Data Science Portfolio

Welcome to My Data Science World!

All machine learning notebook & MLOps Projects:

🦆 Kaggle:

https://www.kaggle.com/kritsadakruapat

🐶 DagsHub:

https://dagshub.com/kkowenn

🤗 HuggingFace:

https://huggingface.co/kritsadaK

Machine learning and Deep learning Projects:

Project 1: Machine learning without any library:

https://github.com/kkowenn/ManualMachineLearning

Project 2: Manual algorithm cosin similarity:

https://www.kaggle.com/code/kritsadakruapat/cosin

Project 3: The Feature Importance score from a decision tree model:

https://github.com/kkowenn/FastworkAnalysisProject

Project 4: Optimizing Performance Metrics to find the best possible combination of hyperparameters to achieve the highest performance. (before tuning vs after tuning)

https://github.com/kkowenn/DataSciencePortfilo/blob/main/Schooltask(WineQuality)/Wine_Quality.ipynb

Project 5: Comparing Regression Models(Linear Regression model with Random Forest Regression, XGBoost Regression and Neural Network) + Exploratory Data Analysis(EDA)

https://www.kaggle.com/code/kritsadakruapat/comparing-regression-models

Project 6: Fraud detection; tuning and interpreting imbalanced data by using SMOTE, RandomizedSearchCV, SHAP

https://www.kaggle.com/code/kritsadakruapat/imbalanced-tuning-interpretability/notebook

java script plot :

https://github.com/kkowenn/DataSciencePortfolio/blob/main/Imbalanced%2BTuning%2BInterpretability/modeltuning%2BInterpretability.ipynb

Project 7: Applying model on web, draw digits on a canvas and classify them using a pre-trained convolutional neural network (CNN) model by using the MNIST dataset.

https://github.com/kkowenn/DigitRecognitionWeb

Project 8: Transfer Learning with ResNet-50 & MobileNetV2 and etc.

https://www.kaggle.com/code/kritsadakruapat/simpletransferlearning-resnet50-unsatisfied

https://www.kaggle.com/code/kritsadakruapat/mobilenetv2-image-classification

Project 9: RNN model (Vanilla RNN, LSTM, GRU, and Bidirectional RNN) converted text to numbers, and tried different model types (RNN) to see which performed best.

https://www.kaggle.com/code/kritsadakruapat/rnnmodelcomparison?scriptVersionId=182686891

Project 10: gans-on-themnist-dataset

https://www.kaggle.com/code/kritsadakruapat/gans-on-themnist-dataset

Computer Vision Projects:

Project 1: Data Augmentation to Fix Overfitting and improve model:

https://www.kaggle.com/code/kritsadakruapat/cnnchangedatacolor-notjustred

https://github.com/kkowenn/AugmentationForDlibModel

Project 2: using mediapose to detect arm to overlay tattoo filter:

https://huggingface.co/spaces/kritsadaK/TattooPoseOverlay

https://github.com/kkowenn/ComputerVisionProject

Project 3: simple yolov (object detection)

https://github.com/kkowenn/SimpleYolov

Project 4: fall detection capture when detect fall with canny edge detection and skeleton and remove the background

https://github.com/kkowenn/Fall-Detection

Natural Language Processing Projects:

Project 1: fundamental Natural Language Processing

https://github.com/kkowenn/basic_nlp

Project 2: Data Augmentation to Fix Overfitting (nlp)

https://github.com/kkowenn/Political-Fake-News-Detector-NLP.git

Project 3: Thai word recommender system by using cosine similarty

https://huggingface.co/spaces/kritsadaK/ThaiSentenceSimilarityApp

Project4: Quantized LoRA-Tuned Llama

https://github.com/kkowenn/DataSciencePortfolio/blob/main/QuantizedLoRATunedLlama/FlexTuneLLM.ipynb

https://huggingface.co/kritsadaK/UltraInteract-Llama-FT

Project5: A LINE chatbot using Retrieval-Augmented Generation (RAG) with Pinecone and OpenAI GPT for intelligent, real-time responses by python

https://github.com/kkowenn/LINE-RAG-AI-Assistant

MLflow Projects:

Project 1: AirFlowProject1: Weather Data ETL Pipeline Using Docker, Docker Compose, Airflow, PostgreSQL, DagsHub, and Python Libraries:

fetch_data_task → process_data_task → upload_data_task

https://dagshub.com/kkowenn/AirFlowProject1

Project 2: Deployment model on Amazon Bucket(s3):

Machine learning model -> Mlflow -> Amazon Web Services

https://dagshub.com/dashboard

https://dagshub.com/kkowenn/DeploymentOnAWSProject1.mlflow/#/experiments/0?searchFilter=&orderByKey=attributes.start_time&orderByAsc=false&startTime=ALL&lifecycleFilter=Active&modelVersionFilter=All+Runs&datasetsFilter=W10%3D

Project 3: Dogecoin Minutely Prediction:

Data Collection (API binance) & Preprocessing -> Model Training & Experiment Tracking (MLFlow) -> Version Control & Pipeline Management (DVC)

https://dagshub.com/kkowenn/End-to-endDogecoinMinutelyPrediction

Project 4: Thailand PM10 Prediction App (stream lit track Log experiments by mlflow )

url(Open Government Data of Thailand) -> mini ETL -> basic ARIMA model -> streamlit display & choose location to predict-> Experiment Tracking -> Version Control & Pipeline Management (DVC)

https://dagshub.com/kkowenn/OpendatathaiMLflow

https://huggingface.co/spaces/kritsadaK/ThailandPM10PredictionApp2022

Big data analysis:

Scala Project: Scala workspace designed for data analysis with a simple multi-root, Spark SQL and visualization, featuring sub-projects for SQL-based analysis, various plot examples, and including essential configurations

https://github.com/kkowenn/Basic_Scala_ForBigData

Hadoop Project: A Hadoop MapReduce application for efficiently counting word occurrences in a text file, comprising a driver, a mapper, and a reducer

https://github.com/kkowenn/Simple-Hadoop

R Project: Exploratory data analysis (EDA) for the dataset in R lanaguge. The main goal is to analyze factors influencing student performance by exploring data distributions, identifying patterns, and examining relationships among variables.

https://www.kaggle.com/code/kritsadakruapat/eda-forrlanguage

SparkMLib Project: Personalized Beer SVD Recommendations Using Spark MLlib

https://www.kaggle.com/code/kritsadakruapat/svdbyusingsparkmllib

ETC:

Visualization tool :

PowerBI: https://github.com/kkowenn/MasterPowerBI

looker studio: https://lookerstudio.google.com/reporting/287f8ccf-c993-4185-bcdd-db1e06d7fc36

Web Scraping with Python Project:

https://github.com/kkowenn/MyScrapingProject