DEV Community: Abhilash Majumder

Adobe ML-2 Engineer Interview Experience

Abhilash Majumder — Sun, 23 Apr 2023 07:17:45 +0000

This is regarding a set of interviews I appeared for Adobe a while back for their ML-2 Engineer (Deep Learning role). It consisted of 4 rounds in totality (each being a technical with the last round exclusively being HR- round 5). I have prepared some videos to give an insight on how to tackle deep learning questions, model architectures, solve statistical problems and apply framework knowledge (including dl algorithmic knowledge ).

Round 1: https://www.youtube.com/watch?v=52kFFMkc7A0&t=1s
Round 2&3: https://www.youtube.com/watch?v=9AcGSPpCzhE&t=1630s

Further rounds and more such interview (related to deep learning) will be posted on the channel. Stay tuned.

deeplearninginterview #mlengineer2 #GAN #datastructures

Cartoonize your images with CartoonGAN

Abhilash Majumder — Thu, 16 Dec 2021 08:04:13 +0000

Cartoonizing images is now easier with CartoonGAN : https://huggingface.co/spaces/abhilash1910/CartoonGAN
Based on CVPR 2018 (http://openaccess.thecvf.com/content_cvpr_2018/papers/Chen_CartoonGAN_Generative_Adversarial_CVPR_2018_paper.pdf)
Youtube : https://youtu.be/PY7XJydYFaA

Leetcode Solutions to Graph Algorithm Problems

Abhilash Majumder — Mon, 30 Aug 2021 07:42:05 +0000

Leetcode contains some of the most popular interview questions asked across organizations. This youtube playlist contains solutions of some of the most important graph algorithms which are asked in interviews and present in Leetcode:
https://www.youtube.com/watch?v=w9wv2jlF3jY&list=PLovuuDh4TFdCJ00I378H-REodw6AWygN1
Do subscribe if found helpful!

My EuroPython '21 talk on Quantum Deep Learning

Abhilash Majumder — Thu, 19 Aug 2021 16:58:26 +0000

My talk at #EuroPython2021 on "Introduction to Quantum Deep Learning" which revolved around parameterized quantum classical circuits and how these help to solve some of the most challenging problems in Deep learning is at the link: https://www.youtube.com/watch?v=V92trw5bK34
Repository: https://github.com/abhilash1910/EuroPython-21-QuantumDeepLearning

Spectral Embeddings Library for Knowledge Graph Embeddings

Abhilash Majumder — Wed, 18 Aug 2021 15:40:36 +0000

SpectralEmbeddings (https://pypi.org/project/SpectralEmbeddings/0.2) is a python package which is used to generate node embeddings from Knowledge graphs using GCN kernels and Graph Autoencoders. Variations include VanillaGCN,ChebyshevGCN and Spline GCN along with SDNE based Graph Autoencoder.
There are broadly 2 categories for Knowledge graph embeddings through this package:

Graph AutoEncoders: This models the first and higher order similarity measures in a graph for each node in a neighborhood. The first and second order similarity measures are created through an Autoencoder circuit which preserves the proximity loss of similarity with reconstruction loss. This model has been implemented along the lines of SDNE . These embeddings not only cover the first order dependencies but also are used to capture second order dependencies between node neighbors. The output of this AutoEncoder network has a dimension of (number of input entries,dimension of embedding space provided).

Graph Convolution Kernels: These embeddings are based on spectral graph convolution kernels which capture node representations through laplacian norm matrices. This part is based on the Graph Convolution Network paper. The GCNs are based on deep neural networks which operate on the node features and the normalized laplacian of the adjacency matrix of input graph. The GCNs are mainly used for node/subgraph classification tasks but here we are interested in capturing only the embeddings from the penultimate layer of the network. For this we create an Embedding based on Tensorflow as node features. We define that the nodes that don’t have predecessors are in layer 0. The embeddings of these nodes are just their features. To calculate the embeddings of layer k we weight the average embeddings of layer k-1 and put it into an activation function. In this kernel there are 3 variations : Vanilla GCN, Chebyshev GCN and Spline GCN embeddings

The repository is at the link: https://github.com/abhilash1910/SpectralEmbeddings/
Colab: https://www.kaggle.com/abhilash1910/spectralembeddings?scriptVersionId=72130689