Build logistic regression, neural network models for classification

[BMVC'23 Oral] Offical repository of "Rethinking Transfer Learning for Medical Image Classification"

Source code for the numerical experiments presented in the paper "Greedy Shallow Networks: An Approach for Constructing and Training Neural Networks".

In recent times, toxicological classification of chemical compounds is considered to be a grand challenge for pharma-ceutical and environment regulators. Advancement in machine learning techniques enabled efficient toxicity predic-tion pipelines. Random forests (RF), support vector machines (SVM) and deep neural networks (DNN) are often ap-plied to model the toxic effects of chemical compounds. However, complexity-accuracy tradeoff still needs to be ac-counted in order to improve the efficiency and commercial deployment of these methods. In this study, we implement a hybrid framework consists of a shallow neural network and a decision classifier for toxicity prediction of chemicals that interrupt nuclear receptor (NR) and stress response (SR) signaling pathways. A model based on proposed hybrid framework is trained on Tox21 data using 2D chemical descriptors that are less multifarious in nature and easy to calcu-late. Our method achieved the highest accuracy of 0.847 AUC (area under the curve) using a shallow neural network with only one hidden layer consisted of 10 neurons. Furthermore, our hybrid model enabled us to elucidate the inter-pretation of most important descriptors responsible for NR and SR toxicity.

Libreria didattica per la creazione, addestramento e test di reti neurali fino a tre strati in linguaggio C

[CVPR 2026] Ultra-Low Bitrate Perceptual Image Compression with Shallow Encoder

[NeurReps 2024, TMLR 2025] Can Kernel Methods Explain How the Data Affects Neural Collapse?

Predicting if a mushroom is edible or poisonous with a shallow neural network with Keras and TensorFlow 2.

Notebooks of programming assignments of Neural Networks and Deep Learning course of deeplearning.ai on coursera in August-2019

This is a project that is derived from my capstone project targeted for cleaning of EEG signals and feature extraction followed by a custom made neural network and analysis of performance by the Network mainly for classifying the valence of an emotion.

In this project, we propose a cervical cancer detection and classification system using CNNs . We employ transfer learning and fine-tuning for enhanced performance. Classifiers like ELM and AE are added to increase the efficiency.

Deep learning Specialization on Coursera

Logistic Regression Implementations - ML, Shallow NN and Enhanced Deep Neural Network for Structured and Unstructured Data Classification

Human Data Analytics (Optional Project)

A shallow CNN model that is trained on X-ray chest images with preprocessing step of adaptive histogram equalization.

High-throughput detection and enumeration of tumor cells in blood using Digital Holographic Microscopy (DHM) and Deep Learning.

study of scene classification with different MLP layer types

Comparative Analysis of Activation Functions in Shallow Neural Networks for Multi-Class Image Classification Using MNIST Digits and CIFAR-10 Datasets with Fixed Architectural Parameters

Challenge of shallow neural network approximation with one-dimensional input.

Car Price Prediction is a machine learning project aimed at developing a model that can predict the selling price of used cars based on various features or attributes.

AI & Cybersecurity laboratory activity reports for the AI & Cybersecurity course of Politecnico di Torino

Design of an one hidden layer neural network using numpy only,

Deploy Deep Learning model for classifying hoax news with Flask

This project encompasses a range of neural and non-neural model implementations to classifiy MNIST digits. The goal is to compare the performance of each technique including details of hyper-parameters, training ans testing errors, training and testing duration and additional parameters used in the analysis.

Implementation of Deep Neural Networks

Image classification using neural networks. Completed for school. Part 1 is a classical 80s style shallow network, and part 2 is a more modern network. For part 2, I added activation functions, implemented L2 Regularization, changed network depth and width, and used Convolutional Neural Nets to improve performance. Check README

This is a classifier for classifying the planar data with one hidden layer.

Ising models and kinetic Ising models applied to calcium imaging recordings from mouse area M2.

Implementation of DNN with Early Stopping from scratch in Python. Evaluation was done on two simple datasets (Blobs and Moons) and on one more challenging dataset (Fashion-MNIST).

Using a shallow neural network in Retinopathy of Prematurity (ROP) image enhancement