From: Jérôme Benoit Date: Sat, 4 Oct 2025 22:37:35 +0000 (+0200) Subject: docs(reforcexy): refine reward space analysis README X-Git-Url: https://git.piment-noir.org/?a=commitdiff_plain;h=7cc8c63ec7968d160958479f76265f99f0232707;p=freqai-strategies.git docs(reforcexy): refine reward space analysis README Signed-off-by: Jérôme Benoit --- diff --git a/ReforceXY/reward_space_analysis/README.md b/ReforceXY/reward_space_analysis/README.md index 80fb4dc..6a75f1f 100644 --- a/ReforceXY/reward_space_analysis/README.md +++ b/ReforceXY/reward_space_analysis/README.md @@ -1,6 +1,6 @@ # 📊 Reward Space Analysis - User Guide -**Analyze and validate ReforceXY reward logic with synthetic data (with built‑in runtime statistical & invariant validations)** +**Analyze and validate ReforceXY reward logic with synthetic data** --- @@ -340,7 +340,7 @@ The analysis generates the following output files: - **Distribution Shift** - Real vs synthetic divergence (KL, JS, Wasserstein, KS) - **Diagnostics Validation Summary** - Pass/fail snapshot of all runtime checks - - Consolidated pass/fail state of every validation layer (invariants, parameter bounds, bootstrap CIs, distribution metrics, diagnostics, hypothesis tests) + - Consolidated pass/fail state of every validation layer (invariants, parameter bounds, bootstrap CIs, distribution metrics, diagnostics, hypothesis tests) ### Data Exports @@ -430,7 +430,7 @@ done python test_reward_space_analysis.py ``` -The unified suite currently contains 32 focused tests (coverage ~84%). Example (abridged) successful run shows all test_* cases passing (see file for full list). Number may increase as validations expand. +The suite currently contains 32 focused tests (coverage ~84%). Example (abridged) successful run shows all test_* cases passing (see file for full list). Number may increase as validations expand. ### Test Categories