OPEN-SOURCE SCRIPT
AetherEdge - PULSE Adaptive Signal Intelligence
🖊️ Overview
PULSE is an all-in-one signal intelligence that grades trend-continuation (confirmation) and reversal (contrarian) signals through a self-learning AI classifier. Each time a signal fires, online learning predicts the probability the move will continue and sorts it into one of four tiers. Unlike conventional tools that merely light up signals, PULSE learns conviction from experience and enables filtering to keep only high-quality signals. An adaptive Smart Trail and smooth gradient visuals render the very breath of the trend.
🔶 Key Features
Confirmation signals: continuation entries from momentum/trend agreement, graded Normal / Strong (+)
Contrarian signals: reversal entries from oscillator extremes, shown as diamond markers
Exit marks: momentum fade after a signal flagged as take-profit candidates with "✕"
Smart Trail: an ATR-based adaptive trail serving as both dynamic support/resistance and a signal filter
[AI Layer] Multi-tier classification: an online logistic regression learns each signal's continuation probability and grades it ★★/★/◇/◆
Tier filter: set a minimum tier to automatically exclude low-conviction signals
Smooth gradient candles: continuous red↔violet↔green coloring by trend score
Vertical gradient fill on the Smart Trail and layered neon glow for dimensional depth
Refined red/green jewel-tone palettes (three) with a gold-framed integrated HUD
Built-in confirmation/contrarian alerts
🧠 Technical Architecture
PULSE's foundation is a composite trend-and-momentum core. A trend score — the fast/slow EMA difference normalized by ATR — provides direction and strength, while RSI-derived smoothed momentum supplies turning points. Confirmation signals fire when momentum crosses zero in the trend direction, promoting to Strong (+) when trend strength exceeds a sensitivity-scaled threshold. Contrarian signals capture momentum reversing from an extreme. The Smart Trail, an ATR-based adaptive trail (supertrend-style), reverses direction when price pierces it, simultaneously serving as dynamic support/resistance and a directional signal filter.
Atop this sits the AI classification layer — a genuine online logistic regression. From four standardized features (trend score, momentum, trend strength, Smart Trail agreement), it predicts the probability of continuation in the signal's direction. When a signal fires, its features and direction are buffered; after a set outcome horizon, a supervised label is generated from whether price actually continued in the signal direction, and the weights are updated via stochastic gradient descent. Using the learned continuation probability, each signal is graded into four tiers: ★★ (strong continuation), ★ (continuation), ◇ (reversal-leaning), and ◆ (strong reversal). Setting a minimum-tier filter automatically excludes low-conviction signals, leaving only high-conviction ones on the chart.
In the visual layer, candles are colored by a continuous gradient tied to the trend score, and the space to the Smart Trail is filled with a vertical gradient, intuitively conveying trend strength and distance from the trail. Layered glow gives signals and the trail a smooth emission.
⚙️ Recommended Settings & Tuning Guide
PULSE is tuned with crypto in mind. On majors like BTC and ETH, Sensitivity near 5 is a starting point. On higher-volatility names such as SOL and XRP, raising Sensitivity to 7–9 suppresses noise signals and stabilizes quality. Use Confirmation mode (trend continuation) as the baseline; in choppy, range-heavy conditions choose Contrarian, or Both to see everything.
Sensitivity is the central frequency parameter — higher is more selective. In the AI layer, the Outcome Horizon controls the evaluation window for "continuation" and the Learning Rate controls adaptation speed. The most practical tuning lever is "Minimum Tier to Show." Start at tier 1 (show all); when signals are too frequent or fakeouts stand out, raise to tier 3 to narrow to high-continuation-probability signals. Combining with the Smart Trail Filter — keeping only signals aligned with the trail — raises conviction further.
💡 How to Use in Practice
The highest-conviction setups are confluences where multiple elements align at once. A Strong (+) confirmation signal accompanied by a ★★ or ★ AI tier and aligned with the Smart Trail direction is a strong continuation-entry candidate. Conversely, a confirmation signal the AI grades ◆ (strong reversal) suggests the trend may be running out of breath — a cue to stand aside or take profit early.
The HUD's continuation probability and model accuracy gauge how much to trust the AI. If model accuracy stably exceeds 50%, it is a sign the tier classification is meaningful. For multi-timeframe trading, pair a higher-timeframe PULSE to read the macro trend and Smart Trail with a lower timeframe for precise entry timing. Contrarian signals serve counter-trend trades in ranges, and exit marks serve as take-profit timing cues — each used according to conditions.
⚠️ Important Notes
The AI classifier requires accumulated data to learn, so near the chart's left edge or when signals are few, model accuracy stays at its initial value (50%) and tier classification remains neutral. Static model accuracy in the HUD is a sign that learning data is still insufficient. Whether model accuracy stably exceeds 50% depends on the instrument, timeframe, and period, and cannot overcome the fundamental constraint of market predictability. Because signals are designed for trend-following, note that confirmation signal accuracy degrades in violently ranging markets.
🚨 Disclaimer
This indicator is provided for educational and informational purposes only and does not constitute financial or investment advice. No signal guarantees future profit, and past performance does not indicate future results. All trading decisions are made at your own risk; please use this tool with sound risk management and your own independent verification.
PULSE is an all-in-one signal intelligence that grades trend-continuation (confirmation) and reversal (contrarian) signals through a self-learning AI classifier. Each time a signal fires, online learning predicts the probability the move will continue and sorts it into one of four tiers. Unlike conventional tools that merely light up signals, PULSE learns conviction from experience and enables filtering to keep only high-quality signals. An adaptive Smart Trail and smooth gradient visuals render the very breath of the trend.
🔶 Key Features
Confirmation signals: continuation entries from momentum/trend agreement, graded Normal / Strong (+)
Contrarian signals: reversal entries from oscillator extremes, shown as diamond markers
Exit marks: momentum fade after a signal flagged as take-profit candidates with "✕"
Smart Trail: an ATR-based adaptive trail serving as both dynamic support/resistance and a signal filter
[AI Layer] Multi-tier classification: an online logistic regression learns each signal's continuation probability and grades it ★★/★/◇/◆
Tier filter: set a minimum tier to automatically exclude low-conviction signals
Smooth gradient candles: continuous red↔violet↔green coloring by trend score
Vertical gradient fill on the Smart Trail and layered neon glow for dimensional depth
Refined red/green jewel-tone palettes (three) with a gold-framed integrated HUD
Built-in confirmation/contrarian alerts
🧠 Technical Architecture
PULSE's foundation is a composite trend-and-momentum core. A trend score — the fast/slow EMA difference normalized by ATR — provides direction and strength, while RSI-derived smoothed momentum supplies turning points. Confirmation signals fire when momentum crosses zero in the trend direction, promoting to Strong (+) when trend strength exceeds a sensitivity-scaled threshold. Contrarian signals capture momentum reversing from an extreme. The Smart Trail, an ATR-based adaptive trail (supertrend-style), reverses direction when price pierces it, simultaneously serving as dynamic support/resistance and a directional signal filter.
Atop this sits the AI classification layer — a genuine online logistic regression. From four standardized features (trend score, momentum, trend strength, Smart Trail agreement), it predicts the probability of continuation in the signal's direction. When a signal fires, its features and direction are buffered; after a set outcome horizon, a supervised label is generated from whether price actually continued in the signal direction, and the weights are updated via stochastic gradient descent. Using the learned continuation probability, each signal is graded into four tiers: ★★ (strong continuation), ★ (continuation), ◇ (reversal-leaning), and ◆ (strong reversal). Setting a minimum-tier filter automatically excludes low-conviction signals, leaving only high-conviction ones on the chart.
In the visual layer, candles are colored by a continuous gradient tied to the trend score, and the space to the Smart Trail is filled with a vertical gradient, intuitively conveying trend strength and distance from the trail. Layered glow gives signals and the trail a smooth emission.
⚙️ Recommended Settings & Tuning Guide
PULSE is tuned with crypto in mind. On majors like BTC and ETH, Sensitivity near 5 is a starting point. On higher-volatility names such as SOL and XRP, raising Sensitivity to 7–9 suppresses noise signals and stabilizes quality. Use Confirmation mode (trend continuation) as the baseline; in choppy, range-heavy conditions choose Contrarian, or Both to see everything.
Sensitivity is the central frequency parameter — higher is more selective. In the AI layer, the Outcome Horizon controls the evaluation window for "continuation" and the Learning Rate controls adaptation speed. The most practical tuning lever is "Minimum Tier to Show." Start at tier 1 (show all); when signals are too frequent or fakeouts stand out, raise to tier 3 to narrow to high-continuation-probability signals. Combining with the Smart Trail Filter — keeping only signals aligned with the trail — raises conviction further.
💡 How to Use in Practice
The highest-conviction setups are confluences where multiple elements align at once. A Strong (+) confirmation signal accompanied by a ★★ or ★ AI tier and aligned with the Smart Trail direction is a strong continuation-entry candidate. Conversely, a confirmation signal the AI grades ◆ (strong reversal) suggests the trend may be running out of breath — a cue to stand aside or take profit early.
The HUD's continuation probability and model accuracy gauge how much to trust the AI. If model accuracy stably exceeds 50%, it is a sign the tier classification is meaningful. For multi-timeframe trading, pair a higher-timeframe PULSE to read the macro trend and Smart Trail with a lower timeframe for precise entry timing. Contrarian signals serve counter-trend trades in ranges, and exit marks serve as take-profit timing cues — each used according to conditions.
⚠️ Important Notes
The AI classifier requires accumulated data to learn, so near the chart's left edge or when signals are few, model accuracy stays at its initial value (50%) and tier classification remains neutral. Static model accuracy in the HUD is a sign that learning data is still insufficient. Whether model accuracy stably exceeds 50% depends on the instrument, timeframe, and period, and cannot overcome the fundamental constraint of market predictability. Because signals are designed for trend-following, note that confirmation signal accuracy degrades in violently ranging markets.
🚨 Disclaimer
This indicator is provided for educational and informational purposes only and does not constitute financial or investment advice. No signal guarantees future profit, and past performance does not indicate future results. All trading decisions are made at your own risk; please use this tool with sound risk management and your own independent verification.
Mã nguồn mở
Theo đúng tinh thần TradingView, tác giả của tập lệnh này đã công bố nó dưới dạng mã nguồn mở, để các nhà giao dịch có thể xem xét và xác minh chức năng. Chúc mừng tác giả! Mặc dù bạn có thể sử dụng miễn phí, hãy nhớ rằng việc công bố lại mã phải tuân theo Nội quy.
Thông báo miễn trừ trách nhiệm
Thông tin và các ấn phẩm này không nhằm mục đích, và không cấu thành, lời khuyên hoặc khuyến nghị về tài chính, đầu tư, giao dịch hay các loại khác do TradingView cung cấp hoặc xác nhận. Đọc thêm tại Điều khoản Sử dụng.
Mã nguồn mở
Theo đúng tinh thần TradingView, tác giả của tập lệnh này đã công bố nó dưới dạng mã nguồn mở, để các nhà giao dịch có thể xem xét và xác minh chức năng. Chúc mừng tác giả! Mặc dù bạn có thể sử dụng miễn phí, hãy nhớ rằng việc công bố lại mã phải tuân theo Nội quy.
Thông báo miễn trừ trách nhiệm
Thông tin và các ấn phẩm này không nhằm mục đích, và không cấu thành, lời khuyên hoặc khuyến nghị về tài chính, đầu tư, giao dịch hay các loại khác do TradingView cung cấp hoặc xác nhận. Đọc thêm tại Điều khoản Sử dụng.