This repository accommodates local LLMs for Agent Laboratory, with which you can let AI agents backed up by local LLMs conduct academic research either autonomously or with human intervention.

AI Scientist backed up by local LLMs

My master research implementation titled 'Multi-Agent Simulated Environments Generated by a Transformer-Based Generative Model'.

A study repo for data structures and algorithms in Python.

programs for graphics using PyOpenGL and PyGame

The repository is to analyze my pitching velocity while training to increase my fastball pitching velocity.

TensorFlow 2 Implementation of "U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation".

TensorFlow 2 implementation of "Deep Learning for NLP and Speech Recognition" written by Uday Kamath, John Liu, and James Whitaker.

Mario played by Reinforcement Learning and implemented with TensorFlow 2.

TensorFlow 2 Implementation of "GANimation: Anatomically-aware Facial Animation from a Single Image".

TensorFlow 2 Implementation of "StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation".

Analyze various GAN algorithms.

TensorFlow Implementation of FOTS, Fast Oriented Text Spotting with a Unified Network.

Japanese-English Transformer translator implemented with TensorFlow 2.

This repo is to gain some knowledge of TensorFlow 1.x and 2.x.

This module can detect human faces and analyze its' sex, race and age.

To predict a pitch velocity from a video.

Python implementation of a book called "How Program Works" written by Hisao Yazawa.

To investigate various DNN text classifiers including MLP, CNN, RNN, BERT approaches.

This module is a part of the app, which I was in charge, and created at a hackathon "【IoTLT初！】AIoTLTハッカソン - データを集めてAIを作ってみよう". The event page is below. https://iotlt.connpass.com/event/103011/

This project includes some codes that are useful to understand PRML, which is a book officially named 'PATTERN RECOGNITION AND MACHINE LEARNING' by Chris Bishop

This repo is to try various object tracking methods.

Try various NMT(Neural Machine Translation) methods such as seq2seq or attention and see the results.

Web application of Word2Vec