PyTorch

Fundamental theory and practice in Data Science (DS). 🧮

Целью проекта является обучение нейросети на данных о погоде, и на основание этих данных предсказывать будущую температуру

Practical tasks on Deep Learning (DL) and Neural Networks (NN). 🤖

The goal of the project is to implement a simple generative image model. First, we study StyleGAN v1, and after examining its limitations, we move on to a simple DiT.

The goal of the project is to implement a DiT-B generative image model.

AI Radiology Copilot (ARC) is a CNN model app to recognize and label brain tumors in MRI scans.

CNN model for Visualizing CNN layers

AI-remove photobackground

Fine-tune wangchanBERTa + FastAPI + Docker

Vintage / Second-hand Marketplace Item Identifier & Price Comparator

Basic Idea: User takes photos of second-hand items (books, clothes, electronics) – especially relevant to students/people buying second-hand in Ireland. ML model recognises item type + approximate category + suggests price range (using scraped data from online marketplaces like DoneDeal, DoneDeal.ie, etc.). Backend stores user offers/requests; allows search, history, notifications for good deals.

Datasets / Data sources: For generic objects: use general vision datasets such as ImageNet – wide variety of everyday objects & categories. For clothes / fashion: Fashion‑MNIST for a start (though low-resolution, but good prototyping). For more specialized classes, get good coverage from public datasets like ImageNet.

对不同尺寸的交通信号灯图片做红黄绿分类

可以训练和使用根据语种的姓名分类和生成模型

Using a Residual Neural Network to predict coordinates from a single image from a Google Street View image within the United States

A clean, modular implementation of Physics-Informed Neural Networks (PINNs) for solving the 1D and 2D wave equation using PyTorch.

binary for .wav

Image classification is the “hello world” of deep learning.