Aryan Singh's World

Data Science, Statistics, ML, Deep Learning


Kabhi toh aana tu bhi mere saffe par alfaaz ki tarah,
Door ghataon mei se chhanti roshni ki parwaaz ki tarah,
Meri duaon ka kaboolnama hai tera muskurana,
Uski pakeezgee hai khoobsurat mere kalaam ki tarah.

Nazron ka jhuk jana riwayton ke lihaz ki tarah,
Kuchh eshsaas se, kuchh jazbaat se ankahi keh jana humraaz ki tarah,
Pehle pehar ki namaaz si masoom hai teri aankhen,
Jaise dekhte hi aagosh mei sama jana, ulfat ke aagaz ki tarah.

Meri khamoshion ko awaaz de jana, meri soch ko aagaz de jana,
Kabhi guzro kisi roz mere darwaze se,
Toh chhukar meri kalam ko likhne ka andaaz de jaana,
Zindagi jeene ka ik bahana dhoondhti hai ye aankhen, inhe zindadili ka ik haseen paigaam de jaana.


Paas nahi ho tum, toh bhi yahin kahin ho tum.

Sab keh gayi ye aankhein, toh bhi ik baat ankahi ho tum.

Zindagi hai ik registaan si banjar, toh bhi pehli baarish ki boondo mei shamil nami haseen ho tum.

Jahaan har raasta hai ek aur paheli ki tarah, vahan kuchh aasan silsilon ki ik dor sarsari ho tum.

Jahaan kuchh aur nahi maut ke sanaate ke siva, vahan zindagi se sarabor ik khushfahmi ho tum.

Jahan aakar jhuk jaana dastoor hai iss zamaane ka, uss dargah ki dehleez ka pak noor ho tum.

Jo chhipta hai jaakar saagar ki baahon mei, uss suraj ki lalima ka guroor ho tum.

Jahan jaade se thithurta hai voh nanha sa baalak, vahan ik daraar se aati hui dhoop ki garmi ho tum.

Mai hun jis makaam ka pyaasa zamaane bhar se, uss makaam ki nawazish mei maqbool ho tum.

Kya dekhna uss husn ka joh itraata hai har pal, meri rooh ke aaine mei jo kaid hai voh tasveer ho tum.

Mai khud ko dhoondho yan kho jaane dun teri yaadon mei, uss khone mei sab kuchh paa jaane ka sukoon ho tum.


Sometimes a dish is not perfect until it is sprinkled with a bit of spice. Similarly a sprinkle of gratitude and small gestures of love make the dish of life tastier than ever.

Memories have a life of their own,
Some get conceived at the moments of utter joy,
Others are sad like a little still born boy,
Yet immortal, yet indelible post the initial ahoy.

Swiping someone off their feet,
With unannounced arrivals and rhapsodic beat.
Tears of joy with a soothing hug,
A gift out of nowhere is hard to shrug.

Candlelit evenings in the arms of a lake,
A filling of love in the surprise cake.
A soup with a smile for the healing touch,
Long drives through the rain when there’s nothing much.

Because life is a flow of emotional streams,
Merging at a juncture in the end of the one act scenes.
So why not create a piggy bank of smiles,
Open it up to cherish the memory lane through the last mile.

The Orange Moon

The Sun struggles for the whole day to meet back with the horizon once it rises out of it. Whereas an orange moon occurs when the Moon is very close to the horizon and they(sun, moon, horizon) finally meet. Similarly sometimes we struggle in life until we think of someone far away with whom we will meet when our descend towards the horizon begins.

Several feet away yet i can hear your heartbeat,
Incense across space and time like the sweetest smell of orchid leaves,
I feel that smile on my silly jokes,
Serene as the dampness of the misty oaks.

Boundaries of light years yet your voice brightens my life,
Giving wings to my heart’s wishes like a million fireflies,
The sounds of my songs in the karaoke bars,
You are the asteroid of my universe brighter than a million stars.

On my run across the garden you’re the stead behind my step,
The chats with the old baker about your favourite crepe,
Remember you by with each whiff of air I breathe,
Hugging the cushions, and your fragrance in my sheets.

When I see the sky my love reflects even more,
As beautiful as your eyes on seeing me at the door,
I am the sun’s hope of meeting the horizon’s gaze,
You are the orange moon shining back my longing rays.

Killing Them Softly

I walk into the world of virulent wolves with nothing but mischief under my sleeve,
What to others is a tumultuous ravine, for me its a premonition to redeem.

They surround, they slander, laying cobwebs for the adversary to concede
Little do they know of the forlorn games I won, neither of the weapon that I conceal.

To their fire I am the ice, to their darkness I am the morning sunrise,
Their constant shenanigans are just fuel to my patience pyre.

Mad I am from in to out but method to the madness is what’s on my priors,
No matter wolves are growling to scare, insecurities gallore in their vain stare.

I am ready for the battle with the ominous device,
Some call it chortle, others laughter, some chuckle, while others smile.

It is this what puts their minds through the turnstile,
Why go hard when killing them softly is so worthwhile.


Sochtan hun ki kaash voh bhi kabhi mere baare mei soche

Ittefaq se hi sahi, kabhi mera bhi khayaal voh apne khwabon mei khoje

Jo band hai khidkian kai roz se,

Voh karne ko aaj aankhon ki guftugu, kuchh lamho ke liye khole.


Kuch aisa tha mausam ka bebakpan, kuchh aisi zindadili teri muskan mei thi
Ki mai toh nikla tha bhar ke zehen mei kuch sawal
Ab toh jawabon ka silsila bhi tujhe janne ka bahana ho gaya.

Kyun kajal tera karta hai abhi bhi mujhse baatein
Kyun adaa teri se mera taroof baar baar ho jaata hai
Ki mai toh baitha tha alfaazon ke bagaan se kuch phool chunne
Teri soch mei ye agaaz bhi nazm likhne ka ik bahana ho gaya.


Aaj fir shaakh par kuchh naye ghonsle dikhee..

Lagta hai kuchh aur parindon ne uchaayion se samjhauta karna seekh liya..

Tensorflow 2.0 in 2 minutes

Tensorflow 2.0-alpha was released a couple of days ago with a bunch of exciting features. It can be installed by following command:

pip install -U –pre tensorflow

In this post I explore the 17 most key features among them. The purpose is to make this short, crisp but touch on all the major pointers.

  1. Improvement in tf.keras high level API: Tensorflow 2.0 takes the compatibility between imperative Keras and DAG driven tensorflow to next level by adopting tf.keras to its core. This will make prototyping and productionizing deep learning models fast. Also, this will engage and bring more developers towards deep learning, keras being more intuitive.
  2. Eager execution by default: No need to create an interactive session to execute the graph. TF 2.0 introduces the eager execution by default moving all the session related boilerplate under the hood.
  3. Compact and improved documentation: TF 2.0 offers better organized documentation. Most of it is available here:
  4. Clarity: 2.0 takes clarity to the next level by removing various duplicate functionalities like multiple versions of GRU and LSTM cells available. 2.0 takes care of choosing the optimum node according to hardware giving developer a single unified library to choose from for instance one implementation of LSTM and one for GRU.
  5. Low Level API: Full low level API in tf.raw_ops with inheritable interfaces for variables, checkpoints and layers to define your own components.
  6. Easy Up-gradation: Conversion script in tf_upgrade_v2 to convert TF 1.0 code into TF 2.0 code automatically. Just write: !tf_upgrade_v2 –infile <input_file> –outfile <output_file>
  7. Backward compatibility: Comes with a separate backward compatibility module tf.compat.v1 for getting the older components.
  8. One optimizer, one losses module, one layers module: Unified optimizers module in tf.keras.optimizer.*. Similarly one losses and one layers module under tf.keras.losses.* and tf.keras.layers.*.
  9. Better graphical visualization: 2.0 gives better graph visualizations in Tensorboard even for keras models.
  10. Easy to distribute: Provides more options for scaling and multi GPU training via tf.distribute.Strategy module. strategy = tf.distribute.MirroredStrategy()
    with strategy.scope():
    <define the model here>
  11. Save and Import keras model: Easy to save and load the keras models by using tf.keras.experimental.*.
  12. Run Keras On TPUs: 2.0 comes with tf.distribute.experimental.TPUStrategy() that will allow the keras code to run on TPUs.
  13. New datasets available: 2.0 comes with new datasets to test the models on in vision, audio and text domain.
  14. More pre-trained models at TF Hub: More pre-trained models from the world of NLP and Vision available at Tensorflow Hub.
  15. Improved error reporting: Improved error reporting with exact line number and full call stack.
  16. TFFederated: TF federated to support federated learning on edge devices.
  17. Swift support and Fast AI: 2.0 to come with a Swift library. Jeremy Howard will be delivering a course on the same.

Source: Tensorflow Dev Summit 2019

Camaraderie Of Silence

In this poem I talk about the times of silence where I sit alone and think about life, goals and the problems that keep me awake at night.

Zero, Vaccum, tranquility, hush are the diamonds in the life’s rough,
Waging lonely battles against life’s frivolous trouphs.
In harrowing night where becons a tryst with the longing to wither the time,
The power of silence rallies the intellect to stifle this crime.

Alone I sit waiting for the train of thoughts,
Sometimes its musings somedays its an aweful lot.
The vine of challenges encircles to thicken the plot,
Then I draw the sword of logic to cut through these knots.

The time that makes the philosophers think, the time that changes the rulers to kings.
It’s that time of the day which solitude brings,
Where silence is the companion in the troubled brinks.

Things go quieter as the times go past,
Mind aches for a companion to share the feel of outcast.
The fire of melancholy burns the cauldron of heart,
Enraging the blood to add fuel for a quest miles apart.

Now I have made friends with the lonesome days,
Now i hold hands with the quiet nights.
For its the time of the day that gives wings to the thought’s flight,
My comate my crony, towards the ultimate light.