ML Signal · StockLens

ML Signal Buy · Sell · Hold predictions powered by machine learning

Label params:

🌐 Universal Model — train on multiple stocks Uses all stocks in your DB with financial data (862 available)

Number of stocks

Algorithm

Label strategy

Last run: {{ univResult.n_stocks }} stocks · {{ univResult.total_rows?.toLocaleString() }} rows

Test symbol (for signal & backtest)

Exchange

The universal model was trained on 50 stocks — predict on any NSE stock

Trained on ({{ univResult.stocks_used?.length }})

Skipped ({{ univResult.stocks_failed?.length }})

No signal yet

Enter a symbol and click Get signal.

{{ signalResult.symbol || sym }} · {{ signalResult.strategy?.toUpperCase() }} {{ signalResult.date }}

{{ signalResult.signal === 'BUY' ? '▲' : signalResult.signal === 'SELL' ? '▼' : '—' }} {{ signalResult.signal }}

Confidence

{{ pct(signalResult.confidence) }}%

Class probabilities

{{ k }}: {{ pct(v) }}%

⚠️ ML signals are probabilistic, not financial advice. Paper trade before using real capital.

No training run yet

Enter a symbol and click Train model.

Bars trained (80%)

OOS bars (20%)

Accuracy (CV)

{{ pct(trainResult.accuracy) }}%

BUY precision

{{ pct(trainResult.buy_precision) }}%

SELL precision

{{ pct(trainResult.sell_precision) }}%

BUY recall

{{ pct(trainResult.buy_recall) }}%

Algorithm

Top 10 features (importance)

Walk-forward CV report

Class	Precision	Recall	F1	Support
{{ cls }}	{{ pct(trainResult.cv_report?.[cls]?.precision) }}%	{{ pct(trainResult.cv_report?.[cls]?.recall) }}%	{{ pct(trainResult.cv_report?.[cls]?.['f1-score']) }}%	{{ trainResult.cv_report?.[cls]?.support ?? '—' }}

No universal model trained yet

Set the number of stocks above and click Train universal model.

Stocks trained

Total rows

Train rows (80%)

OOS rows (20%)

Accuracy (OOS)

{{ pct(univResult.accuracy) }}%

BUY precision

{{ pct(univResult.buy_precision) }}%

SELL precision

{{ pct(univResult.sell_precision) }}%

Algorithm

Top 10 features (importance)

OOS classification report

Class	Precision	Recall	F1	Support
{{ cls }}	{{ pct(univResult.cv_report?.[cls]?.precision) }}%	{{ pct(univResult.cv_report?.[cls]?.recall) }}%	{{ pct(univResult.cv_report?.[cls]?.['f1-score']) }}%	{{ univResult.cv_report?.[cls]?.support ?? '—' }}

To get signals from this model: type UNIVERSAL in the Symbol field and click Get signal or Backtest.

No backtest yet

Train a model first, then click Backtest.

Out-of-sample period: {{ backtestResult.oos_start_date }} → {{ backtestResult.oos_end_date }} ({{ backtestResult.oos_bars }} bars — model never saw this data)

Total return

{{ backtestResult.total_return_pct }}%

Sharpe ratio

Max drawdown

{{ backtestResult.max_drawdown_pct }}%

Win rate

{{ backtestResult.win_rate_pct }}%

Total trades

Avg win

{{ backtestResult.avg_win_trade_pct }}%

Avg loss

{{ backtestResult.avg_loss_trade_pct }}%

Engine

Evaluation checklist

{{ activeTrainResult?.buy_precision >= 0.55 ? '✓' : '✗' }} BUY precision ≥ 55% (got {{ activeTrainResult ? pct(activeTrainResult.buy_precision)+'%' : '—' }})

{{ activeTrainResult?.sell_precision >= 0.55 ? '✓' : '✗' }} SELL precision ≥ 55% (got {{ activeTrainResult ? pct(activeTrainResult.sell_precision)+'%' : '—' }})

{{ backtestResult.sharpe_ratio >= 0.8 ? '✓' : '✗' }} Sharpe ratio ≥ 0.8 (got {{ backtestResult.sharpe_ratio }})

{{ backtestResult.max_drawdown_pct <= 20 ? '✓' : '✗' }} Max drawdown ≤ 20% (got {{ backtestResult.max_drawdown_pct }}%)

{{ backtestResult.win_rate_pct >= 50 ? '✓' : '✗' }} Win rate ≥ 50% (got {{ backtestResult.win_rate_pct }}%)

☐ Paper trade 60 days before using real capital

No models saved yet

Train a model to see it listed here.

Saved models ({{ modelsList.length }})

Symbol	Strategy	Saved at
{{ m.symbol }}	{{ m.strategy_name?.toUpperCase() }}	{{ formatTs(m.ts) }}