Skip to content
View dkwik's full-sized avatar

Block or report dkwik

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse

Pinned Loading

  1. rag-knowledge-chatbot-django rag-knowledge-chatbot-django Public

    Knowledge chatbot using Agentic Retrieval Augmented Generation (RAG) techniques. Full-stack proof of concept built on langchain, llama-index, django, pgvector, with multiple advanced RAG techniques…

    Jupyter Notebook 30 7

  2. nixtla-forecasting-at-scale nixtla-forecasting-at-scale Public

    Robust forecast that allows for the creation of hundreds of time-series forecasts, with analysis of performance at different combinations of granularity, time-horizon, and algorithms.

    Jupyter Notebook

  3. xgboost-shap-business-analysis xgboost-shap-business-analysis Public

    Trained multivariate model with XGboost + shapley values to make blackbox model explainable. Analysis used for account management and c-suite analysis. Shapley value was wrangled to become useful a…

    Jupyter Notebook

  4. digital-garden digital-garden Public

    My personal 'digital garden' where I allow myself to "think with the garage door open". I use this as a canvas for my inner thoughts, working on lateral thinking through the dense linkage of ideas.

    JavaScript

  5. streamlit_app streamlit_app Public

    A streamlit UI prototype with a form submission, a table pulled from Google Cloud Platform's BigQuery, and other interactive features. This was quick prototyping work to get familiar with streamlit…

    Python

  6. AI-Image-Classification-Neural-Network AI-Image-Classification-Neural-Network Public

    Fine-tuning a facial emotion classifier from scratch with fast.ai. Uses VGG16 architecture, training on 20 epochs with a learning rate of 0.002, achieving a 70.24% accuracy.

    Jupyter Notebook 1