Skip to main content Skip to secondary navigation

CIFAR10 Training

Main content start
 Submission DateModelTime to 94% AccuracyCost (USD)Max AccuracyHardwareFramework

Dec 2019

Custom Resnet 9

Santiago Akle Serrano, Hadi Pour Ansari, Vipul Gupta, Dennis DeCoste

source

0:00:10N/A94.23%Tesla V100 * 8 GPU / 32 GB / 40 CPUPytorch 1.1.0

Jan 2020

Custom ResNet 9

Ajay Uppili Arasanipalai

source

0:00:11N/A94.05%IBM AC922 + 4 * Nvidia Tesla V100 (NCSA HAL)PyTorch 1.1.0

Oct 2019

Kakao Brain Custom ResNet9

clint@KakaoBrain

source

0:00:28N/A94.04%Tesla V100 * 4 GPU / 488 GB / 56 CPU (Kakao Brain BrainCloud)PyTorch 1.1.0

May 2019

BaiduNet9P

Baidu USA GAIT LEOPARD team: Baopu Li, Zhiyu Cheng, Yingze Bao

source

0:00:45$0.1194.18%Baidu Cloud Tesla 8*V100-16GB/448 GB/96 CPUPyTorch v1.0.1 and PaddlePaddle

Oct 2019

Kakao Brain Custom ResNet9

clint@KakaoBrain

source

0:00:58N/A94.20%Tesla V100 * 1 GPU / 488 GB / 56 CPU (Kakao Brain BrainCloud)PyTorch 1.1.0

May 2019

BaiduNet9

Baidu USA GAIT LEOPARD team: Baopu Li, Zhiyu Cheng, Yingze Bao

source

0:01:12$0.0294.10%Baidu Cloud Tesla V100*1-16GB/56 GB/12 CPUPyTorch v1.0.1 and PaddlePaddle

Apr 2019

Custom ResNet 9

Ajay Uppili Arasanipalai

source

0:01:14N/A94.06%IBM AC922 + Nvidia Tesla V100 (Nimbix np9g1)PowerAI 1.6.0 + PyTorch 1.0.1

Nov 2018

Custom ResNet 9

David Page, myrtle.ai

source

0:01:15$0.0694.08%V100 (AWS p3.2xlarge)pytorch 0.4.0

Aug 2019

BaiduNet9

Chuan Li

source

0:01:42$0.0494.02%Lambda GPU Cloud - 4x GTX 1080 Tifastai / Pytorch 1.0.0

Apr 2018

Custom Wide Resnet

fast.ai + students team: Jeremy Howard, Andrew Shaw, Brett Koonce, Sylvain Gugger

source

0:02:54$1.1894.39%8 * V100 (AWS p3.16xlarge)fastai / pytorch

Apr 2018

Resnet18 + minor modifications

bkj

source

0:05:41$0.2994.34%V100 (AWS p3.2xlarge)pytorch 0.3.1.post2

Apr 2018

Custom Wide Resnet

fast.ai + students team: Jeremy Howard, Andrew Shaw, Brett Koonce, Sylvain Gugger

source

0:06:45$0.2694.20%Paperspace Volta (V100)fastai / pytorch

Apr 2018

KervResNet34

Chen Wang

source

0:35:37N/A95.29%1 GPU (Nvidia GeForce GTX 1080 Ti)PyTorch 0.3.1

Jan 2018

ResNet50

DIUX

source

1:07:55$3.4694.60%p3.2xlargetensorflow 1.5, tensorpack 0.8.1

Oct 2017

ResNet 164 (without bottleneck)

Stanford DAWN

source

2:31:42N/A94.46%1 P100 / 512 GB / 56 CPU (DAWN Internal Cluster)PyTorch v0.1.12

Feb 2018

ResNet 164 (without bottleneck)

Stanford DAWN

source

2:47:50N/A94.18%1 P100 / 512 GB / 56 CPU (DAWN Internal Cluster)TensorFlow v1.3

Oct 2017

ResNet 164 (with bottleneck)

Stanford DAWN

source

3:01:52N/A94.82%1 P100 / 512 GB / 56 CPU (DAWN Internal Cluster)PyTorch v0.1.12

Jan 2018

ResNet50

DIUX

source

3:18:50$3.7894.51%g3.4xlargetensorflow 1.5, tensorpack 0.8.1

Oct 2017

ResNet 164 (without bottleneck)

Stanford DAWN

source

3:20:27N/A94.19%1 P100 / 512 GB / 56 CPU (DAWN Internal Cluster)TensorFlow v1.2

Oct 2017

ResNet 164 (with bottleneck)

Stanford DAWN

source

3:29:30N/A94.46%1 P100 / 512 GB / 56 CPU (DAWN Internal Cluster)TensorFlow v1.2

Oct 2017

ResNet 164 (with bottleneck)

Stanford DAWN

source

9:03:29$8.7694.91%1 K80 / 30 GB / 8 CPU (Google Cloud)PyTorch v0.1.12

Oct 2017

ResNet 164 (with bottleneck)

Stanford DAWN

source

9:16:32$8.3594.61%1 K80 / 61 GB / 4 CPU (Amazon EC2 [p2.xlarge])PyTorch v0.1.12

Oct 2017

ResNet 164 (without bottleneck)

Stanford DAWN

source

9:37:52$9.3194.37%1 K80 / 30 GB / 8 CPU (Google Cloud)PyTorch v0.1.12

Oct 2017

ResNet 164 (without bottleneck)

Stanford DAWN

source

10:02:45$9.7194.31%1 K80 / 30 GB / 8 CPU (Google Cloud)TensorFlow v1.2

Oct 2017

ResNet 164 (without bottleneck)

Stanford DAWN

source

10:32:14$9.4894.32%1 K80 / 61 GB / 4 CPU (Amazon EC2 [p2.xlarge])PyTorch v0.1.12

Oct 2017

ResNet 164 (with bottleneck)

Stanford DAWN

source

10:53:08$9.8094.58%1 K80 / 61 GB / 4 CPU (Amazon EC2 [p2.xlarge])TensorFlow v1.2

Oct 2017

ResNet 164 (with bottleneck)

Stanford DAWN

source

11:00:22$10.6494.45%1 K80 / 30 GB / 8 CPU (Google Cloud)TensorFlow v1.2

Oct 2017

ResNet 164 (without bottleneck)

Stanford DAWN

source

3 days, 7:26:54$42.3594.37%60 GB / 16 CPU (Google Cloud [n1-standard-16])PyTorch v0.1.12

Oct 2017

ResNet 164 (without bottleneck)

Stanford DAWN

source

3 days, 22:09:47$50.1994.04%60 GB / 16 CPU (Google Cloud [n1-standard-16])TensorFlow v1.2

Oct 2017

ResNet 164 (with bottleneck)

Stanford DAWN

source

4 days, 1:10:39$51.8094.79%60 GB / 16 CPU (Google Cloud [n1-standard-16])PyTorch v0.1.12

Oct 2017

ResNet 164 (with bottleneck)

Stanford DAWN

source

4 days, 6:48:08$54.7994.58%60 GB / 16 CPU (Google Cloud [n1-standard-16])TensorFlow v1.2