
10 3 51.683627 19.348487 23097.741604 6.533362
10 4 52.691641 18.978342 23121.733904 6.860301
10 5 52.760739 18.953487 23012.316942 6.557144
10 6 53.070833 18.842742 23209.119081 6.877601
10 7 52.512492 19.043088 23119.111538 6.905422
10 8 52.026126 19.221112 22915.400028 6.558336
10 9 51.058478 19.585386 22974.707842 6.754749
32 2 True 300 300 10 0 53.463848 18.704228 38056.073189 6.247517
10 1 55.518123 18.012136 38525.581360 6.233811
10 2 53.360695 18.740386 37970.025539 6.246682
10 3 53.299815 18.761791 39445.714474 6.284975
10 4 55.700301 17.953224 38235.680580 6.226789
10 5 56.074075 17.833553 38272.370815 6.632596
10 6 55.090850 18.151835 38257.955074 6.576888
10 7 53.035958 18.855132 38142.452240 6.234635
10 8 52.651272 18.992893 38927.487135 6.274536
10 9 54.079601 18.491261 38045.373678 6.560322
tensorrt-dynamic 1 2 True 300 300 10 0 36.257188 27.580738 4492.929935 9.842515
10 1 38.533234 25.951624 4472.224474 9.761214
10 2 35.761031 27.963400 4458.845854 9.777784
10 3 35.651299 28.049469 4430.665731 9.676337
10 4 37.579665 26.610136 4474.393845 9.764075
10 5 36.388357 27.481318 4479.102850 9.887338
10 6 36.222431 27.607203 4483.099461 9.891748
10 7 37.665721 26.549339 4507.941723 9.858489
10 8 36.447488 27.436733 4423.889160 9.681463
10 9 35.044818 28.534889 4492.774725 9.820461
2 2 True 300 300 10 0 43.085229 23.209810 4354.706287 8.053243
10 1 43.896201 22.781014 4459.623098 8.087158
10 2 46.957910 21.295667 4429.211617 8.065522
10 3 44.158935 22.645473 4382.166386 8.145511
10 4 45.924461 21.774888 4426.285744 8.045971
10 5 48.671370 20.545959 4482.997894 8.146882
10 6 43.603233 22.934079 4484.375715 8.114696
10 7 45.575895 21.941423 4462.842941 8.069873
10 8 47.062235 21.248460 4417.350054 8.044362
10 9 45.151506 22.147655 4489.324808 8.040249
4 2 True 300 300 10 0 50.611835 19.758224 4439.113617 7.771552
10 1 49.317193 20.276904 4527.236700 7.988602
10 2 49.987385 20.005047 4455.612898 7.809043
10 3 50.471302 19.813240 4472.424984 8.042753
10 4 52.434863 19.071281 4436.103582 7.985413
10 5 49.616329 20.154655 4438.630819 7.978261
10 6 49.702170 20.119846 4454.273701 7.984698
10 7 50.655845 19.741058 4485.294819 8.020997
10 8 49.991555 20.003378 4437.453270 7.853121
10 9 49.685684 20.126522 4419.140100 7.975161
8 2 True 300 300 10 0 51.926814 19.257873 4410.434723 6.691709
10 1 51.487227 19.422293 4386.659861 6.940439
10 2 53.588660 18.660665 4428.605318 6.895855
10 3 52.078895 19.201636 4450.277805 6.945342
10 4 55.108358 18.146068 4473.469496 6.976351
10 5 51.361366 19.469887 4481.755972 6.971747
10 6 52.652573 18.992424 4534.485817 6.936476
10 7 49.769403 20.092666 4488.675356 6.753743
10 8 51.072817 19.579887 4422.151327 6.929055
10 9 51.597595 19.380748 4452.725649 6.948888
16 2 True 300 300 10 0 55.482276 18.023774 4458.913326 7.088885
10 1 52.765676 18.951714 4457.708359 6.583475
10 2 52.935200 18.891022 4399.269819 6.477609
10 3 52.940294 18.889204 4466.050148 6.877258
10 4 52.952868 18.884718 4449.568748 6.659865
10 5 53.795974 18.588752 4397.683859 6.574184
10 6 54.138782 18.471047 4499.639511 6.823711
10 7 53.431773 18.715456 4428.323269 6.462626
10 8 53.892400 18.555492 4493.390560 6.541923
10 9 51.793480 19.307449 4420.616150 6.868303
32 2 True 300 300 10 0 53.517334 18.685535 4455.427647 6.305601
10 1 54.348094 18.399909 4469.662666 6.552197
10 2 54.623958 18.306985 4455.507517 6.439712
10 3 53.370943 18.736787 4434.459925 6.262217
10 4 54.644907 18.299967 4389.331341 6.248925
10 5 53.379540 18.733770 4404.261351 6.401818
10 6 55.426256 18.041991 4452.430964 6.259941
10 7 53.753897 18.603303 4411.975145 6.275486
10 8 53.469108 18.702388 4428.040504 6.598327
10 9 54.834806 18.236592 4407.277822 6.508533
yolo-v3 cpu 1 2 True 416 416 10 0 7.459803 134.051800 826.650381 12.364388
10 1 7.406113 135.023594 761.429787 11.766911
10 2 7.377169 135.553360 767.985821 11.433959
10 3 7.297196 137.038946 786.037683 12.368202
10 4 7.427570 134.633541 792.571783 12.083530
10 5 7.437487 134.454012 789.049864 11.571765
10 6 7.439809 134.412050 768.500090 11.892796
10 7 7.557768 132.314205 769.306660 11.978507
10 8 7.334212 136.347294 762.162685 11.768222
10 9 7.507874 133.193493 727.417946 11.743665
2 2 True 416 416 10 0 7.557522 132.318497 828.713655 10.014474
10 1 7.692893 129.990101 750.771284 10.156333
10 2 7.741839 129.168272 777.485132 10.104835
10 3 7.821765 127.848387 794.567108 10.232568
10 4 7.825603 127.785683 788.365364 10.231614
10 5 7.866818 127.116203 768.076181 10.247052
10 6 7.857776 127.262473 765.919685 10.197282
10 7 7.776648 128.590107 774.108171 10.263085
10 8 7.818790 127.897024 771.886349 9.763479
10 9 7.880068 126.902461 770.621300 10.260940
4 2 True 416 416 10 0 8.070625 123.906136 801.921129 9.931475
10 1 8.023575 124.632716 783.029556 9.824723
10 2 8.057489 124.108136 758.070469 9.692401
10 3 8.038337 124.403834 783.883572 9.921670
10 4 7.983725 125.254810 768.750191 9.572893
10 5 7.725530 129.440963 794.973135 10.110289
10 6 8.027645 124.569535 766.256332 9.763539
10 7 8.059495 124.077260 785.444498 9.889036
10 8 7.952784 125.742137 751.116037 9.540707
10 9 7.957868 125.661790 754.434824 9.631872
8 2 True 416 416 10 0 7.987237 125.199735 812.329292 8.724585
10 1 8.096920 123.503745 793.829918 8.596435
10 2 7.979053 125.328153 777.197599 8.909374
10 3 8.075955 123.824358 754.701853 8.658290
10 4 8.084819 123.688608 754.005432 8.825153
10 5 7.982700 125.270903 772.009850 8.698374
10 6 8.006997 124.890774 765.104055 8.629158
10 7 8.002761 124.956876 766.089916 8.645639
10 8 8.030900 124.519050 783.808708 8.621648
10 9 8.100998 123.441577 770.650148 8.544594
16 2 True 416 416 10 0 7.783413 128.478348 826.050043 8.524381
10 1 7.861007 127.210170 760.484934 8.538470
10 2 7.858170 127.256095 775.201321 8.575514
10 3 7.817914 127.911359 790.518761 8.449160
10 4 7.801944 128.173187 777.793169 8.474402
10 5 7.843042 127.501547 768.007994 8.512363
10 6 7.851994 127.356187 759.419203 8.807898
10 7 7.819607 127.883658 777.924061 8.742295
10 8 7.841469 127.527133 769.105196 8.449383
10 9 7.803614 128.145754 758.054733 8.371010
32 2 True 416 416 10 0 7.779773 128.538452 3009.296656 8.442033
10 1 7.789319 128.380932 754.206419 8.477602
10 2 7.796772 128.258213 794.567108 8.463278
10 3 7.788845 128.388740 754.859209 8.251451
10 4 7.737622 129.238665 774.708271 8.363865
10 5 7.766231 128.762580 778.465033 8.292355
10 6 7.751496 129.007362 795.592308 8.367561
10 7 7.779671 128.540143 757.143021 8.209728
10 8 7.780789 128.521666 770.535707 8.316364
10 9 7.778688 128.556386 769.662380 8.382883
cpu-prebuilt 1 2 True 416 416 10 0 7.505080 133.243084 770.616055 11.433601
10 1 7.578552 131.951332 776.219130 11.618733
10 2 7.637742 130.928755 741.240501 11.794209
10 3 7.466376 133.933783 742.387056 11.749029
10 4 7.641596 130.862713 765.636683 11.710882
10 5 7.463095 133.992672 742.483616 12.343168
10 6 7.476318 133.755684 744.406462 12.012601
10 7 7.632752 131.014347 746.438503 12.103677
10 8 7.545843 132.523298 746.113300 12.069225
10 9 7.554990 132.362843 759.110928 11.991858
2 2 True 416 416 10 0 7.788873 128.388286 780.621052 10.141850
10 1 7.864502 127.153635 745.664358 10.001600
10 2 7.868057 127.096176 758.759022 9.752929
10 3 7.704983 129.786134 742.192984 9.903431
10 4 7.846619 127.443433 745.921612 10.131240
10 5 7.761947 128.833652 774.803638 10.340035
10 6 7.772966 128.651023 727.174759 10.143161
10 7 7.880601 126.893878 763.806820 10.171413
10 8 7.838436 127.576470 743.127346 9.914815
10 9 7.817129 127.924204 741.799593 9.956062
4 2 True 416 416 10 0 8.047622 124.260306 733.114004 9.597838
10 1 8.085912 123.671889 787.828684 9.945691
10 2 7.898123 126.612365 729.163647 9.665459
10 3 7.930611 126.093686 758.428097 9.828657
10 4 8.129674 123.006165 744.267941 9.881586
10 5 8.020768 124.676347 742.209673 9.817600
10 6 8.054070 124.160826 763.856173 9.695083
10 7 8.112507 123.266459 758.080959 9.725362
10 8 7.953040 125.738084 740.075827 9.908676
10 9 8.010194 124.840915 740.096569 9.770602
8 2 True 416 416 10 0 8.000002 124.999970 735.028505 8.663520
10 1 7.942802 125.900149 743.657351 8.951396
10 2 7.997963 125.031829 728.087425 8.647516
10 3 7.973471 125.415891 748.322010 8.727342
10 4 8.024255 124.622166 747.579813 8.622497
10 5 7.932480 126.063973 729.033232 8.685499
10 6 7.984263 125.246376 743.060589 8.726448
10 7 7.970761 125.458539 755.054474 8.622214
10 8 8.084295 123.696625 742.125750 8.901402
10 9 7.907932 126.455307 748.281717 8.654654
16 2 True 416 416 10 0 7.754572 128.956184 749.789000 8.496441
10 1 7.800863 128.190950 735.840559 8.662008
10 2 7.751390 129.009128 730.700254 8.602113
10 3 7.757177 128.912881 747.271299 8.516140
10 4 7.817014 127.926081 762.900352 8.661099
10 5 7.802087 128.170833 768.960953 8.776374
10 6 7.761160 128.846720 777.239323 8.589797
10 7 7.745810 129.102051 761.098623 8.519493
10 8 7.774493 128.625751 746.984482 8.693248
10 9 7.731438 129.342049 750.234842 8.449718
32 2 True 416 416 10 0 7.699597 129.876934 804.311514 8.393414
10 1 7.700581 129.860327 746.842146 8.290410
10 2 7.709005 129.718430 755.871296 8.331478
10 3 7.710922 129.686184 743.375301 8.428369
10 4 7.711723 129.672714 741.960049 8.426096
10 5 7.717937 129.568309 733.941793 8.431178
10 6 7.732280 129.327968 742.064476 8.472528
10 7 7.689008 130.055785 749.837399 8.439362
10 8 7.694193 129.968144 746.048212 8.480441
10 9 7.715597 129.607596 764.533758 8.533400
cuda 1 2 True 416 416 10 0 33.015097 30.289173 756.636620 11.920929
10 1 32.440804 30.825377 720.205069 11.973023
10 2 32.638720 30.638456 710.690260 11.832714
10 3 32.494337 30.774593 736.654997 11.603832
10 4 32.126474 31.126976 747.756481 12.080669
10 5 32.239821 31.017542 743.244410 11.530995
10 6 32.753670 30.530930 735.150099 11.778235
10 7 32.872009 30.421019 731.841564 12.060523
10 8 32.599907 30.674934 733.180046 11.737704
10 9 32.617146 30.658722 734.117746 11.719584
2 2 True 416 416 10 0 41.050600 24.360180 767.584801 9.929657
10 1 41.081157 24.342060 731.180906 10.029256
10 2 41.724819 23.966551 738.259077 9.691656
10 3 41.070095 24.348617 715.700865 9.936154
10 4 41.777393 23.936391 733.265638 10.015130
10 5 41.593860 24.042010 736.417055 10.241807
10 6 41.215585 24.262667 728.220940 9.880304
10 7 41.324006 24.199009 733.928204 9.844542
10 8 41.747661 23.953438 716.781855 9.973347
10 9 41.212953 24.264216 719.955921 9.980798
4 2 True 416 416 10 0 48.843390 20.473599 775.704145 9.679258
10 1 48.176473 20.757020 735.307455 9.795904
10 2 48.078589 20.799279 728.606462 9.511650
10 3 47.934766 20.861685 737.453222 9.882748
10 4 48.733577 20.519733 725.023508 9.634823
10 5 48.025603 20.822227 735.261679 9.669513
10 6 48.073767 20.801365 738.612890 10.017842
10 7 47.925181 20.865858 746.457815 9.713411
10 8 47.957923 20.851612 730.696201 9.512872
10 9 48.148268 20.769179 733.915806 9.631932
8 2 True 416 416 10 0 49.872077 20.051301 778.191328 8.593932
10 1 50.699549 19.724041 731.018305 8.717254
10 2 45.246364 22.101223 731.523991 8.585066
10 3 49.813587 20.074844 738.673449 8.677602
10 4 50.128452 19.948751 735.694170 8.610740
10 5 50.059500 19.976228 723.472834 8.506611
10 6 45.941941 21.766603 732.049465 8.674756
10 7 49.874375 20.050377 742.737532 8.846745
10 8 45.806910 21.830767 741.208792 8.561283
10 9 45.206192 22.120863 727.003574 8.646011
16 2 True 416 416 10 0 57.646532 17.347097 776.599169 8.510336
10 1 57.404123 17.420352 738.630295 8.519799
10 2 57.397888 17.422244 734.946012 8.575954
10 3 57.219991 17.476410 760.316133 8.636869
10 4 57.370802 17.430469 745.356560 8.588895
10 5 57.377423 17.428458 733.257771 8.496560
10 6 58.166468 17.192036 731.886148 8.646935
10 7 57.602642 17.360315 731.846333 8.623302
10 8 57.351435 17.436355 742.234945 8.781746
10 9 57.514867 17.386809 735.044956 8.636773
32 2 True 416 416 10 0 59.267480 16.872659 766.610622 8.285634
10 1 59.953289 16.679652 734.985352 8.291133
10 2 52.717345 18.969089 737.916231 8.209787
10 3 52.571791 19.021608 756.889105 8.212481
10 4 52.565800 19.023776 738.118410 8.357562
10 5 59.229582 16.883455 740.912437 8.296773
10 6 52.939918 18.889338 719.966888 8.205108
10 7 59.452879 16.820043 738.403320 8.346349
10 8 52.687959 18.979669 743.752003 8.253429
10 9 52.519684 19.040480 732.797861 8.360378
tensorrt 1 2 True 416 416 10 0 32.256185 31.001806 11862.448454 11.571169
10 1 32.623996 30.652285 11716.783524 11.469364
10 2 32.519783 30.750513 11688.913107 11.564612
10 3 32.938353 30.359745 11777.283192 11.774659
10 4 32.098446 31.154156 11653.762102 11.580467
10 5 32.552593 30.719519 11586.960077 11.241317
10 6 32.957765 30.341864 11689.930916 11.301637
10 7 32.500883 30.768394 11901.330471 11.224627
10 8 32.231645 31.025410 11764.339685 11.325717
10 9 32.658034 30.620337 11768.515825 11.432052
2 2 True 416 416 10 0 41.332762 24.193883 11705.509424 9.914577
10 1 41.581695 24.049044 11819.612503 9.901166
10 2 40.985416 24.398923 11785.220623 9.689987
10 3 41.102091 24.329662 11800.042629 9.913146
10 4 40.803798 24.507523 11820.136547 9.708226
10 5 41.586849 24.046063 11747.811556 10.325491
10 6 41.356807 24.179816 11744.561195 9.793341
10 7 41.803208 23.921609 11721.886873 9.730399
10 8 41.358031 24.179101 11648.236513 9.858608
10 9 41.076732 24.344683 12031.765938 9.903312
4 2 True 416 416 10 0 48.808434 20.488262 11897.512436 9.983242
10 1 48.520195 20.609975 11693.880796 9.728044
10 2 47.965738 20.848215 11817.900896 9.706616
10 3 47.937642 20.860434 11917.607069 9.643972
10 4 48.326764 20.692468 11596.016884 9.389549
10 5 48.561625 20.592391 11940.795422 9.587616
10 6 48.615238 20.569682 11711.360693 9.485662
10 7 48.106437 20.787239 11645.892620 9.732932
10 8 48.652880 20.553768 11828.845024 9.661496
10 9 48.276980 20.713806 11688.106298 9.573609
8 2 True 416 416 10 0 50.509593 19.798219 11651.144028 8.472502
10 1 50.355265 19.858897 11687.902927 8.901447
10 2 50.044045 19.982398 11889.169455 8.548275
10 3 49.831045 20.067811 11802.555561 8.690566
10 4 49.940217 20.023942 11680.688381 8.634180
10 5 50.684692 19.729823 11726.379395 8.637890
10 6 50.666554 19.736886 11607.134819 8.601561
10 7 49.614715 20.155311 11641.483545 8.741736
10 8 45.288991 22.080421 11725.649595 8.792400
10 9 49.951368 20.019472 11664.314985 8.556172
16 2 True 416 416 10 0 57.895791 17.272413 11757.075310 8.454420
10 1 57.394599 17.423242 11903.642178 8.384913
10 2 57.901486 17.270714 11781.234503 8.474089
10 3 57.257827 17.464861 11862.584114 8.416958
10 4 57.900886 17.270893 11722.855806 8.738019
10 5 57.874970 17.278627 11779.921055 8.446746
10 6 58.184522 17.186701 11815.725088 8.558780
10 7 57.248497 17.467707 11799.353600 8.438826
10 8 57.323218 17.444938 11851.527452 8.447371
10 9 58.060687 17.223358 11700.683117 8.359380
32 2 True 416 416 10 0 59.918013 16.689472 11949.890852 8.227546
10 1 52.954268 18.884219 11936.690807 8.245900
10 2 59.252540 16.876914 11856.063366 8.250069
10 3 59.453221 16.819946 11819.866657 8.245450
10 4 52.823782 18.930867 11652.352810 8.260824
10 5 59.937814 16.683958 11699.779034 8.226626
10 6 59.195336 16.893223 11725.577354 8.300830
10 7 59.434976 16.825110 11632.177114 8.197069
10 8 53.711670 18.617928 11683.812618 8.247960
10 9 59.461808 16.817518 11692.517519 8.225493
tensorrt-dynamic 1 2 True 416 416 10 0 33.722234 29.654026 10214.663744 11.405230
10 1 33.344760 29.989719 10163.130760 11.441231
10 2 32.207885 31.048298 10030.030966 11.518478
10 3 33.368635 29.968262 9938.854694 11.760235
10 4 32.228426 31.028509 9924.241066 11.911869
10 5 33.119637 30.193567 10087.437153 11.619210
10 6 32.753670 30.530930 9938.045979 11.651874
10 7 34.520452 28.968334 10115.472555 11.228919
10 8 32.833924 30.456305 9982.085705 11.464119
10 9 31.725999 31.519890 10169.567823 11.517763
2 2 True 416 416 10 0 36.674076 27.267218 10154.089928 9.788334
10 1 34.905992 28.648376 10108.564377 9.919822
10 2 38.682671 25.851369 10135.166883 9.725690
10 3 37.770189 26.475906 10128.602505 9.730458
10 4 38.220550 26.163936 10192.887545 9.962976
10 5 35.161893 28.439879 10098.555088 9.943366
10 6 34.940304 28.620243 10168.661118 9.785891
10 7 38.465914 25.997043 10097.043514 9.609044
10 8 38.029604 26.295304 9929.991961 10.007560
10 9 38.326197 26.091814 10142.686605 9.811342
4 2 True 416 416 10 0 40.641496 24.605393 11178.858280 9.646863
10 1 40.764834 24.530947 11536.217451 9.618133
10 2 40.525459 24.675846 10812.536478 9.865493
10 3 40.309304 24.808168 10403.804064 9.634823
10 4 36.667102 27.272403 11289.476871 9.460688
10 5 40.309110 24.808288 10769.871712 9.692043
10 6 40.637657 24.607718 10642.199755 9.489983
10 7 40.178886 24.888694 10468.506336 9.643734
10 8 39.946703 25.033355 10529.506445 9.732276
10 9 40.112889 24.929643 10472.413301 9.521335
8 2 True 416 416 10 0 41.125919 24.315566 10136.513948 8.508503
10 1 40.832945 24.490029 10148.459196 8.661255
10 2 40.678350 24.583101 10214.094400 8.521974
10 3 40.708652 24.564803 10059.355497 8.677438
10 4 37.725898 26.506990 10028.834343 8.544073
10 5 40.822265 24.496436 9911.667585 8.550242
10 6 40.799978 24.509817 10052.909613 8.539915
10 7 40.863677 24.471611 10235.522270 8.539960
10 8 40.510879 24.684727 10168.982029 8.608490
10 9 40.664990 24.591178 10139.875412 8.745968
16 2 True 416 416 10 0 38.212976 26.169121 10636.604786 8.877695
10 1 38.328517 26.090235 10321.853638 8.753181
10 2 41.660788 24.003386 10284.081936 8.482546
10 3 41.403296 24.152666 10617.761612 8.488871
10 4 41.592596 24.042740 10567.626476 8.564413
10 5 41.526041 24.081275 10933.218002 8.424073
10 6 40.946929 24.421856 10555.508852 8.465439
10 7 41.796725 23.925319 10436.465979 8.487754
10 8 41.767356 23.942143 10323.004246 8.584604
10 9 42.019910 23.798242 10236.896515 8.871555
32 2 True 416 416 10 0 42.181365 23.707151 10011.900425 8.353427
10 1 42.641372 23.451403 10015.968561 8.194283
10 2 42.212262 23.689799 10113.837481 8.239280
10 3 38.904930 25.703683 10202.510595 8.308575
10 4 38.740709 25.812641 10094.983816 8.282896
10 5 42.082953 23.762591 10029.658556 8.198120
10 6 42.069419 23.770235 10193.944931 8.187093
10 7 42.496558 23.531318 10064.801931 8.159630
10 8 38.847153 25.741912 10182.497501 8.184604
10 9 42.213536 23.689084 10184.962988 8.262347

Plot the experimental data

Plot accuracy

In [15]:
def plot_accuracy(df, groupby_level='batch_enabled',
                  accuracy_metric=['mAP','mAP_large','mAP_medium','mAP_small'],
                  save_fig=False, save_fig_name='accuracy.resizing.', resize_type=['internal','external'],
                  title='', figsize=[default_figwidth, 8], rot=90, colormap=cm.autumn):
    # Bars.
    df_bar = pd.DataFrame(
        data=df[accuracy_metric].values, columns=accuracy_metric,
        index=pd.MultiIndex.from_tuples(
            tuples=[ (m,be,nr) for (m,b,bs,bc,be,ih,iw,nr) in df.index.values ],
            names=[ 'model','batch_enabled','num_reps' ]
        )
    )

    # Plot.
    mean = df_bar.groupby(level=df_bar.index.names[:-1]).mean()
    std = pd.Series()
    axes = mean \
            .groupby(level=groupby_level) \
            .plot(yerr=std, kind='bar', grid=True, rot=rot,
                  figsize=figsize, width=default_barwidth, fontsize=default_fontsize, colormap=colormap)

    xlabel = 'Model'
    xtics  = df_bar.index.get_level_values('model').drop_duplicates()
    ylabel = 'mAP %'
    for count, ax in enumerate(axes):
        # Title.
        ax.set_title('Accuracy with %s resizing' % resize_type[count])
        # X label.
        ax.set_xlabel(xlabel)
        # X ticks.
        ax.set_xticklabels(xtics)
        # Y axis.
        ax.set_ylabel(ylabel)
        if save_fig:
            save_fig_path = os.path.join(save_fig_dir, '%s.%s' % (save_fig_name+resize_type[count], save_fig_ext))
            ax.figure.savefig(save_fig_path, dpi=default_figdpi, bbox_inches='tight')        

plot_accuracy(dfs, save_fig=True)