Repository navigation
feed: ship a single committed 1m feed; derive chart feeds locally - #5
Merged
Merged
Conversation
Collapse the four shipped OHLCV CSVs into one: - data/ohlcv_ETH-USDT-USDT_1m.csv (Git LFS) — Binance ETH-USDT-USDT perp 1m bars, full exchange history (2020 listing → window end). This is the former *_1m_warmup6m.csv, renamed now that "warmup6m" no longer describes it. - The 15m chart feeds (full-history and window-only) are now derived deterministically from it into data/derived/ (gitignored) by the engine repo's scripts/derive_corpus_feeds.py; run_corpus.sh and run_strategy.py invoke it automatically. - Remove data/build_deep_warmup_feed.py: it pointed at private infrastructure (internal R2 endpoint/bucket) and does not belong in a public repo. The committed feed is the reproducibility artifact. - Probe feed pins updated: ltf-bool-array-bull-majority-01 → committed 1m; composite-trendmaster-three-tier-ema-state-01 and vwap-bands-mean-reversion-2sigma-01 (cold-start) → derived window slice. README/LEGAL/draft-probes docs updated. Known value deltas vs the old 15m files, from deriving out of 1m klines: two warmup-region bars on the 2024-10-28 outage day, a 0.01 open on the 1756449000000 bar (Binance's 15m kline disagrees with its own 1m klines during a flat outage), and a fuller partial final bucket. Engine parity gate re-verified at excellent=251/anomaly=1. Co-Authored-By: Claude Fable 5 <[email protected]> Claude-Session: https://claude.ai/code/session_01CnvqmHPmgUpeu2fz6A1mMU
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The corpus now ships exactly one OHLCV file:
data/ohlcv_ETH-USDT-USDT_1m.csv(Git LFS) — Binance ETH-USDT-USDT perp 1m bars, full exchange history (2020 listing → window end). It is the former*_1m_warmup6m.csv, renamed now that "warmup6m" no longer describes its depth.What changed
data/ohlcv_ETH-USDT-USDT_15m.csv,..._15m_warmup6m.csv, and the old window-only..._1m.csv. The 15m chart feeds (full-history + comparison-window slice) are now derived deterministically intodata/derived/(gitignored) by the engine repo'sscripts/derive_corpus_feeds.py, invoked automatically byrun_corpus.sh/run_strategy.py.data/build_deep_warmup_feed.py: it pointed at private infrastructure (internal R2 endpoint/bucket) and does not belong in a public repo. The committed feed is the reproducibility artifact.ltf-bool-array-bull-majority-01→ committed 1m; the two cold-start probes → derived window slice. Docs (README/LEGAL/draft-probes) updated.engine_trades.csvrefreshed: cumulative-PnL ripples (≈0.02 USD) from three bar-value deltas inherent to deriving 15m out of 1m klines (two 2024-10-28 outage-day warmup bars, one 0.01 in-window open where Binance's 15m kline disagrees with its own 1m klines, and a fuller final partial bucket).Gate (engine
feed/single-1m-derive): ctest 78/78; corpus sweep 252/252 ok;verify_corpus.py --all= excellent=251 / anomaly=1 — exact baseline hold.🤖 Generated with Claude Code
https://claude.ai/code/session_01CnvqmHPmgUpeu2fz6A1mMU