Basic Gesture Recognition Using mmWave Sensor - TI AWR1642

Intrusion Detection System - IDS example using Dense, Conv1d and Lstm layers in Keras / TensorFlow

MPS like shaders for audio processing. Conv1d, Spectrogram.

cross-platform modular neural network inference library, small and efficient

Griffin: Mixing Gated Linear Recurrences with Local Attention for Efficient Language Models

Augmenting character level Transformers with Causal Dilated Conv1D layers

Cross-platform mobile Neural network C library for training and inference on the device. CPU only. It fits for time-series data.

VAITP - Vulnerability Attack and Injection Tool for Python

Implemented a Research paper "PubMed 200k RCT: a Dataset for Sequential Sentence Classification in Medical Abstracts. Use Hybrid Embedding (char + token + positional).

GAN for Spectral Data Augmentation

Goal is to detect peak in given signal(s) in directory \dataset with computation intelligence and not just signal processing. Here, we have used 1D CNN from Keras to achieve it.

Donors Choose

predict price of bitcoin using time series

A project on Topic Modelling and Text Summarization of NIPS research papers

This project predicts sunspot activity using an LSTM model for time series data. Built with TensorFlow and Keras, it uses Huber loss for outlier handling and MAE for performance evaluation. The dataset, sourced from Kaggle or SIDC, spans over 270 years of monthly sunspot data.

A Comprehensive Analysis Framework for Material Thickness Estimation Using TOF Data

Anomaly detection with Dense, LSTM, and Conv1D autoencoders — reconstruction error, adaptive thresholds, ROC/PR analysis, and error heatmaps.

📈 Conv1D + LSTM forecasting with synthetic data, feature engineering, and trend alignment. Full deep learning pipeline.

This project explores the application of advanced neural network architectures, including Multi-Layer Perceptron (MLP), Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU), to forecast traffic volume. 👍👍✅

Identification of Credit Card Fraud Utilizing Hybrid Deep Learning Models with Improved Precision and Minimized False Positives