Skip to main content

​
Entry points

These are the top-level entry points (not in scripts/):
ScriptPurpose
trading.pyStart the live trading bot
backtest.pyRun a backtest
train_ml_model.pyTrain one or more models
# trading.py
python trading.py           # live
python trading.py --dry     # dry run — no real orders

# backtest.py
python backtest.py --balance 500 --no-swap --leverage 500 --spread 16.95 --oos-only --no-chart

# train_ml_model.py
python train_ml_model.py --ensemble --position --sltp --risk
python train_ml_model.py --ensemble --shap --refresh
python train_ml_model.py --risk   # retrain risk only

​
scripts/

​
onboarding.py

First-time setup wizard. Run once. 
python scripts/onboarding.py --balance 500
  • Prompts for .env credentials
  • Pulls latest models from HF Hub
  • Runs risk questionnaire (6 questions → profile 1–5)
  • Runs first weekly optimize

​
weekly_optimize.py

Autonomous 13-phase weekly pipeline. 
python scripts/weekly_optimize.py                           # full + questionnaire
python scripts/weekly_optimize.py --profile balanced        # skip questionnaire
python scripts/weekly_optimize.py --profile 3               # by profile number
python scripts/weekly_optimize.py --skip-retrain            # sweep only (~45 min)
python scripts/weekly_optimize.py --from-phase 5            # resume after crash
python scripts/weekly_optimize.py --dry-run                 # show plan, don't run
Profiles: steady_income (1), conservative (2), balanced (3), growth (4), aggressive (5).

​
retrain.py

Standalone retrain with optional steps. 
python scripts/retrain.py --all                # all 4 models
python scripts/retrain.py --signal             # signal ensemble only
python scripts/retrain.py --signal --shap      # with SHAP analysis
python scripts/retrain.py --refresh            # refresh dataset first

​
sweep.py

Config parameter sweep scoped to a risk profile. 
python scripts/sweep.py --profile 3                    # balanced
python scripts/sweep.py --profile 4 --oos-only         # OOS sweep
python scripts/sweep.py --profile 3 --dry-run          # show grid
python scripts/sweep.py --profile 3 --output results/  # save CSV
The sweep tests combinations of confidence_threshold, risk_percent, sl_multiplier, tp_multiplier. Results are ranked by the composite score.

​
sweep_signal.py

Signal-focused sweep — tests confidence thresholds and session filters. 
python scripts/sweep_signal.py
python scripts/sweep_signal.py --oos-only --save

​
sweep_pos_model.py

Position model parameter sweep. 
python scripts/sweep_pos_model.py
python scripts/sweep_pos_model.py --threshold 0.6

​
optimize_loop.py

Continuous optimization loop — runs SHAP → tune → retrain → OOS in a loop until a target score is reached or max iterations is hit. 
python scripts/optimize_loop.py --target 20 --max-iter 5
python scripts/optimize_loop.py --profile 3 --target 15
_PKL_FILES in optimize_loop.py must include .onnx files alongside .pkl files. If only .pkl is listed, the revert step restores wrong inference files (the ONNX models keep the old version).

​
performance_monitor.py

Detects performance regression against a baseline. 
python scripts/performance_monitor.py                    # compare vs stored baseline
python scripts/performance_monitor.py --baseline 20.0   # custom baseline score
python scripts/performance_monitor.py --alert-telegram   # send alert if regression

​
dashboard.py

Live terminal dashboard showing current position, equity, today’s P&L, and recent trade log. 
python scripts/dashboard.py
python scripts/dashboard.py --refresh 10   # refresh every 10 seconds

​
compare_configs.py

Diff two config.json files and show what changed. 
python scripts/compare_configs.py config.json results/config_best.json

​
config_sync.py

Sync the current config.json to another machine over SSH. 
python scripts/config_sync.py --host trading-vm --user ubuntu

​
notify.py

Send a manual Telegram notification. 
python scripts/notify.py "deploy complete — bot restarted"
python scripts/notify.py --level error "MT5 connection failed"

​
run_agent.sh

Launch a Claude Code agent for multi-step tasks. 
TASK="run backtest and report OOS score" ./scripts/run_agent.sh
TASK="sweep confidence threshold 0.55-0.70 and apply best" ./scripts/run_agent.sh
Use the TASK= variable form. Avoid multiline shell continuations for the TASK assignment.

​
ml/hf_hub.py

Hugging Face Hub model management. 
python ml/hf_hub.py --pull                # download latest
python ml/hf_hub.py --push                # upload after retrain
python ml/hf_hub.py --push --tag phase-16 # push with version tag
python ml/hf_hub.py --list                # show available revisions