I'm currently specializing on building performance focused Deep Learning models using C++ over Python.
- MLNotify Contributor (300+ stars): No need to keep checking your training - just one import line and you'll know the second it's done. [Code]
- A Hitchhiker's Guide to Lending Club Loan Data (100+ upvotes) [Kaggle Notebook]
- Statistical Sidewalk for Exploratory Data Analysis: A Kaggle notebook sharing statistical analysis techniques written in Python [Kaggle Notebook]
- PiHard: Mathematics in Python ℼ( ◡̀_◡́)ᕤ [Code]
- MNIST DCGAN using C++ [Code] [Article]
- NLP using Deep Learning [Code]
- NLTK Breakdown: A breakdown of the Natural Language Toolkit (NLTK) library in Python [Code]
- PyTorch Refreshers: Implementation of various Machine Learning and Deep Learning algorithms in PyTorch [Code]
- Machine Learning Refreshers: Implementation of various Machine Learning algorithms along with their nuances [Code]
- Segment Anything Model (SAM) Breakdown: A study of the Segment Anything Model (SAM) by Meta AI [Code]
- Uncovering variants of GLU for improving Transformers [Article]
- Gated Linear Unit — Enabling stacked convolutions to out-perform RNNs [Article]
- Python Configuration Management using Hydra by Meta [Article]
- Fixed Positional Embeddings using Sinusoidal Embeddings [Article]
- Adaptive Embedding for Transformer XL [Article]
- Position-wise Feed Forward Network For Positional Embeddings In Transformers [Article]
- CLIP by OpenAI Explained - A concise explanation on how the CLIP model is pre-trained and used for zero-shot image classification [Article]
- Transformer Models by Google Brain Explained With PyTorch Implementation [Article]
You can send me a mail at pragyansubedi@gmail.com or send me a message on LinkedIn.