Glimpse Of Intel AI & HPC Summit : Application’s Of Artificial Intelligence – Part 3

Glimpse Of Intel AI & HPC Summit : Application’s Of Artificial Intelligence – Part 3

Read my previous article Part 2

Intel demonstrated some of the great applications of Artificial Intelligence where Intel’s hardwares and softwares out performing to solve complex problems in realtime using AI. Intel is working with many customers to solve their complex problems. 

Application’s Of Artificial Intelligence

1. Tumour detection in large 3-D medical images

Identifying tumour from huge medical images is yet another complex problems, today we have algorithms which does this job better than doctors.

2. Disease Risk Prediction Using Genetic Variation Data

3. Classification and predictions of protein sequences

4. Intelligent Knowledge Management 

This is another interesting application where, the algorithms analyse the huge set of documents and help the management to answer for complex queries.

5. Oil Rig Inspector Assist System

This is a very complex problem, identifying corrosion on nut-bolts under deep sea water is a complex problem solve for oil mining industries. Currently it is a manual process, where video cameras are fixed in oil rig systems and several hundred people will manually go through thousands of frames of videos manually and identify the corrosion on nut-bolts.

Now Intel is working with such industries to automate the process by analysing the video frames. thousands of video frames will be analysed in just few hours. The algorithms are capable of identifying the corrosion with great precision.

6. Self driving Cars and Car automation

BMW CARASSO is the system based on ai, the company is using to collect data from several sensors to debug their hard systems and to achieve the quality in their product. Company is using this system only for their internal quality improvements and to understand the roads and behaviour of the cars.

Intel has lot of open source code base which can be used for many such applications, kindly see the source code of Intel’s Nervana in Github

simple steps towards saving water

simple steps towards saving water

I am not going to tell again Water is precious, We have to save water ! I am not going to tell again Importance of water and why we have to save water ! Because since my childhood and ofcourse you too, listening all these and thats nothing new for us.

I am writing this article just to share about a simple experiment we did at our home and a simple analysis based on the data we collected. Now it is upto the reader whether to follow it or not because you know the importance of water and you all know that what you are doing is wrong and you are leaving nothing to your kids.

Few months back we stopped using washing machine and we started using a small hand held machine to wash our cloths. one fine day I realised something very important, that is when we were using washing machine we were wasting lot of water, because all these modern washing machines will consume too much water and we do not re-use that water. Since we started using small hand held machine these days we realised how much water we are saving using this machine and we started re-using the water after washing the cloths.

after doing it for few days  we thought to collect the data so that we can do some simple calculations and do some analysis on it. The analysis gave us an astonishing result, so thought to share with all of you.

Here is what we did :

1.we stopped using washing machine which consumes too much amount of water

water-consuming-washing-machine
water-consuming-washing-machine

2. we started using small hand held washing machine which consumes very minimum amount of water

hand-held-small-washing-machine
hand-held-small-washing-machine

3. finally re-using the water after washing the cloths

re-using-water
re-using-water

These are the simple steps we followed to save enough water. How much water we can save if we all together join hands ? Let us come to the analysis part.

Some data we collected for the period of 30 days

So the above table shows how much buckets of water we are using for washing the cloths and re-using the same water. total for around 30 days we re-used 264 liters of water.

Now considering certain data which helps us to do some calculations 

  1. Average amount of water consumed by any modern washing machine = 170.3 liters per load
  2. Average amount of water used by flush in toilets = 4 liters
  3. Some important data regarding Bengaluru water supply – 

    Bengaluru Water supply
    Bengaluru Water supply

Keeping above data for reference we did some simple calculations as follows

Converting all the data for a month.

170.3 ltr per load in washing machine -> So per month even if we use 10 times = 1700 liters ( assuming in a family of 3-4 people we use atleast 10 times in a month)

4 liters per flush and if we use 6 times per day then 24 liters per day and 744 liters per month ( assuming we use toilet flush at least 6 times in a day )

So total per month both washing machine and toilet flush together it consumes 2500 liters of water

Now here we are just using 264 liters of water for waching cloths and we are re-using the water after washing cloths which means we are able to manage both washing cloths and toilet with just 264 liters of water.

That means almost 90% of total consumption is optimised, means 2300 liters water is saved or reused per month.

2300 liters of water saved per month 

Let us calculate it for 12 months -> 2300 x 12 = 27600 liters

Now considering another stats 8.5 million people served daily in Banglore 

If all 8.5 million people save water as suggested above, we can save 8.5 * 2300 = 19550 ML 

Another stats given is 1350MLD water served in Bengaluru daily

so now

total_water_saved_by_8.5M_people / water_served_per_day_in_Bengaluru

19550ML/1350ML = 14.48

which means if all 8.5M people together save 19550ML of water then we can serve these 8.5M people for extra 14.48 days.

 

Android Gradle : multiDexEnabled property

Android Gradle : multiDexEnabled property

Android Build Error

Error:Execution failed for task ‘:app:transformClassesWithDexForDebug’.
> com.android.build.api.transform.TransformException: com.android.ide.common.process.ProcessException: java.util.concurrent.ExecutionException: com.android.dex.DexIndexOverflowException: method ID not in [0, 0xffff]: 65536

Solution: Add the following multiDex property in your build.gradle.
multiDexEnabled true
so your build.gradle looks like follows

 

apply plugin: 'com.android.application'

android {
    compileSdkVersion 25
    buildToolsVersion "25.0.1"
    defaultConfig {
        applicationId "dhi_ti.com.repositoryapp"
        minSdkVersion 11
        targetSdkVersion 25
        versionCode 1
        versionName "1.0"
        testInstrumentationRunner "android.support.test.runner.AndroidJUnitRunner"
        multiDexEnabled true
    }
    buildTypes {
        release {
            minifyEnabled false
            proguardFiles getDefaultProguardFile('proguard-android.txt'), 'proguard-rules.pro'
        }
    }
}

dependencies {
    compile fileTree(dir: 'libs', include: ['*.jar'])
    androidTestCompile('com.android.support.test.espresso:espresso-core:2.2.2', {
        exclude group: 'com.android.support', module: 'support-annotations'
    })
    compile 'com.android.support:appcompat-v7:25.1.1'
    testCompile 'junit:junit:4.12'
    compile project(path: ':dhilib')
}

What is  multiDexEnabled true ? Why it is required ?

As per the documentation from Google, As the Android platform has continued to grow, so has the size of Android apps. When your app and the libraries it references reach a certain size, you encounter build errors that indicate your app has reached a limit of the Android app build architecture.

About the 64K reference limit

(APK) files contain executable bytecode files in the form of Dalvik Executable (DEX) files, which contain the compiled code used to run your app. 

The Dalvik Executable specification limits the total number of methods that can be referenced within a single DEX file to 65,536—including Android framework methods, library methods, and methods in your own code. In the context of computer science, the term Kilo, K, denotes 1024 (or 2^10). Because 65,536 is equal to 64 X 1024, this limit is referred to as the ’64K reference limit’.

Reference : https://developer.android.com/studio/build/multidex.html#about

 

Sanskrita Vyakarana : separation, distinction,analysis, explanation

Sanskrita Vyakarana : separation, distinction,analysis, explanation

Sanskrit is called as a perfect language, an unambiguous language and hence NASA preferred it for Artificial Intelligence. But why so ? How it is different from any other language in the world? Why it is called as a most scientific language? What is so unique in sanskrit language ?

To answer all the above questions, we have to really look into the vyakarana in sanskrit. Though I don’t have enough knowledge in sanskrit, I am writing these articles after exploring a little bit about sanskrit with the help of few online courses of sanskrit ( Vyoma Linguistic Labs : Learn Sanskrit From home )

Vyakarana in Sanskrit:

In english if I say Read it is a just a word and you can not give any more explanation for that, is there any logic behind formation of this word ? Is there any explanation for the word Read‘ or ‘Reading ? Definitely we don’t have any answer for this question.

In sanskrit there is nothing which has no explanation, there is a logic behind formation of each and every word. Let me take simple example

  1. There is a word पठति in sanskrit, which means Read but as I said above for every word there is an explanation, let us see how this word formed

पठति  = पठ्‌ + + ति 

पठ्‌ – धातुः [अर्थं निर्दिशति]

अ – विकरणप्रत्ययः [गणं निर्दिशति]

ति – तिङ्‌प्रत्ययः [लकारं निर्दिशति] |

2. भ्वादिः (प्रथमः गणः = भू इत्यादयः धातवः; “भवतिइत्यादीनि क्रियापदानि) | यस्य गणस्य आदौ भू-धातुः अस्ति, सः भ्वादिगणः |

विकरण प्रत्ययः शप्‌ → अ

उदाखाद्‌ (खादति), पठ्‌ (पठति), क्रीड्‌ (क्रीडति), वद्‌ (वदति); शुच्‌ (शोचति), वृत्‌ (वर्तते); जि (जयति), स्रु (स्रवति), हृ (हरति)

Hope these simple examples help us to understand how rich the sanskrit language is. All I am trying to tell here is learn sanskrit, we have plenty of online courses, videos, mobile applications but I prefer the following pointers because these mobile apps for sanskrit are really very nice.

Online tutorials For Learning Sanskrit

Learning Sanskrit From Home

Sanskrit Learning Online Courses : http://www.sanskritfromhome.in/

Sanskrit Learning YouTube Videos : https://www.youtube.com/channel/UCp5mvCwXR-drTRyzNUUjdZg

Sanskrit Learning Android Mobile Applications : https://play.google.com/store/apps/developer?id=Vyoma_labs&hl=en

For CBSC Students : www.cbsesanskrit.com

Read More about Sanskrit : Interesting Facts About Sanskrit