update benchmark results

Author: Coderx7

ImageNet/training_scripts/imagenet_training/results/add_acc_to_benchmarks.py

@@ -1,3 +1,4 @@
+import os
 import pandas as pd
 import argparse
 
@@ -11,7 +12,7 @@
 )
 parser.add_argument(
     "--bench-csv",
-    default="./benchmark_inference_GTX1080_fp32_small_torch1.10.csv",
+    default="",
     type=str,
     metavar="FILENAME",
     help="the csv file for which you want to add accuracy",
@@ -23,10 +24,18 @@ def add_acc_to_csv(imagenet_results, csv_filename):
     df_imagenet_accs = df_imagenet_results[["model", "top1", "top5"]]
     df_csv = pd.read_csv(csv_filename)
     df_csv_acc = pd.merge(df_csv, df_imagenet_accs, on=["model"])
-    df_csv_acc.to_csv(csv_filename.replace(".csv", "_with_accuracy.csv"), index=False)
-    print(f"{csv_filename} is done")
+    df_csv_acc.to_csv(csv_filename.replace(".csv", "_acc.csv"), index=False)
+    print(f"--{csv_filename:<60} is processed.")
 
 
 if __name__ == "__main__":
     args = parser.parse_args()
-    add_acc_to_csv(args.imagenet_results, args.bench_csv)
+    if args.bench_csv:
+        add_acc_to_csv(args.imagenet_results, args.bench_csv)
+    else:
+        print('Fetching all benchmark logs and adding accuracy to all in bulk...')
+        current_dir = os.path.dirname(os.path.abspath(__file__))
+        for file in os.listdir(current_dir):
+            if 'bench' in file and file.endswith('.csv'):
+                add_acc_to_csv(args.imagenet_results, file)
+        print(f'all done.')

ImageNet/training_scripts/imagenet_training/results/bench_infer_GTX1080_fp32_normal_known_t1.10.csv

@@ -0,0 +1,37 @@
+model,infer_samples_per_sec,infer_step_time,infer_batch_size,infer_img_size,infer_gmacs,infer_macts,param_count
+simplenetv1_small_m1_075,2893.91,88.444,256,224,0.76,1.49,3.29
+simplenetv1_small_m2_075,2478.41,103.276,256,224,0.96,1.71,3.29
+vit_tiny_r_s16_p8_224,2337.23,109.515,256,224,0.44,2.06,6.34
+simplenetv1_5m_m1,2105.06,121.594,256,224,1.35,1.98,5.75
+simplenetv1_5m_m2,1754.25,145.914,256,224,1.69,2.28,5.75
+resnet18,1750.38,146.235,256,224,1.82,2.48,11.69
+regnetx_006,1620.25,157.982,256,224,0.61,3.98,6.2
+mobilenetv3_large_100,1491.86,171.579,256,224,0.23,4.41,5.48
+tf_mobilenetv3_large_minimal_100,1476.29,173.391,256,224,0.22,4.4,3.92
+tf_mobilenetv3_large_075,1474.77,173.567,256,224,0.16,4.0,3.99
+ghostnet_100,1390.19,184.13,256,224,0.15,3.55,5.18
+tinynet_b,1345.82,190.2,256,188,0.21,4.44,3.73
+tf_mobilenetv3_large_100,1325.06,193.181,256,224,0.23,4.41,5.48
+mnasnet_100,1183.69,216.252,256,224,0.33,5.46,4.38
+mobilenetv2_100,1101.58,232.375,256,224,0.31,6.68,3.5
+simplenetv1_9m_m1,1048.91,244.043,256,224,2.78,3.28,9.51
+resnet34,1030.4,248.426,256,224,3.67,3.74,21.8
+deit_tiny_patch16_224,990.85,258.344,256,224,1.26,5.97,5.72
+efficientnet_lite0,977.76,261.802,256,224,0.4,6.74,4.65
+simplenetv1_9m_m2,900.45,284.28,256,224,3.56,3.73,9.51
+tf_efficientnet_lite0,876.66,291.999,256,224,0.4,6.74,4.65
+dla34,834.35,306.803,256,224,3.07,5.02,15.74
+mobilenetv2_110d,824.4,310.505,256,224,0.45,8.71,4.52
+resnet26,771.1,331.973,256,224,2.36,7.35,16.0
+repvgg_b0,751.01,340.855,256,224,3.41,6.15,15.82
+crossvit_9_240,606.2,422.278,256,240,1.85,9.52,8.55
+vgg11,576.32,444.176,256,224,7.61,7.44,132.86
+vit_base_patch32_224_sam,561.99,455.502,256,224,4.41,5.01,88.22
+vgg11_bn,504.29,507.616,256,224,7.62,7.44,132.87
+densenet121,435.3,588.073,256,224,2.87,6.9,7.98
+vgg13,363.69,703.869,256,224,11.31,12.25,133.05
+vgg13_bn,315.85,810.498,256,224,11.33,12.25,133.05
+vgg16,302.84,845.295,256,224,15.47,13.56,138.36
+vgg16_bn,265.99,962.41,256,224,15.5,13.56,138.37
+vgg19,259.82,985.288,256,224,19.63,14.86,143.67
+vgg19_bn,229.77,1114.113,256,224,19.66,14.86,143.68

ImageNet/training_scripts/imagenet_training/results/bench_infer_GTX1080_fp32_normal_known_t1.10_acc.csv

@@ -0,0 +1,37 @@
+model,infer_samples_per_sec,infer_step_time,infer_batch_size,infer_img_size,infer_gmacs,infer_macts,param_count,top1,top5
+simplenetv1_small_m1_075,2893.91,88.444,256,224,0.76,1.49,3.29,67.784,87.718
+simplenetv1_small_m2_075,2478.41,103.276,256,224,0.96,1.71,3.29,68.506,88.15
+vit_tiny_r_s16_p8_224,2337.23,109.515,256,224,0.44,2.06,6.34,71.792,90.822
+simplenetv1_5m_m1,2105.06,121.594,256,224,1.35,1.98,5.75,71.548,89.94
+simplenetv1_5m_m2,1754.25,145.914,256,224,1.69,2.28,5.75,72.03,90.324
+resnet18,1750.38,146.235,256,224,1.82,2.48,11.69,69.744,89.082
+regnetx_006,1620.25,157.982,256,224,0.61,3.98,6.2,73.86,91.672
+mobilenetv3_large_100,1491.86,171.579,256,224,0.23,4.41,5.48,75.766,92.544
+tf_mobilenetv3_large_minimal_100,1476.29,173.391,256,224,0.22,4.4,3.92,72.25,90.63
+tf_mobilenetv3_large_075,1474.77,173.567,256,224,0.16,4.0,3.99,73.436,91.344
+ghostnet_100,1390.19,184.13,256,224,0.15,3.55,5.18,73.974,91.46
+tinynet_b,1345.82,190.2,256,188,0.21,4.44,3.73,74.976,92.184
+tf_mobilenetv3_large_100,1325.06,193.181,256,224,0.23,4.41,5.48,75.518,92.604
+mnasnet_100,1183.69,216.252,256,224,0.33,5.46,4.38,74.658,92.112
+mobilenetv2_100,1101.58,232.375,256,224,0.31,6.68,3.5,72.97,91.02
+simplenetv1_9m_m1,1048.91,244.043,256,224,2.78,3.28,9.51,73.792,91.486
+resnet34,1030.4,248.426,256,224,3.67,3.74,21.8,75.114,92.284
+deit_tiny_patch16_224,990.85,258.344,256,224,1.26,5.97,5.72,72.172,91.114
+efficientnet_lite0,977.76,261.802,256,224,0.4,6.74,4.65,75.476,92.512
+simplenetv1_9m_m2,900.45,284.28,256,224,3.56,3.73,9.51,74.23,91.748
+tf_efficientnet_lite0,876.66,291.999,256,224,0.4,6.74,4.65,74.832,92.174
+dla34,834.35,306.803,256,224,3.07,5.02,15.74,74.62,92.072
+mobilenetv2_110d,824.4,310.505,256,224,0.45,8.71,4.52,75.038,92.184
+resnet26,771.1,331.973,256,224,2.36,7.35,16.0,75.3,92.578
+repvgg_b0,751.01,340.855,256,224,3.41,6.15,15.82,75.16,92.418
+crossvit_9_240,606.2,422.278,256,240,1.85,9.52,8.55,73.96,91.968
+vgg11,576.32,444.176,256,224,7.61,7.44,132.86,69.028,88.626
+vit_base_patch32_224_sam,561.99,455.502,256,224,4.41,5.01,88.22,73.694,91.01
+vgg11_bn,504.29,507.616,256,224,7.62,7.44,132.87,70.36,89.802
+densenet121,435.3,588.073,256,224,2.87,6.9,7.98,75.584,92.652
+vgg13,363.69,703.869,256,224,11.31,12.25,133.05,69.926,89.246
+vgg13_bn,315.85,810.498,256,224,11.33,12.25,133.05,71.594,90.376
+vgg16,302.84,845.295,256,224,15.47,13.56,138.36,71.59,90.382
+vgg16_bn,265.99,962.41,256,224,15.5,13.56,138.37,73.35,91.504
+vgg19,259.82,985.288,256,224,19.63,14.86,143.67,72.366,90.87
+vgg19_bn,229.77,1114.113,256,224,19.66,14.86,143.68,74.214,91.848

ImageNet/training_scripts/imagenet_training/results/bench_infer_GTX1080_fp32_normal_known_t1.11.csv

@@ -0,0 +1,37 @@
+model,infer_samples_per_sec,infer_step_time,infer_batch_size,infer_img_size,infer_gmacs,infer_macts,param_count
+simplenetv1_small_m1_075,2632.69,97.222,256,224,0.76,1.49,3.29
+simplenetv1_small_m2_075,2276.71,112.424,256,224,0.96,1.71,3.29
+simplenetv1_5m_m1,2122.88,120.574,256,224,1.35,1.98,5.75
+vit_tiny_r_s16_p8_224,2114.01,121.08,256,224,0.44,2.06,6.34
+simplenetv1_5m_m2,1808.81,141.512,256,224,1.69,2.28,5.75
+regnetx_006,1569.23,163.12,256,224,0.61,3.98,6.2
+resnet18,1564.65,163.594,256,224,1.82,2.48,11.69
+resnet34,975.71,262.351,256,224,3.67,3.74,21.8
+tf_mobilenetv3_large_minimal_100,943.92,271.189,256,224,0.22,4.4,3.92
+deit_tiny_patch16_224,921.07,277.916,256,224,1.26,5.97,5.72
+simplenetv1_9m_m1,915.22,279.692,256,224,2.78,3.28,9.51
+resnet26,835.45,306.402,256,224,2.36,7.35,16.0
+dla34,815.16,314.023,256,224,3.07,5.02,15.74
+simplenetv1_9m_m2,803.52,318.578,256,224,3.56,3.73,9.51
+tf_mobilenetv3_large_075,798.98,320.387,256,224,0.16,4.0,3.99
+repvgg_b0,772.16,331.514,256,224,3.41,6.15,15.82
+ghostnet_100,740.79,345.555,256,224,0.15,3.55,5.18
+mobilenetv3_large_100,717.75,356.649,256,224,0.23,4.41,5.48
+tf_mobilenetv3_large_100,695.46,368.08,256,224,0.23,4.41,5.48
+mobilenetv2_100,624.46,409.93,256,224,0.31,6.68,3.5
+tinynet_b,613.99,416.926,256,188,0.21,4.44,3.73
+crossvit_9_240,576.43,444.093,256,240,1.85,9.52,8.55
+mnasnet_100,522.9,489.551,256,224,0.33,5.46,4.38
+vit_base_patch32_224_sam,508.95,502.975,256,224,4.41,5.01,88.22
+vgg11,487.59,525.002,256,224,7.61,7.44,132.86
+densenet121,461.35,554.857,256,224,2.87,6.9,7.98
+vgg11_bn,454.48,563.254,256,224,7.62,7.44,132.87
+mobilenetv2_110d,451.89,566.488,256,224,0.45,8.71,4.52
+efficientnet_lite0,428.07,598.004,256,224,0.4,6.74,4.65
+tf_efficientnet_lite0,414.3,617.891,256,224,0.4,6.74,4.65
+vgg13,298.11,858.716,256,224,11.31,12.25,133.05
+vgg13_bn,277.48,922.559,256,224,11.33,12.25,133.05
+vgg16,254.87,1004.418,256,224,15.47,13.56,138.36
+vgg16_bn,238.54,1073.189,256,224,15.5,13.56,138.37
+vgg19,224.34,1141.111,256,224,19.63,14.86,143.67
+vgg19_bn,212.07,1207.101,256,224,19.66,14.86,143.68

ImageNet/training_scripts/imagenet_training/results/bench_infer_GTX1080_fp32_normal_known_t1.11_NCHW.csv

@@ -0,0 +1,37 @@
+model,infer_samples_per_sec,infer_step_time,infer_batch_size,infer_img_size,infer_gmacs,infer_macts,param_count,top1,top5
+simplenetv1_small_m1_075,2632.69,97.222,256,224,0.76,1.49,3.29,67.784,87.718
+simplenetv1_small_m2_075,2276.71,112.424,256,224,0.96,1.71,3.29,68.506,88.15
+simplenetv1_5m_m1,2122.88,120.574,256,224,1.35,1.98,5.75,71.548,89.94
+vit_tiny_r_s16_p8_224,2114.01,121.08,256,224,0.44,2.06,6.34,71.792,90.822
+simplenetv1_5m_m2,1808.81,141.512,256,224,1.69,2.28,5.75,72.03,90.324
+regnetx_006,1569.23,163.12,256,224,0.61,3.98,6.2,73.86,91.672
+resnet18,1564.65,163.594,256,224,1.82,2.48,11.69,69.744,89.082
+resnet34,975.71,262.351,256,224,3.67,3.74,21.8,75.114,92.284
+tf_mobilenetv3_large_minimal_100,943.92,271.189,256,224,0.22,4.4,3.92,72.25,90.63
+deit_tiny_patch16_224,921.07,277.916,256,224,1.26,5.97,5.72,72.172,91.114
+simplenetv1_9m_m1,915.22,279.692,256,224,2.78,3.28,9.51,73.792,91.486
+resnet26,835.45,306.402,256,224,2.36,7.35,16.0,75.3,92.578
+dla34,815.16,314.023,256,224,3.07,5.02,15.74,74.62,92.072
+simplenetv1_9m_m2,803.52,318.578,256,224,3.56,3.73,9.51,74.23,91.748
+tf_mobilenetv3_large_075,798.98,320.387,256,224,0.16,4.0,3.99,73.436,91.344
+repvgg_b0,772.16,331.514,256,224,3.41,6.15,15.82,75.16,92.418
+ghostnet_100,740.79,345.555,256,224,0.15,3.55,5.18,73.974,91.46
+mobilenetv3_large_100,717.75,356.649,256,224,0.23,4.41,5.48,75.766,92.544
+tf_mobilenetv3_large_100,695.46,368.08,256,224,0.23,4.41,5.48,75.518,92.604
+mobilenetv2_100,624.46,409.93,256,224,0.31,6.68,3.5,72.97,91.02
+tinynet_b,613.99,416.926,256,188,0.21,4.44,3.73,74.976,92.184
+crossvit_9_240,576.43,444.093,256,240,1.85,9.52,8.55,73.96,91.968
+mnasnet_100,522.9,489.551,256,224,0.33,5.46,4.38,74.658,92.112
+vit_base_patch32_224_sam,508.95,502.975,256,224,4.41,5.01,88.22,73.694,91.01
+vgg11,487.59,525.002,256,224,7.61,7.44,132.86,69.028,88.626
+densenet121,461.35,554.857,256,224,2.87,6.9,7.98,75.584,92.652
+vgg11_bn,454.48,563.254,256,224,7.62,7.44,132.87,70.36,89.802
+mobilenetv2_110d,451.89,566.488,256,224,0.45,8.71,4.52,75.038,92.184
+efficientnet_lite0,428.07,598.004,256,224,0.4,6.74,4.65,75.476,92.512
+tf_efficientnet_lite0,414.3,617.891,256,224,0.4,6.74,4.65,74.832,92.174
+vgg13,298.11,858.716,256,224,11.31,12.25,133.05,69.926,89.246
+vgg13_bn,277.48,922.559,256,224,11.33,12.25,133.05,71.594,90.376
+vgg16,254.87,1004.418,256,224,15.47,13.56,138.36,71.59,90.382
+vgg16_bn,238.54,1073.189,256,224,15.5,13.56,138.37,73.35,91.504
+vgg19,224.34,1141.111,256,224,19.63,14.86,143.67,72.366,90.87
+vgg19_bn,212.07,1207.101,256,224,19.66,14.86,143.68,74.214,91.848

ImageNet/training_scripts/imagenet_training/results/bench_infer_GTX1080_fp32_normal_known_t1.11_NCHW_acc.csv

@@ -0,0 +1,37 @@
+model,infer_samples_per_sec,infer_step_time,infer_batch_size,infer_img_size,infer_gmacs,infer_macts,param_count
+vit_tiny_r_s16_p8_224,1880.84,136.09,256,224,0.44,2.06,6.34
+simplenetv1_small_m1_075,1585.06,161.491,256,224,0.76,1.49,3.29
+simplenetv1_small_m2_075,1303.45,196.384,256,224,0.96,1.71,3.29
+simplenetv1_5m_m1,1182.83,216.412,256,224,1.35,1.98,5.75
+deit_tiny_patch16_224,988.98,258.833,256,224,1.26,5.97,5.72
+simplenetv1_5m_m2,877.27,291.797,256,224,1.69,2.28,5.75
+resnet18,873.32,293.112,256,224,1.82,2.48,11.69
+crossvit_9_240,600.23,426.481,256,240,1.85,9.52,8.55
+vit_base_patch32_224_sam,568.15,450.56,256,224,4.41,5.01,88.22
+tinynet_b,526.6,486.118,256,188,0.21,4.44,3.73
+resnet26,525.93,486.735,256,224,2.36,7.35,16.0
+tf_mobilenetv3_large_075,501.4,510.547,256,224,0.16,4.0,3.99
+resnet34,491.53,520.806,256,224,3.67,3.74,21.8
+simplenetv1_9m_m1,479.06,534.361,256,224,2.78,3.28,9.51
+regnetx_006,477.32,536.307,256,224,0.61,3.98,6.2
+dla34,469.89,544.784,256,224,3.07,5.02,15.74
+repvgg_b0,453.58,564.372,256,224,3.41,6.15,15.82
+ghostnet_100,405.05,632.006,256,224,0.15,3.55,5.18
+simplenetv1_9m_m2,402.67,635.734,256,224,3.56,3.73,9.51
+tf_mobilenetv3_large_minimal_100,402.04,636.729,256,224,0.22,4.4,3.92
+mobilenetv3_large_100,397.73,643.626,256,224,0.23,4.41,5.48
+tf_mobilenetv3_large_100,383.01,668.367,256,224,0.23,4.41,5.48
+mobilenetv2_100,293.62,871.86,256,224,0.31,6.68,3.5
+densenet121,293.49,872.248,256,224,2.87,6.9,7.98
+mnasnet_100,260.29,983.481,256,224,0.33,5.46,4.38
+vgg11,259.27,987.367,256,224,7.61,7.44,132.86
+vgg11_bn,247.67,516.792,128,224,7.62,7.44,132.87
+mobilenetv2_110d,229.73,1114.341,256,224,0.45,8.71,4.52
+efficientnet_lite0,223.31,1146.379,256,224,0.4,6.74,4.65
+tf_efficientnet_lite0,218.86,1169.677,256,224,0.4,6.74,4.65
+vgg13,151.9,842.666,128,224,11.31,12.25,133.05
+vgg13_bn,142.62,897.488,128,224,11.33,12.25,133.05
+vgg16,122.08,1048.441,128,224,15.47,13.56,138.36
+vgg16_bn,115.59,1107.345,128,224,15.5,13.56,138.37
+vgg19,101.94,1255.594,128,224,19.63,14.86,143.67
+vgg19_bn,97.15,1317.532,128,224,19.66,14.86,143.68