📄️ Historical Data
HistoricalData is Limen's stateful file-backed data surface. It has four public retrieval methods:
📄️ Data Bars
Limen currently supports threshold-based bar formation over existing kline data. This is an optional preprocessing step inside a manifest: you start from regular time-based klines, then aggregate consecutive rows until a volume, trade-count, or liquidity threshold is reached.
📄️ Single File Decoder
A Single File Decoder (SFD) is the unit of experiment definition in Limen. It is a Python module that packages the parameter space together with either a declarative manifest or fully custom preparation and model functions.
📄️ Built-In SFDs
Limen ships a small set of foundational SFDs under limen.sfd.foundational_sfd. These are the packaged decoders you can run immediately without authoring your own experiment module first.
📄️ Experiment Manifest
The Experiment Manifest is Limen's declarative pipeline builder. Instead of hand-writing prep() and threading every step yourself, you describe how data should be fetched, split, transformed, targeted, scaled, and handed to a model.
📄️ Universal Experiment Loop
The Universal Experiment Loop (UEL) is Limen's experiment runner. It takes an SFD, executes parameter combinations, streams a CSV log, and then exposes post-run analysis surfaces such as confusion metrics and backtest results.
📄️ Advanced Search
Advanced search is Limen's artifact-rich experiment path. It adds a mutable parameter domain, a search strategy abstraction, checkpointing, resumability, and mid-run feedback on top of the normal Universal Experiment Loop.
📄️ Reducers And Feedback
Reducers and feedback are the control layer inside Limen's advanced search path. They are what let a run react to its own results instead of sweeping the full original domain unchanged.
📄️ Log
Log is Limen's post-run analysis layer. It sits on top of a finished experiment and turns raw round results into round-level prediction tables, benchmark-style summaries, backtest summaries, and parameter-correlation views.
📄️ Benchmark
In Limen, benchmark is the prediction-quality layer between the raw experiment log and the trading backtest.
📄️ Backtest
Backtest is Limen's trading-economics layer. It takes prediction outputs and asks the next question after benchmark:
📄️ Trainer
Trainer is Limen's promotion layer for finished experiment rounds. It takes a completed artifact-rich experiment directory, reconstructs the manifest and round parameters, validates selected permutations, and retrains them into reusable Sensor objects.
📄️ Conserved Flux Renormalization
Conserved Flux Renormalization (CFR) is a trade-data feature that summarizes how traded value and trade-size entropy behave across multiple time scales inside each bar.