Here’s a bit of what I do.
Discover how memory optimization techniques like weight quantization enable the deployment of powerful language models in resource-constrained environments. Explore the trade-offs between precision and memory efficiency with practical experiments, and learn how bfloat16 and int8 quantization impact model performance....
This project presents an analysis and clustering of COVID-19 data worldwide. The result is an hierarchical clustering with respect to active cases, active cases per population and active cases per population and per area.
The problem addressed in this research paper is a binary classification on a time series. The aim is to detect the emergence of collusion in Prisoner Dilemma games played by two Reinforcement Learning algorithms using a Random Forest Classifier....
This project automates the process of sending is connection invites and notes, fetching 1st level connections, sending messages to new connections and sending inmails to those who do not accept the invitation. The automation is done using selenium and linkedin-api. The responses data is then analyzed with NLP using pre-trained RoBERTa in the transformers library of Huggingface to understand which messages were more effective.