Afshin Karimi's repositories
Fine-tuning-an-LLM-using-LoRA
📚 Text Classification with LoRA (Low-Rank Adaptation) of Language Models - Efficiently fine-tune large language models for text classification tasks using the Stanford Sentiment Treebank (SST-2) dataset and the LoRA technique.
LungTumor-Segmentation
Automatically segment lung cancer in CTs
3D-Liver-Segmentation
Automatically segment the liver and liver tumors in CT scans with 3D-UNET
Chat-Bot-using-Streamlit-and-OpenAI
This repository hosts a user-friendly chatbot application, powered by the GPT-3.5 model and built with the Streamlit framework. Engage in dynamic conversations, ask questions, and enjoy interactive responses with ease. Customize and extend its capabilities to suit your specific requirements.
Image2AudioStoryConverter
Convert images into captivating audio stories using image-to-text, language models, and text-to-speech technologies. Upload images, generate stories, and listen to the narrated audio clips. A fun project combining AI and creativity.
Atrium-Segmentation
Automatically segment the left Atrium in cardiac images
Cardiac-Detection
Predict a bounding box around the heart in X-ray images.
FaceEmotionRecognition
Build a Face Emotion Recognition (FER) Algorithm
Pneumonia-Classification
Identify size of pneumonia in X-Ray images
Machine-Learning-App-Deployment-on-Amazon-EKS-with-MLOps-Practices
This repository demonstrates the deployment of a machine learning app on Amazon EKS using Docker and Kubernetes, with the incorporation of MLOps practices such as version control, CI/CD, and infrastructure as code.
Similar-Paper-Reccomendation
This repository contains an application designed to recommend scientific papers that are most similar to a given input paragraph. The application uses the llama and weaviate libraries to achieve this.
train_scheduling_assistant
This project utilizes a fine-tuned Large Language Model (LLM) to generate train scheduling information from unstructured textual data, providing an interactive UI via Streamlit.
ReAct-Framework-Implementation
This repository implements the "ReAct" framework, combining OpenAI's GPT-3.5 Turbo model and LangChain for dynamic reasoning and interaction with external data sources.
GAN-digits
Implement GAN and DCGAN (Deep Learning Course Project)
Financial-Risk-Assessment-using-LightGBM
This project is a demonstration of a Financial Risk Assessment application that predicts risk tolerance based on user-provided financial data. It combines machine learning models (Logistic Regression and LightGBM), FastAPI for the backend, and Streamlit for the frontend to create an interactive web application.
Heart-disease-prediction-using-SVM
In this project, Support Vector Machines (SVM) algorithm implemented that determines which patient is in danger and which is not.
Keypoint-Description-and-Matching
Advance computer vision course project (Fall 2020)
localGPT
Chat with your documents on your local device using GPT models. No data leaves your device and 100% private.
matching-Algorithms
Advance computer vision course project (Fall 2020)
Predict_house_prices_KR
Predict house prices with kernel regression
ResNet_Pytorch
Implement ResNet architecture and test on CIFAR10 dataset
Sentiment-Analysis
Machine Learning Course Final Project (Sharif University of Technology)
XGboost-on-Iris
Clustering on Iris dataset using XGBoost and Gradient Boost