Explore live open-source projects and AI models.

Randomly partitions time series segments into train, development, and test sets; Trains multiple models optimizing parameters for development set, final cross-validation in test set; Calculates model’s annualized return, improvement from buy/hold, percent profitable trades, profit factor, max dr... Read more

The official implementation of ICLR2025 paper for shift-invariant neural networks

Time series analysis on NIFTY-50 index stock data. The project visualises trends in stock data and explores various time-series analysis models such as ARIMA, ARMA and deep learning models such as LTSM and RNN to predict stock prices with the highest accuracy.

Codebase for the Paper "Deep Semi-supervised Learning (SSL) for Time Series Classification (TSC)" to appear at the ICMLA '21

An implementation of the DeepAnT model, a deep learning approach for unsupervised anomaly detection in time series data, using Python.

Longsword Stance RESTful Service Working demo: https://www.youtube.com/watch?v=v7hvOyPQ0EM Longsword Stance RESTful Service: Invokes prediction results with real-time multivariate time series data. Using Flask and python, the pre-trained bidirectional LSTM deep learning model is loaded to memory... Read more

Analysis of Air Pollution prediction and time-series forecast of PM2.5 Pollutant using Machine Learning Algorithms (SVM, Decision Tree and Random Forest) and Deep Learning Algorithms (CNN, Bi-LSTM). Also considered for improved performances is random search hyper-parameter tuning using Ray-Tune w... Read more