Enter photometric light curve data — flux and time — and the model will determine the number of planets in a star's orbit.
Each line: flux time
lightkurve downloads Kepler/TESS light curves. Normalization, NaN removal, noise filtering.
Brightness and time normalization to the range [0, 1]. Saving prepared data to learn*.txt.
Rust + tch-rs trains a recurrent network. Auto-saving the best model to model.ot by MSE.
Light curve input → LSTM memory layers → Linear hidden→1 → number of planets. Under 10 ms.
The primary language — maximum speed and memory safety without GC.
PyTorch C++ API — LSTM, tensor operations, automatic differentiation.
Loading and processing Kepler/TESS data via a Python pipeline.
GPU-accelerated training and inference up to ×10 on compatible cards.
Open source code, free commercial use.
Captures transit patterns spread across weeks and months. Robust to noise in real-world data.