Min Htet Myet (Mattral)'s repositories
Paraphraser-App
Analyze and Paraphrase Texts
Predict-Bike-Sharing-Demand-with-AutoGluon
Predicting Bike Sharing Demand with AutoGluon
summary-widget
Integrate your CodersRank profile summary to your personal website
ML-AI-Algorithms-from-scratch
implementation of several Supervised, Unsupervised, Bayesian, Neural Networks and Reinforcement Learning Algorithms from scratch with only numpy and basic libs.
Assembly-Testing
Assembly Language for simple operations
Lung-Disease-Detection-from-Medical-Images
Deployed models. Takes X-Rays or CT scans for accurate disease detection, includes several architectures for each disease.
HandWritten-Text-Recognizer
Streamlit Web Interface for Handwritten Text Recognition (HTR), Optical Character Recognition (OCR) implemented with TensorFlow and trained on the IAM off-line HTR dataset. The model takes images of single words or text lines (multiple words) as input and outputs the recognized text.
NOTE-Best-Practices-for-Computer-Vision
Best Practices for Computer Vision
YouTube-Video-Summarizer-Using-Whisper-and-LangChain
YouTube Video Summarizer Using Whisper and LangChain
Markdown-Notes
Notes for writing markdown format (for writing ReadMe.md files)
Blockchain-and-IPFS-Video-Upload-Simulation
local demo simulation designed to mimic the process of uploading video files to IPFS and interacting with a smart contract for storing and retrieving IPFS hashes. This simulation is intended for demonstration purposes, to showcase the concept without requiring actual blockchain or IPFS network interactions.
LLM-Improving-Trained-Models-with-RLHF
Experimented with the three essential Reinforcement Learning with Human Feedback (RLHF) process stages. It starts by revisiting the Supervised Fine-Tuning (SFT) process, then proceeds with the training of a reward model, and finally concludes with the reinforcement learning phase. We explored and applied methods such as 4-bit quantization and LoRA
AI-Road-Inspection
Omdena Mexico Project
Fine-Tuning-using-LoRA-and-SFT
Lets dive deeper into the mechanics of LoRA, a powerful method for optimizing the fine-tuning process of Large Language Models, its practical uses in various fine-tuning tasks, and the open-source resources that simplify its implementation.
CareNetDetect
demo Web application for CareNet Disease Detection
ActivityRecognition-ExerciseMannerPrediction
Using devices such as Jawbone Up, Nike FuelBand, and Fitbit it is now possible to collect a large amount of data about personal activity relatively inexpensively. The goal of this project is to predict the manner in which they did the exercise.
RoboticsAcademy
Learn Robotics with JdeRobot
Recreating-the-Bing-Chatbot
explore the idea of finding the best articles from the Internet as the context for a chatbot to find the correct answer. We will use LangChain’s integration with Google Search API and the Newspaper library to extract the stories from search results. This is followed by choosing and using the most relevant options in the prompt.
Using-the-Open-Source-GPT4All-Model-Locally
GPT4All is trained on top of Facebook’s LLaMA model. The main contribution of GPT4All models is the ability to run them on a CPU. Testing these models is practically free because the recent PCs have powerful Central Processing Units.
Google-Gemma-Experiments
Gemma LLM experiments
QA-Chatbot-over-Documents-with-Sources
building a Question Answering (QA) Chatbot that works over documents and provides sources of information for its answers. Our QA Chatbot uses a chain (specifically, the RetrievalQAWithSourcesChain), and leverages it to sift through a collection of documents, extracting relevant information to answer queries.
Cpp-Web-Assembly-Credit-Card-Validator
C++ Web assembly project with Emscripten.
LLAMA2-with-Replicate
This chatbot is created using the open-source Llama 2 LLM model from Meta. Particularly, we're using the Llama2-7B model deployed by the Andreessen Horowitz (a16z) team and hosted on the Replicate platform. This app was refactored from a16z's implementation of their LLaMA2 Chatbot to be light-weight for deployment to the Streamlit Community Cloud
Supercharge-Your-Blog-Posts-Automatically-with-LangChain-and-Google-Search
leveraging Google search results to enrich the prompt to the model by incorporating additional information. The demonstration showcased the utilization of embedding vectors to identify content that shares a similar meaning or context—also the process of adding relevant information to a prompt to achieve better output.
Alzheimers-Disease-Detection-from-Brain-Scan-Images
Analyzing Brain Scan Images for the Early Detection and Diagnosis of Alzheimer's Disease+Parkinson disease
Guarding-Against-Undesirable-Outputs-with-the-Self-Critique-Chain
LLM can occasionally generate undesirable outputs. A couple of well-known examples of this behaviour are harmful or hallucinating content. It is important to employ a mechanism to make sure the model’s responses are appropriate in the production environment.