AlexNet is a CNN network developed in 2012 by Alex Krizhevsky using five-layer convolution and three-layer ReLU layer, and won the ImageNet competition (ILSVRC). AlexNet proves the effectiveness in classification (15.3% error rate) of CNN, against the 25% error rate by previous image recognition tools. The emergence of this network marks a milestone for deep learning applications in the computer vision field.
AlexNet is also a common performance indicator tool for deep learning framework. TensorFlow provides the alexnet_benchmark.py tool to test GPU and CPU performance. This document uses AlexNet as an example to illustrate how to run a GPU application in Alibaba Cloud Container Service easily and quickly.
Create a GN5 GPU cluster in Container Service console.Create a gn4 GPU cloud server cluster
This operation is based on the Container Service Beijing HPC or GN4 type GPU ECS instance.
- Log on to the Container Service console.
- Click in the left-side navigation pane.
- Enter the application name (alexNet in the example) and select the Beijing HPC or GN4 ECS cluster, and click Next step.
- Configure the application.
registry.cn-beijing.aliyuncs.com/tensorflow-samples/alexnet_benchmark:1.0.0-devel-gpuin the Image Name field.
- In the Container section, enter the command in the Command field. For example, enter
python /alexnet_benchmark.py --batch_size 128 -num_batches 100.
- Click the button in the Label section. Enter the Alibaba Cloud
gpuextension label. Enter
aliyun.gpuin the Tag Name field, and the number of scheduling GPUs (
1in this example) in the Tag Value field.
- Click Create after completing the settings.
You can view the created alexNet application on the Application List page.
In this way, you can check the performance of AlexNet on EGS or HPC by means of the container Log Service in Container Service console.