Hi, I'm Imran...Welcome to my portfolio !
"I am a passionate and dedicated machine learning and deep learning professional with a strong background in developing and implementing cutting-edge AI solutions. With 3 years of experience, I have developed a deep understanding of neural networks, computer vision, and natural language processing. Throughout my career, I have worked on a variety of projects and have honed my skills in developing high-quality, scalable and efficient machine learning models"
In this portfolio, you will find a showcase of my work. I am confident that my skills and expertise in the field of deep learning and machine learning make me a valuable asset to any team, and I look forward to contributing my skills to new and challenging projects in the future.
Now I'm working as a freelancer for UpWork as Machen Learning Developer.Publications Talks Twitter Github
My expertise can be used to clean, analyse, interpret and visualize your data.
Knowledge of popular deep learning frameworks such as TensorFlow, Keras, PyTorch, and Caffe.+
Knowledge of NLP techniques such as text classification, sentiment analysis, and language translation.+
Knowledge of cloud computing platforms such as AWS, GCP, and Azure, and experience with deploying machine learning models on the cloud..+
From Data to Viz - A classification of chart types based on input data format
July 2018 - 6 minutes read
I’m delighted to announce a new dataviz project called Data to Viz. It is a classification of chart types based on input data format. It comes in the form of a decision tree leading to a set of potentially appropriate visualizations to represent the dataset. - Read more
Pimp my RMD - A collection of tips for R Markdown
July 2018 - 2 minutes read
R markdown creates interactive reports from R code. I’ve created a document that provides a few tips I use on a daily basis to improve the appearance of my html outputs (my memory aid). This document is built using R Markdown and hosted on Github. - Read more
A classification of charts based on their input format, and a lot of tips and tricks.Visit
Hundreds of R charts, always with reproducible code provided.Visit
A gallery of charts made with python, with a focus on Matplotlib and Seaborn.Visit
D3 is the ultimate tool for data visualization on the web. Tedious but limitless.Visit
Dashboard for Terres Inovia
A dashboard allowing to quickly access information in Terres Inovia database
The Como Project
Webpage describing a study of interconnection between mental disorders.
Published in JAMA Psychiatry
Visualize how all the co-authors of my previous supervisor are inter
A dataviz challenge on water pollution in France. Third price at the GreenTech Challenge.
Online tool to estimate standard error of allele frequency in pool-sequencing experiments. Published in Molecular Ecology
"[...] Yan did an absolutely brilliant job. He has multiple skills and qualities making him a great data science project member, and even leader : excellent knowledge and learning ability of diverse statistical and data management tools obviously, and a real knack for data viz."
"Yan always performed with outstanding skills and impressive capacity for adaptation and efficiency. His job was crucial for the publication of important results. [...] I thus highly recommend him for any work concering framing and analysis of huge data sets."
"Yan has taught me everything I know about data visualization!
He taught my group how to display data in clever ways, and has been responsible for our recent
interactive website related to psychiatric epidemiology.
Also, he is a super-fast and efficient bioinformatician. He has set up a pipeline to handle large genetic datasets and apply modern analytic methods to explore the genetic architecture of traits and disorders.
He is a super-bright guy – I can highly recommend him!"