Featured

AWS DL (Deep Learning) Containers: Rapidly Deploy Custom Machine Learning Environments

Published

2 years ago

April 11, 2019

AWS recently released a service called AWS DL Containers aimed at deep learning researchers. The service involves pre-configuring and validating Docker images and then pre-installing them with deep learning frameworks that can be used to deploy custom ML environments in a rapid manner.

AWS DL Containers currently support Apache MXNet and TensorFlow, but will keep adding more frameworks like Facebook’s PyTorch. Unveiled at the Santa Clara AWS Summit in March 2019, DL Containers can be used for inferencing and training purposes. They’re the company’s response to EKS and ECS users’ call to AWS to create a simple way to deploy TensorFlow workloads to the cloud.

Support for more services will follow, per AWS Chief Evangelist Jeff Barr. He also said that the images would be made available for free, and can be used pre-configured, or customized to suit the needs of the workload by adding packages and libraries.

There are several types of AWS DL Containers available, all of which are based on combinations of the following criteria:

Framework – TensorFlow or MXNet.
Mode – Training or Inference. You can train on a single node or on a multi-node cluster.
Environment – CPU or GPU.
Python Version – 2.7 or 3.6.
Distributed Training – Availability of the Horovod framework.
Operating System – Ubuntu 16.04.

How to Use Amazon Deep Learning Containers

The setup part is fairly simple. In the example shown by Barr, the user creates an ECS cluster with an instance like p2.8xlarge, as shown below:

$ aws ec2 run-instances –image-id ami-0ebf2c738e66321e6
–count 1 –instance-type p2.8xlarge
–key-name keys-jbarr-us-east …

Next check that the cluster is running and that the ECS Container Agent is active, and create a text file for task definition:

{
“requiresCompatibilities”: [
“EC2”
],
“containerDefinitions”: [
{
“command”: [
“tensorflow_model_server –port=8500 –rest_api_port=8501 –model_name=saved_model_half_plus_two_gpu –model_base_path=/models/saved_model_half_plus_two_gpu”
],
“entryPoint”: [
“sh”,
“-c”
],
“name”: “EC2TFInference”,
“image”: “841569659894.dkr.ecr.us-east-1.amazonaws.com/sample_tf_inference_images:gpu_with_half_plus_two_model”,
“memory”: 8111,
“cpu”: 256,
“resourceRequirements”: [
{
“type”: “GPU”,
“value”: “1”
}
],
“essential”: true,
“portMappings”: [
{
“hostPort”: 8500,
“protocol”: “tcp”,
“containerPort”: 8500
},
{
“hostPort”: 8501,
“protocol”: “tcp”,
“containerPort”: 8501
},
{
“containerPort”: 80,
“protocol”: “tcp”
}
],
“logConfiguration”: {
“logDriver”: “awslogs”,
“options”: {
“awslogs-group”: “/ecs/TFInference”,
“awslogs-region”: “us-east-1”,
“awslogs-stream-prefix”: “ecs”
}
}
}
],
“volumes”: [],
“networkMode”: “bridge”,
“placementConstraints”: [],
“family”: “Ec2TFInference”
}

The task definition is then registered using container specs like Mode and Environment, and the revision number is captured. In this case, 3:

The task definition and revision number are then used to create a service, after which you can navigate to the task in the console and find the external binding for port 8501.

After training the model on the function y=ax+b, with a being 0.5 and b being 2, inferences can be run for 1.0, 2.0 and 5.0 as inputs, and the predicted values will be 2.5, 3.0 and 4.5.

$ curl -d ‘{“instances”: [1.0, 2.0, 5.0]}’
-X POST http://xx.xxx.xx.xx:8501/v1/models/saved_model_half_plus_two_gpu:predict
{
“predictions”: [2.5, 3.0, 4.5
]
}

The example Barr gave was intended to show users how easy it is to perform inferencing using the new DL Containers service and a pre-trained model. The flexible nature of the service allows users to launch a training model, do the training and then run inferences.

The big cloud players are now introducing ready-to-deploy frameworks and hardware acceleration. Google announced TensorFlow 2.0 alpha for its GCP, and Microsoft and NVIDIA recently announced some key integrations:

“By integrating NVIDIA TensorRT with ONNX Runtime and RAPIDS with Azure Machine Learning service, we’ve made it easier for machine learning practitioners to leverage NVIDIA GPUs across their data science workflows,” according to Kari Briski, Senior Director of Product Management for Accelerated Computing Software at NVIDIA.

AWS Deep Learning Containers is the latest addition to the broad and deep list of services aimed at data scientists and deep learning researchers. They’re available through Amazon ECR at no cost, as is the case with most of their free services that only charge you for resource usage.

Source: https://1reddrop.com/2019/04/11/aws-dl-deep-learning-containers-rapidly-deploy-custom-machine-learning-environments/?utm_source=rss&utm_medium=rss&utm_campaign=aws-dl-deep-learning-containers-rapidly-deploy-custom-machine-learning-environments

Related Topics:Amazon blockchain environment

Up Next

White House Sets The Most Unclear and Ambiguous AI Goal Yet

Don't Miss

Patent Applications per Country

Blockchain

ETH Mining Pool Sparkpool To Shut Down Amid China’s Pressure

Published

9 hours ago

September 28, 2021

Republished by Plato

ETH mining pool Sparkpool will shut down on September 30 due to the increasing regulatory pressure in China as the crackdown is taking another victim along so let’s read more in our latest Ethereum news today.

The second biggest ETH mining pool Sparkpool announced that it halted access to new users in China and the company also plans to abandon all of its services as of September 30^th. Based in Hangzhou, China, the ETH mining pool is the second biggest mining pool responsible for more than 20% of the total hash rate. The company plans to seize all operations due to teh recent Chinese crackdown as the announcement published reads that it suspended access to the new users that are located in Mainland China.

People’s Bank Of China Official, btc, bitcoin, mining

This decision could have been expected to a large extent because of the open hostile stance against the entire crypto industry. The latest news that came from the country showed that all digital assets trading is banned. This forced other exchanges like Huobi as one of the biggest exchanges to stop providing any of the services to Chinese users however Sparkpool decided to go one step further. The statement reads that a complete shutdown for all of the services and operations for all the existing users has been planned for September 30^th.

The company then argued that the decision came under the premise of ensuring the safety of the users’ assets. Sparkpool promised to provide even more specific information on clearing and refunding rules in the next announcements or in emails to its users. The closure of the second-biggest ETH mining pool would lead to damages to the ETH hash rate and this metric is quite crucial for the network’s security as it reached a new ATH but it could drop in the next few weeks.

Ethereum Slides To $2K, btc, bitcoin, altcoins,

As recently reported, Though this is not the first time that China invoked chaos on the market, the prices of most coins dropped which only exacerbated the recent global market fears that were caused by the potential collapse of the property giant Evergrande. The Chinese company dubbed Evergrande was the main suspect of the recent market crash and the Chinese unstable markets because the real estate giant lost a debt of $305 billion. Some are worried that the Chinese property developer will not be able to pay back what it owes and it could lead to a massive collapse.

DC Forecasts is a leader in many crypto news categories, striving for the highest journalistic standards and abiding by a strict set of editorial policies. If you are interested to offer your expertise or contribute to our news website, feel free to contact us at [email protected]

PlatoAi. Web3 Reimagined. Data Inteligence Amplifed.
Click here for Free Trial.
Source: https://www.dcforecasts.com/ethereum-news/eth-mining-pool-sparkpool-to-shut-down-amid-chinas-pressure/

Blockchain

Virgil Griffith Pleads Guilty Conspiracy Charges Over His North Korean Trip

Published

9 hours ago

September 28, 2021

Republished by Plato

Virgil Griffith pleads guilty to charges of conspiracy over his trip to North Korea for a blockchain conference a few years ago as we reported in our latest crypto news today.

Ethereum developer Virgil Griffith pleads guilty to charges in an agreement with a federal prosecutor and his trial will start today in New York. Griffth was arrested in 2019 after he went to North Korea’s capital Pyongyang and gave a talk at a blockchain conference there. Now, he is facing a maximum sentence of 20 years in prison over the conspiracy to violate the Emergency Economy Powers Act and he took the plea deal that will see him face about six years in prison.

Virgil Griffith’s trial starts today. The government wants to put him in prison for the alleged crime of speaking freely to the wrong audience.

Trials like this are where the law is made, where we find out how strong our civil rights really are. I stand with Virgil.#FreeVirgil

— Jake Chervinsky (@jchervinsky) September 27, 2021

The US Attorney’s Office for the Southern District of New York argued that Griffith violated the international emergency economic powers act which prohibits exporting goods, services, or technology to North Koreans and in a statement, the US Attorney Geoffrey S. Berman alleged that Griffith provided technical information to North Korea, knowing that the information will help them to launder money and to evade sanctions, as Berman explained:

“Griffith jeopardized the sanctions that both Congress and the president have enacted to place maximum pressure on North Korea’s dangerous regime.”

I refuse to take the convenient path of throwing Virgil under the bus, because I firmly believe that that would be wrong. I’m signing. Reasoning below.https://t.co/E44p5caeJO

— vitalik.eth (@VitalikButerin) December 1, 2019

Griffith also entered the country in 2019 without approval from the US, despite being warned about traveling to North Korea as the criminal complaint alleged. In their turn, Griffith’s lawyers argued that the information that was provided didn’t qualify as a service and can be obtained by anyone on the internet which is why he claims his speech was protected by the First Amendment. Since his arrest, the crypto community split on the subject with some arguing that the actions he took were ill-advised. Ethereum’s co-founder Vitalik Buterin was among those that voiced their support for Griffith in 2019 as he tweeted at that time:

“I refuse to take the convenient path of throwing Virgil under the bus, because I firmly believe that that would be wrong.”

In the meantime, a petition to release Griffith from prison generated only 516 signatures in two years. As previously reported, After being denied bail, Griffht was granted a bond order for $1 million at the end of December 2019. The US Attorney’s Office, however, says Virgil Griffith violated bail after he tried to access his crypto account on Coinbase via a third party- his mother. He had been granted bail on the condition that he won’t access his accounts and will remain under house arrest with his parents in Alabama.

PlatoAi. Web3 Reimagined. Data Inteligence Amplifed.
Click here for Free Trial.
Source: https://www.dcforecasts.com/ethereum-news/virgil-griffith-pleads-guilty-conspiracy-charges-over-his-north-korean-trip/

Blockchain

Motivational Movies for Those in Business

Published

12 hours ago

September 28, 2021

Republished by Plato

Running a business will demand a lot from you. It will be no walk in the park. Even after doing everything right, you may still not be successful. From the sleepless nights to the financial sacrifices to the massive losses. Nothing can prepare you for what you will encounter. Therefore it is worth getting inspiration from those who would have tried it out before and see how they overcame certain obstacles.

The two movies that we are going to list below will really change the way that you think about business. Yes, we understand that right now, you may not see your business growing big because of the various hurdles you are facing. However, going on and staying true to yourself and your business will see it staying afloat. While we are at it, we would like to mention that the internet is the future of business. Therefore, you might want to consider moving your business online. Try to copy how online casinos do it and you will be surprised by the results.

The Founder, 2016

McDonald’s is known worldwide and have had their name written in the history books. It is no surprise there is a movie telling of how McDonald’s came to be. You will see how a willingness to succeed, hard work and commitment will in most cases lead to success. Your willpower and drive is the most critical in establishing whether you win or don’t. McDonald’s success story is sure to inspire you

The Social Network, 2010

Everyone with a Facebook account is sure to be familiar with this success story. It is about Mark Zuckerberg who is the founder of Facebook. Through this movie, we learn of his success story. As an entrepreneur, you will learn that starting your own business does not mean that you will achieve instant success and that you need to work hard in order to have your goals. The movie will also teach you that you will probably be met with opposition and how to use that opposition to your advantage. Did you know that you win with crazyvegas online casino if you follow them on their social network to see the completion on their page.

PlatoAi. Web3 Reimagined. Data Intelligence Amplified.

Click here to access.

Source: https://nulltx.com/motivational-movies-for-those-in-business/