Shengwei-Peng
An Electrical Engineering Graduate Student at National Taiwan University of Science and Technology
National Taiwan University of Science and TechnologyTaipei City, Taiwan
Pinned Repositories
Chinese-Extractive-Question-Answering
A project for Chinese Extractive Question Answering (QA), focusing on selecting relevant text spans from a given context to answer specific questions. It supports a two-step process for paragraph and span selection or an end-to-end solution using fine-tuned models like BERT.
Chinese-Multimodal-Sentiment-Analysis
Chinese-Multimodal-Sentiment-Analysis is an innovative project that integrates advanced deep learning techniques to analyze sentiments in Chinese language data. Utilizing a sophisticated blend of textual, audio, and visual modalities, this project offers a comprehensive approach to understanding and interpreting complex emotional expressions.
DataDriven-IGBT-Inverter-Modeling
This repository contains the code and dataset for the paper "Data-Driven Modeling of Three-Phase IGBT Two-Level Inverter for Electrical Drives using ANNs". The project employs MATLAB to develop and train Artificial Neural Networks (ANNs) to model a three-phase IGBT two-level inverter for electrical drives.
InceptionResNet-SSVEP-BCI
Implementation of InceptionResNet for SSVEP-based Brain-Computer Interface (BCI)
LLM-based-Finance-Agent
An intelligent agent utilizing Large Language Models (LLMs) for automated financial news retrieval and stock price prediction.
PV-Power-Generation-Forecasting
A project focused on forecasting solar photovoltaic (PV) power generation using regional microclimate data. Implements machine learning models like CatBoost, LightGBM, and XGBoost for predictions, leveraging environmental features like temperature, humidity, wind speed, and solar radiation.
Speech-Emotion-Recognition
A Speech Emotion Recognition project using CREMA, RAVDESS, SAVEE, and TESS datasets. Implements machine learning techniques to classify emotions from speech.
Stock-Price-Prediction
A repository for predicting stock prices using machine learning techniques. Includes data preprocessing, model training, evaluation, and visualization.
TOCFL-MultiBench
TOCFL-MultiBench: A multimodal benchmark for evaluating Chinese language proficiency using text, audio, and visual data with deep learning. Features Selective Token Constraint Mechanism (STCM) for enhanced decoding stability.
UAV-GenerativeAI-Navigation-Images
Generative AI for UAV Navigation in Natural Environments. This project utilizes GAN and Diffusion models to generate realistic images of roads and rivers from UAV perspectives.
Shengwei-Peng's Repositories
Shengwei-Peng/DataDriven-IGBT-Inverter-Modeling
This repository contains the code and dataset for the paper "Data-Driven Modeling of Three-Phase IGBT Two-Level Inverter for Electrical Drives using ANNs". The project employs MATLAB to develop and train Artificial Neural Networks (ANNs) to model a three-phase IGBT two-level inverter for electrical drives.
Shengwei-Peng/Chinese-Multimodal-Sentiment-Analysis
Chinese-Multimodal-Sentiment-Analysis is an innovative project that integrates advanced deep learning techniques to analyze sentiments in Chinese language data. Utilizing a sophisticated blend of textual, audio, and visual modalities, this project offers a comprehensive approach to understanding and interpreting complex emotional expressions.
Shengwei-Peng/LLM-based-Finance-Agent
An intelligent agent utilizing Large Language Models (LLMs) for automated financial news retrieval and stock price prediction.
Shengwei-Peng/PV-Power-Generation-Forecasting
A project focused on forecasting solar photovoltaic (PV) power generation using regional microclimate data. Implements machine learning models like CatBoost, LightGBM, and XGBoost for predictions, leveraging environmental features like temperature, humidity, wind speed, and solar radiation.
Shengwei-Peng/UAV-GenerativeAI-Navigation-Images
Generative AI for UAV Navigation in Natural Environments. This project utilizes GAN and Diffusion models to generate realistic images of roads and rivers from UAV perspectives.
Shengwei-Peng/TOCFL-MultiBench
TOCFL-MultiBench: A multimodal benchmark for evaluating Chinese language proficiency using text, audio, and visual data with deep learning. Features Selective Token Constraint Mechanism (STCM) for enhanced decoding stability.
Shengwei-Peng/CNC-Predictive-Maintenance
A predictive maintenance system for CNC machines using machine learning algorithms to forecast failures and optimize maintenance schedules.
Shengwei-Peng/InceptionResNet-SSVEP-BCI
Implementation of InceptionResNet for SSVEP-based Brain-Computer Interface (BCI)
Shengwei-Peng/Speech-Emotion-Recognition
A Speech Emotion Recognition project using CREMA, RAVDESS, SAVEE, and TESS datasets. Implements machine learning techniques to classify emotions from speech.
Shengwei-Peng/Stock-Price-Prediction
A repository for predicting stock prices using machine learning techniques. Includes data preprocessing, model training, evaluation, and visualization.
Shengwei-Peng/Chinese-Extractive-Question-Answering
A project for Chinese Extractive Question Answering (QA), focusing on selecting relevant text spans from a given context to answer specific questions. It supports a two-step process for paragraph and span selection or an end-to-end solution using fine-tuned models like BERT.
Shengwei-Peng/Chinese-News-Summarization
A project for Chinese news summarization using state-of-the-art pre-trained models like mT5 and GPT-2. This tool facilitates abstractive summarization, offering flexible training and inference workflows, and includes advanced decoding strategies for improved summarization quality.
Shengwei-Peng/Classical-Chinese-Translation
A project for bidirectional translation between Classical Chinese and modern languages. Utilizing advanced techniques like Low-Rank Adaptation (LoRA) and Parameter-Efficient Fine-Tuning (PEFT), it enables seamless translation of ancient texts while optimizing computational efficiency.
Shengwei-Peng/House-Price-Prediction
AI-based house price prediction. Utilizing machine learning to accurately estimate real estate values based on key influencing factors.
Shengwei-Peng/Shengwei-Peng