Learn for Docker Networking

You have been using docker for quite a bit of time and you have been connecting to other containers using ssh and the IP address of the container, which is fine really if this is a one-time thing but what if you have to connect to the container again and again.

Docker offers a solution to this problem through a network net, for those who are wondering about what this is. Consider it like a group where you can address machines with names.

#docker 

What is GEEK

Buddha Community

Docker Explained: Docker Architecture | Docker Registries

Following the second video about Docker basics, in this video, I explain Docker architecture and explain the different building blocks of the docker engine; docker client, API, Docker Daemon. I also explain what a docker registry is and I finish the video with a demo explaining and illustrating how to use Docker hub

In this video lesson you will learn:

• What is Docker Host
• What is Docker Engine
• Learn about Docker Architecture
• Learn about Docker client and Docker Daemon
• Docker Hub and Registries
• Simple demo to understand using images from registries

#docker #docker hub #docker host #docker engine #docker architecture #api

Docker Bridge Network

Hello readers,

This blog will tell you about the docker bridge network. Its use with some basic use cases also how to bridge networks different from the network.

One of the reasons Docker containers and services are so powerful is that you can connect them together, or connect them to non-Docker workloads.

Docker Network

Docker’s networking subsystem is pluggable, using drivers. Several drivers exist by default, and provide core networking functionality:

Type of docker Network :

• bridge: The default network driver. If you don’t specify a driver, this is the type of network you are creating. Bridge networks allow your applications to run in standalone containers that need to communicate.
• host: For standalone containers, remove network isolation between the container and the Docker host, and use the host’s networking directly. The host is available for swarm services on Docker 17.06 and higher.
• overlay: Overlay networks connect multiple Docker daemons together and enable swarm services to communicate with each other. You can also use overlay networks to facilitate communication between a swarm service and a standalone container, or between two standalone containers on different Docker daemons.
• macvlan: Macvlan networks allow you to assign a MAC address to a container, making it appear as a physical device on your network. The Docker daemon routes traffic to containers by their MAC addresses.
• none: For this container, disable all networking. Usually used in conjunction with a custom network driver.

Let create a bridge network with name alpine-net . Start two alpine container and ping between them to check network connectivity.

$ docker network create --driver bridge alpine-net
962f373dd1cc9b9d532287b898e56d51fe1e5cf09fe90208ccbf34e51ea4511c

#devops #docker #docker bridge network #docker network

Docker manifest - A peek into image's manifest.json files

docker manifest – An experimental feature !

The image manifest provides a configuration and a set of layers for a container image.

This is an experimental feature. To enable this feature in the Docker CLI, one can edit the config.json file found in ~/.docker/config.json like :

{
        "auths": {
                "https://index.docker.io/v1/": {
                        "auth": "XXXXXXX"
                }
        },
        "HttpHeaders": {
                "User-Agent": "Docker-Client/19.03.8 (linux)"
        },
        "experimental": "enabled",
        "debug": true
}

What is ‘docker manifest’ ?

The docker manifest command does not work independently to perform any action. In order to work with the docker manifest or manifest list, we use sub-commands along with it. This manifest sub-command can enable us to interact with the image manifests. Furthermore, it also gives information about the OS and the architecture, that a particular image was built for.

A single manifest comprises of information about an image, it’s size, the layers and digest.

A manifest list is a list of image layers (manifests) that are, created by specifying one or more image names. It can then be used in the same way as an image name in docker pull and docker run commands.

Commands to get started with :

After enabling this feature, one would be able to access the following command :

These commands are easy to use. It basically avoids the need for pulling and running and then testing the images locally, from a docker registry.

Next, to inspect an image manifest, follow this syntax,

 docker manifest inspect image-name

.

#devops #docker #devops #docker #docker learning #docker-image

Deep learning on graphs: successes, challenges, and next steps

TL;DR This is the first in a [series of posts] where I will discuss the evolution and future trends in the field of deep learning on graphs.

Deep learning on graphs, also known as Geometric deep learning (GDL) [1], Graph representation learning (GRL), or relational inductive biases [2], has recently become one of the hottest topics in machine learning. While early works on graph learning go back at least a decade [3] if not two [4], it is undoubtedly the past few years’ progress that has taken these methods from a niche into the spotlight of the ML community and even to the popular science press (with Quanta Magazine running a series of excellent articles on geometric deep learning for the study of manifolds, drug discovery, and protein science).

Graphs are powerful mathematical abstractions that can describe complex systems of relations and interactions in fields ranging from biology and high-energy physics to social science and economics. Since the amount of graph-structured data produced in some of these fields nowadays is enormous (prominent examples being social networks like Twitter and Facebook), it is very tempting to try to apply deep learning techniques that have been remarkably successful in other data-rich settings.

There are multiple flavours to graph learning problems that are largely application-dependent. One dichotomy is between node-wise and graph-wise problems, where in the former one tries to predict properties of individual nodes in the graph (e.g. identify malicious users in a social network), while in the latter one tries to make a prediction about the entire graph (e.g. predict solubility of a molecule). Furthermore, like in traditional ML problems, we can distinguish between supervised and unsupervised (or self-supervised) settings, as well as transductive and inductive problems.

Similarly to convolutional neural networks used in image analysis and computer vision, the key to efficient learning on graphs is designing local operations with shared weights that do message passing [5] between every node and its neighbours. A major difference compared to classical deep neural networks dealing with grid-structured data is that on graphs such operations are permutation-invariant, i.e. independent of the order of neighbour nodes, as there is usually no canonical way of ordering them.

Despite their promise and a series of success stories of graph representation learning (among which I can selfishly list the [acquisition by Twitter] of the graph-based fake news detection startup Fabula AI I have founded together with my students), we have not witnessed so far anything close to the smashing success convolutional networks have had in computer vision. In the following, I will try to outline my views on the possible reasons and how the field could progress in the next few years.

**Standardised benchmarks **like ImageNet were surely one of the key success factors of deep learning in computer vision, with some [6] even arguing that data was more important than algorithms for the deep learning revolution. We have nothing similar to ImageNet in scale and complexity in the graph learning community yet. The [Open Graph Benchmark] launched in 2019 is perhaps the first attempt toward this goal trying to introduce challenging graph learning tasks on interesting real-world graph-structured datasets. One of the hurdles is that tech companies producing diverse and rich graphs from their users’ activity are reluctant to share these data due to concerns over privacy laws such as GDPR. A notable exception is Twitter that made a dataset of 160 million tweets with corresponding user engagement graphs available to the research community under certain privacy-preserving restrictions as part of the [RecSys Challenge]. I hope that many companies will follow suit in the future.

**Software libraries **available in the public domain played a paramount role in “democratising” deep learning and making it a popular tool. If until recently, graph learning implementations were primarily a collection of poorly written and scarcely tested code, nowadays there are libraries such as [PyTorch Geometric] or [Deep Graph Library (DGL)] that are professionally written and maintained with the help of industry sponsorship. It is not uncommon to see an implementation of a new graph deep learning architecture weeks after it appears on arxiv.

Scalability is one of the key factors limiting industrial applications that often need to deal with very large graphs (think of Twitter social network with hundreds of millions of nodes and billions of edges) and low latency constraints. The academic research community has until recently almost ignored this aspect, with many models described in the literature completely inadequate for large-scale settings. Furthermore, graphics hardware (GPU), whose happy marriage with classical deep learning architectures was one of the primary forces driving their mutual success, is not necessarily the best fit for graph-structured data. In the long run, we might need specialised hardware for graphs [7].

**Dynamic graphs **are another aspect that is scarcely addressed in the literature. While graphs are a common way of modelling complex systems, such an abstraction is often too simplistic as real-world systems are dynamic and evolve in time. Sometimes it is the temporal behaviour that provides crucial insights about the system. Despite some recent progress, designing graph neural network models capable of efficiently dealing with continuous-time graphs represented as a stream of node- or edge-wise events is still an open research question.

#deep-learning #representation-learning #network-science #graph-neural-networks #geometric-deep-learning #deep learning

Neural Networks Intuitions: 9. Distance Metric Learning

Welcome back to my series _Neural Networks Intuitions. _In this ninth segment, we will be looking into deep distance metric learning, the motivation behind using it, wide range of methods proposed and its applications.

Note: All techniques discussed in this article comes under Deep Metric Learning (DML) i.e distance metric learning using neural networks.

---

Distance Metric Learning:

Distance Metric Learning means learning a distance in a low dimensional space which is consistent with the notion of semantic similarity. (as given in [No Fuss Distance Metric Learning using Proxies])

What does the above statement mean w.r.t image domain?

It means learning a distance in a low dimensional space(non-input space) such that similar images in the input space result in similar representation(low distance) and dissimilar images result in varied representation(high distance).

Okay, this sounds exactly what a classifier does. Isn’t it? Yes.

So how is this different from supervised image classification? Why different terminology?

Metric learning addresses the problem of open-set setup in machine learning i.e generalize to new examples at test time.

This is not possible by a feature-extractor followed by fully connected layer Classification network.

Why?

This is a very important question. The answer is as follows:

1. A classifier learns**class-specific features and not necessarily generic features.**
2. A classifier with a standard cross entropy loss maximizes inter-class distances such that the features before FC layer are linearly separable.

#metric-learning #deep-learning #siamese-networks #triplet-loss #representation-learning #deep learning