Machine Learning π€
I can work will all kinds of datasets. Structured datasets including tabular data, like CSV, Excel, or databases like SQLite, or unstructured including images, sounds, or even 3D datasets. Text too!
Techniques
- Classification, Regression, Time Series Analysis
- Supervised & Unsupervised
- Hyperparamter Tuning & Cross Validation
- Data Cleaning & Preprocessing
- Feature Selection & Engineering
- Ensemble Learning & Stacking
Tools
- Sklearn, TPOT- ML Models
- Numpy, Pandas - Data Wrangling
- Jupyter Notebook - Data Exploration
- Maptlotlib, Seaborn, Plotly - Data visualizations
Classification
- Logistic Regression
- K-Nearest Neighbors (K-NN)
- Support Vector Machine (SVM)
- Kernel SVM
- Naive Bayes
- Decision Tree Classification
- Random Forest Classification
Clustering
- K-Means Clustering
- Hierarchical Clustering
Regression
- Simple Linear Regression
- Multiple Linear Regression
- Polynomial Regression
- Support Vector Regression (SVR)
- Decision Tree Regression
- Random Forest Regression
- XGBoost
Dimensionality Reduction
- Principal Component Analysis (PCA)
- TSNE
- Variational autoencoders (Deep Learning)
Deep Learning π
This is my superpower. I have worked a LOT with Deep Learning, neural networks, and stuffs like that. I have used a lot of different tools & techniques, simply because there is no single winner, like for ex. Tensorflow is great, but if you want to convert your idea into working things as fast as possible, with great performance, FastAI is my best bet!
Techniques:
- Image classification, Regression & Transfer Learning. Using Keras sequential/function API
- Ensemble Learning & Stacking, multiple input & outputs model
- Advance Hyperparameter Tuning, Bayesian Search
- Efficient Data Pipelines using
tf.data
& training using Multiple GPUs & TPUs.
- Showing Final Results, making Reports.
- Object detection & Image segmentation including architectures like UNET
- GAN, CycleGAN etc.
Tools
- Tensorflow, Keras, FastAI v2, Pytorch, Detectron2 - For Deep Learning & Computer Vision
- Hugging face, tokenizer, ktrain - Natural Language Processing
- OpenCV - Image Preprocessing
- Weights & biases, Tensorboard - For Experimentation
- FastAPI, Flask, Streamlit, Docker - For production
- Git, Github, Github Actions - Version Control & CI/CD
- AWS, Heroku, GCP - Deploying Services
My Environment π
I use Anaconda for managing different environments and sometimes use virtual environments. Jupyter notebook/lab for data science stuff and VSCode for making python scripts.
Web Development π±
- HTML5, CSS3, Javascript (ES6, ES7, ES8, ES9 )
- Bootstrap, flexbox
- Npm, Node JS. and Express JS.
- React JS
Game Development π³
- Unity
Programming Languages π»
- Python - Machine Learning, Deep Learning, Data Science
- Javascript - Web Development
- C# and C++ - Game Development & IOT, Robotics
Courses I have taken π
- Machine Learning A-Z : Hands-On Python & R In Data Science βSuper Data Science
- Deep Learning A-Z: Hands-On Artificial Neural Networks βSuper Data Science
- Deep Learning and Computer Vision A-Z: OpenCV, SSD & GANs βSuper Data Science
- Learning to learn: Efficient Learning β Zero to mastery
- Complete Python Developer in 2020 β Zero to mastery
- Complete Web Developer in 2020 β Zero to mastery
- Complete Machine Learning and Data Science β Zero to mastery
- Machine Learning β Andrew Ng
- Deep Learning Specialization β Andrew Ng
- DeepLearning.AI TensorFlow Developer Professional Certificate β Laurence Moroney
- Deep Reinforcement Learning β Udacity
Some wonderful series πΊ
- The essence of linear algebra β 3blue1brown
- Essence of calculus β 3blue1brown
- Neural networks β 3blue1brown
- The Age of AI β Youtube Originals
Books I am reading/read π
- Life 3.0: Being Human in the Age of Artificial Intelligence β Max Tegmark
- Approaching (Almost) Any Machine Learning Problem β Abhishek Thakur
- Reinforcement Learning: An Introduction ByΒ Richard S. Sutton,Β Andrew G. BartoΒ Β· 2018
- Structures: Or Why Things Don't Fall Down β J. E. Gordon
Other interesting courses π©
( not fully done )
- Practical Deep Learning for Coders β FastAI
- Deep Reinforcement Learning 2.0 β Super Data Science
- Modern NLP in python β Super Data Science
- Master Coding Interviews: Data Structures & Algorithms β Zero To Mastery
Hobbies π
- Robotics
- Science
- Rocketry/Model Rocketry & Astronomy
- AI & Programming
- Football
Planning to learn π«
- Deep Reinforcement learning ( more! )
- Robotics & IOT
- Creating synthetic image dataset
- Mobile App development
βWhat companies I would love to contribute/work and how I can contribute my best to them
Deep Learning and especially areas of computer vision are my superpowers. Haveing done many data science & kaggle competitions, i always try to do which includes some sort of images, classification, segmentation,detection all that stuff, being done web development courses also, i have quit xp in deploying ml model into production, I always first try to put an idea into an working things as fast as possible and then iterating to improve it!
Also I love robotics, companied which are mix of computer vision & robotics are 100000% welcome!