ptdeco is a library for model optimization by decomposition built on top of PyTorch

a collection of computer vision projects&tools. 计算机视觉方向项目和工具集合。

This repository offers a robust solution for multilabel image classification. Utilizing advanced neural networks like VGG16, VGG19, ResNet50, InceptionV3, DenseNet121, and MobileNetV2, the project achieves precise classification across 107 diverse categories.

Learn linear quantization techniques using the Quanto library and downcasting methods with the Transformers library to compress and optimize generative AI models effectively.

Analyzed customer churn using transaction data. Built ML model to predict lapses. Dataset includes customer status, collection/redemption info, and program tenure. Delivered business presentation outlining modeling approach, findings, and churn reduction strategies.

This repository contains code and resources for a project focused on predicting traffic volume using Temporal Convolutional Networks (TCNs). Leveraging the Metro Interstate Traffic Volume dataset from 2012-2018, the project aims to develop an accurate model for short- to medium-term traffic volume forecasting in Minneapolis-St Paul, MN.

Develop a tool in Google Colab using machine learning and neural networks to select applicants for funding with the best chance of success based on the source data provided by the organization.

本仓库包含了完整的深度学习应用开发流程，以经典的手写字符识别为例，基于LeNet网络构建。推理部分使用torch、onnxruntime以及openvino框架💖

Benchmarking bank data to enhance marketing strategies. Models: Decision Tree and Random Forest. Libraries: Pandas, Matplotlib, Seaborn, Scikit-Learn, Numpy. Findings: Customer patterns and seasonal behaviors.

"Vitis-AI-YOLOv3-TF2-Quantization-Evaluation" Repo for quantization of YOLOv3 on Vitis-AI using TF2, aimed to deploy model on edge devices with limited resources. Includes training & quantization scripts and evaluation metrics. Experiment with different configurations.

Practical experience in hyperparameter tuning techniques using the Keras Tuner library. Hyperparameter tuning plays a crucial role in optimizing machine learning models, and this project offers hands-on learning opportunities. Exploring different hyperparameter tuning methods, including random search, grid search, and Bayesian optimization

Some DNN model optimization experiments and notebooks

Affordable GNN using Topological Contraction

Determing the eligibility for granting home loan. ML classification models are used, in order to predict if loans are apporoved or not, based on customers's data.

TOP13% solution for the Titanic-Kaggle competition using a Gradient Boosting Classifier. Moreover, implementation of a Streamlit App to play with the models.

Aprendizagem e Extração de Conhecimento

Vision-lanugage model example code.

Model Optimization using Batch Normalization and Dropout Techniques

Nonprofit foundation Alphabet Soup wants a tool that can help it select the applicants for funding with the best chance of success in their ventures. Using machine learning and neural networks, you’ll use the features in the provided dataset to create a binary classifier that can predict whether applicants will be successful if funded.

Machine Learning: model optimization through hyperparameters