📊 Trading

Evolve multi-stage trading pipelines that maximize Sharpe ratio.

---

Prerequisites

1. VEOX Server: Start the local VEOX server (requires Docker):

   docker run -d \
     --name veox-enclave-server \
     -p 8090:8090 \
     714044927654.dkr.ecr.us-east-2.amazonaws.com/doug/single_enclave/veox-enclave-server:latest
   

   See the Quick Start for detailed server setup, health checks, and Docker Compose instructions.

2. Python SDK: Install the veox package via PyPI:

   pip install veox
   

---

What It Does

The trading family is the most complex, evolving 19-stage pipelines:

Signal → Feature Engineering (×4) → Regime Detection → Risk → Position Sizing →
Execution → Portfolio → + 9 more stages

Each candidate is backtested across 367+ financial datasets using the proprietary VEOX trading evaluator. This runs against server-side data and does not require client-side datasets.

Quick Start

from veox import VeoxEvolver

evolver = VeoxEvolver("trading")
evolver.fit(max_generations=5, population_size=30, max_poll_time=900)
print(f"Best Sharpe: {evolver.best_fitness_:.4f}")
print(f"Pipeline:    {evolver.best_pipeline_}")

Full Example

VEOX FreeVEOX Pro

from veox import VeoxEvolver

# 1. Connect
evolver = VeoxEvolver("trading", api_url="http://127.0.0.1:8090")
evolver.health_check()

# 2. Evolve (trading uses server-side financial datasets)
evolver.fit(
    max_generations=5,
    population_size=30,
    timeout_per_eval=120,    # Trading evals are longer
    max_poll_time=900,       # 15 min wall-clock
)

# 3. Inspect
print(f"Best Sharpe: {evolver.best_fitness_:.4f}")
print(f"Pipeline:    {evolver.best_pipeline_}")
print(f"Evaluations: {evolver.result_.total_evals}")
print(f"Throughput:  {evolver.result_.evals_per_minute:.1f} evals/min")

# 4. Save
evolver.save("trading_results.json")

from veox import VeoxEvolver

# 1. Connect
evolver = VeoxEvolver("trading", api_url="http://127.0.0.1:8090")
evolver.health_check()

# 2. Evolve (trading uses server-side financial datasets)
evolver.fit(
    max_generations=5,
    population_size=30,
    num_islands=4,              # 💎 PRO FEATURE: 4 parallel islands
    timeout_per_eval=120,    # Trading evals are longer
    max_poll_time=900,       # 15 min wall-clock
)

# 3. Inspect
print(f"Best Sharpe: {evolver.best_fitness_:.4f}")
print(f"Pipeline:    {evolver.best_pipeline_}")
print(f"Evaluations: {evolver.result_.total_evals}")
print(f"Throughput:  {evolver.result_.evals_per_minute:.1f} evals/min")

# 4. Save
evolver.save("trading_results.json")

Live dashboard — Sharpe ratio fitness scatter, champion trend, and trading pipeline leaderboard.

19-Stage Pipeline

#	Slot	Purpose
1	required	Mandatory base stage
2	feature_1	Primary feature engineering
3	feature_2	Secondary features
4	feature_3	Tertiary features
5	feature_4	Quaternary features
6	signal	Signal generation
7	regime	Market regime detection
8	risk	Risk management
9	sizing	Position sizing
10	execution	Order execution
11	portfolio	Portfolio management
12	filter	Trade filtering
13	timing	Entry/exit timing
14	hedge	Hedging strategy
15	correlation	Correlation analysis
16	volatility	Volatility modeling
17	momentum	Momentum signals
18	mean_reversion	Mean reversion signals
19	ensemble	Strategy ensemble

Fitness Configuration

Parameter	Value
Primary Metric	Sharpe Ratio
Aggregation	mean − 0.5 × std (penalize variance)
Direction	Maximize
Exception Penalty	−99.0
Datasets	367+ financial instruments

💎 VEOX Pro Activation

To unlock VIP Evaluators and Pro Algorithms (like PaperKit and Generative routines), you must authenticate your local node with a VEOX License Token.

from veox import VeoxEvolver

evolver = VeoxEvolver("trading", api_url="http://127.0.0.1:8090")

# 1. Fetch your unique Hardware Fingerprint
fingerprint = evolver.get_system_fingerprint()
print(f"My VEOX Node Fingerprint: {fingerprint}")
# Example Output: My VEOX Node Fingerprint: 476ad03474b31e3c84d07df9088d93f0

# 2. Provide this fingerprint to your VEOX Admin to receive a JWT Token
jwt_token = "eyJ0b2tlbiI6ICJVRExK...EXPIRES"  # Replace with your token

# 3. Activate the Enclave
if evolver.activate_license(jwt_token):
    print("VIP Features Unlocked!")
    # evolver.fit(...) will now utilize full Pro capabilities

Multiple Datasets

Trading supports both server-side datasets and client-uploaded OHLCV data:

import pandas as pd
import numpy as np
from veox import VeoxEvolver

# Generate synthetic OHLCV data (must have a 'close' column)
def make_ohlcv(name, length=1200, seed=42):
    rng = np.random.default_rng(seed)
    close = 100 + np.cumsum(rng.normal(0, 1, length))
    return pd.DataFrame({
        "open": close + rng.normal(0, 0.5, length),
        "high": close + abs(rng.normal(0, 1, length)),
        "low": close - abs(rng.normal(0, 1, length)),
        "close": close,
        "volume": rng.integers(1000, 10000, length),
    })

df1 = make_ohlcv("ETH", seed=42)
df2 = make_ohlcv("BTC", seed=99)

evolver = VeoxEvolver("trading")
evolver.fit(
    data=[df1, df2],
    target_column="close",      # Trading requires a close/price column
    max_generations=3,
)
print(f"Best Sharpe (averaged): {evolver.best_fitness_:.4f}")

Server-side data

By default, trading uses proprietary financial datasets bundled with the server. You can mix server-side and uploaded data.

Tips

• Patience: Trading evals are 10-100× slower than classification.
• Start with 1 generation: max_generations=1 to verify the pipeline boots.
• Server-side data: Trading uses proprietary financial datasets on the server.
• Client data: Upload your own OHLCV DataFrames — must include a close column.
• 6-Phase ramp: The engine automatically ramps complexity over generations.