Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
HyperView is a terminal-first TradingView strategy lab for downloading market data, backtesting Python strategies with Pine-like behavior, and optimizing SL/TP parameters.
| Date | Stars |
|---|---|
| 2026-07-31 | 970 |
| 2026-08-03 | 971 |
| 2026-08-06 | 973 |
Today
+2 stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# HyperView **Turn TradingView ideas into testable, terminal-speed trading systems.** HyperView is for the moment a TradingView strategy stops being a chart experiment and starts needing hard evidence. It pulls historical candles straight from TradingView's websocket, runs your strategy logic in Python, and backtests with fill behavior designed to closely mirror Pine Script, so the results you tune locally in python still match the results you'll see on TradingView's strategy tester. Instead of bouncing between Pine scripts, CSV exports, and improvised notebooks, HyperView gives you one clean loop: pull up to 40K bars, build on [TA-Lib](https://github.com/TA-Lib/ta-lib-python)'s 150+ indicators, simulate realistic SL/TP execution, and let Bayesian optimization (Optuna TPE) hunt for better parameter ranges. No API keys. No browser automation. No spreadsheet cleanup. Just faster iteration, sharper validation, and a workflow built for traders who want to develop strategies like engineers. ## Prerequisites - **Python 3.11+** - **TA-Lib** — installed automatically by `pip install`. Pre-built wheels ship for Python 3.9–3.14 on Windows, macOS, and Linux. - **rich** — installed automatically. Powers the styled terminal output (colored tables, progress indicators, panels). - **Firefox** *(optional)* — If you have a TradingView paid plan, HyperView can read your Firefox session cookies to download up to **40K candles**. Without it, the websocket still downloads up to **5K candles** anonymously. To use this, just log in to [tradingview.com](https://www.tradingview.com) in Firefox before downloading data. ## Quick Start ```bash # Install in editable mode (creates the `hyperview` CLI command, installs all dependencies including TA-Lib) pip install -e . # Download data for specific pairs hyperview download-data --pairs NASDAQ:NFLX NASDAQ:AAPL --timeframe 1h --session extended # Or define your pairs in config.json and download multiple timeframes at once: hyperview download-data --timeframe 1h 15m # Run a single backtest (uses config pairlist) hyperview backtest --sl 3.23 --tp 13.06 --mode long # Or target a specific symbol using values from a hyperopt preset file hyperview backtest --symbol NASDAQ:NFLX --preset-file results/adx_stochastic_presets.json # Hyper-optimize SL/TP across all pairs in config hyperview hyperopt --mode long # List cached data and registered strategies hyperview list-data hyperview list-strategies ``` You can also run via `python -m hyperview` instead of the `hyperview` command. Python bytecode is redirected into the project-level `.pycache/` directory, so runtime imports do not create scattered `__pycache__` folders under `hyperview/` or `strategy/`. ## How It Works 1. **Download** — Connects to TradingView's websocket using your existing Firefox session cookies. Supports up to 40K historical bars on paid plans with automatic backfill. 2. **Signal** — Runs a pluggable strategy (e.g. the included MACD+RSI or ADX+Stochastic) in pure Python with TA-Lib indicator parity. 3. **Backtest** — Simulates trades bar-by-bar using TradingView-parity fill assumptions (next-bar-open entry, intrabar SL/TP exit ordering). Multi-pair runs produce a true **PORTFOLIO** aggregate row with combined equity-curve statistics. 4. **Hyper-Optimize** — Runs Bayesian optimization (Optuna TPE) across SL/TP combinations, then updates a strategy preset file with the best result for each pair/context. ## Terminal Output Both the backtest and hyperopt commands produce styled terminal output using [rich](https://github.com/Textualize/rich): - **Backtest summary** — A bordered table with colored directional arrows (▲ green for gains, ▼ red for losses) on Return, Drawdown, Expectancy, and Worst Trade, using readable short labels that fit a normal terminal width. When multiple pairs are run, a **PORTFOLIO** row is appended with mathematically correct aggregate statistics computed from a combined equity curve (not simple averages). - *
Excerpt of 14,595 characters
Read on GitHub
183
2
1
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:06d5c712006c520b, topic:quantitative-finance