📈 Regression

Evolve ML pipelines that maximize R² on regression tasks.

---

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 regression family genetically evolves multi-stage ML pipelines for continuous value prediction:

Scaler → Expander₁ → Expander₂ → Reducer → Regressor₁ → Regressor₂ → Fusion

Candidates are scored using R² (coefficient of determination) via K-fold cross-validation.

Quick Start

from sklearn.datasets import make_regression
import pandas as pd
from veox import VeoxEvolver

# Generate a regression dataset
X, y = make_regression(n_samples=500, n_features=10, n_informative=6,
                        noise=15, random_state=42)
df = pd.DataFrame(X, columns=[f"f{i}" for i in range(10)])
df["target"] = y

evolver = VeoxEvolver("regression")
evolver.fit(data=df, target_column="target", max_generations=5)
print(f"Best R²: {evolver.best_fitness_:.4f}")

Full Example

VEOX FreeVEOX Pro

from sklearn.datasets import make_regression
import pandas as pd
from veox import VeoxEvolver

# 1. Generate challenging regression dataset
X, y = make_regression(
    n_samples=1000,
    n_features=15,
    n_informative=8,
    noise=20,            # Significant noise
    random_state=42,
)
df = pd.DataFrame(X, columns=[f"feature_{i}" for i in range(15)])
df["target"] = y

print(f"Dataset: {df.shape[0]} rows × {df.shape[1]} cols")
print(f"Target range: [{y.min():.1f}, {y.max():.1f}]")

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

# 3. Evolve
evolver.fit(
    data=df,
    target_column="target",
    max_generations=10,
    population_size=50,
    timeout_per_eval=30,
    max_poll_time=600,
)

# 4. Inspect
print(f"Best R²:     {evolver.best_fitness_:.4f}")
print(f"Pipeline:    {evolver.best_pipeline_}")
print(f"Evaluations: {evolver.result_.total_evals}")

# 5. Save
evolver.save("regression_results.json")

from sklearn.datasets import make_regression
import pandas as pd
from veox import VeoxEvolver

# 1. Generate challenging regression dataset
X, y = make_regression(
    n_samples=1000,
    n_features=15,
    n_informative=8,
    noise=20,            # Significant noise
    random_state=42,
)
df = pd.DataFrame(X, columns=[f"feature_{i}" for i in range(15)])
df["target"] = y

print(f"Dataset: {df.shape[0]} rows × {df.shape[1]} cols")
print(f"Target range: [{y.min():.1f}, {y.max():.1f}]")

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

# 3. Evolve
evolver.fit(
    data=df,
    target_column="target",
    max_generations=10,
    population_size=50,
    num_islands=4,              # 💎 PRO FEATURE: 4 parallel islands
    timeout_per_eval=30,
    max_poll_time=600,
)

# 4. Inspect
print(f"Best R²:     {evolver.best_fitness_:.4f}")
print(f"Pipeline:    {evolver.best_pipeline_}")
print(f"Evaluations: {evolver.result_.total_evals}")

# 5. Save
evolver.save("regression_results.json")

Live dashboard — R² fitness scatter, champion trend, and pipeline leaderboard.

Fitness Configuration

Parameter	Value
Primary Metric	R² (coefficient of determination)
K-Fold Splits	3
Direction	Maximize
Clamped Range	[-1.0, 1.0]
Exception Penalty	−2.0

Open-Source Datasets to Try

from sklearn.datasets import fetch_california_housing, make_friedman1
import pandas as pd
from veox import VeoxEvolver

# California Housing (real-world)
data = fetch_california_housing(as_frame=True)
df = data.frame
df = df.rename(columns={"MedHouseVal": "target"})

evolver = VeoxEvolver("regression")
evolver.fit(data=df, target_column="target", max_generations=5)

# Friedman #1 (nonlinear benchmark — 5 informative, 5 noise)
X, y = make_friedman1(n_samples=1000, n_features=10, noise=1.0)
df = pd.DataFrame(X, columns=[f"f{i}" for i in range(10)])
df["target"] = y

evolver = VeoxEvolver("regression")
evolver.fit(data=df, target_column="target", max_generations=5)

💎 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("regression", 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

Train on several datasets to evolve regressors that generalize across different distributions:

from sklearn.datasets import make_regression
import pandas as pd
from veox import VeoxEvolver

# Two regression datasets with different signal-to-noise ratios
def make_reg_df(seed, noise=10.0):
    X, y = make_regression(n_samples=600, n_features=15, n_informative=8,
                           noise=noise, random_state=seed)
    df = pd.DataFrame(X, columns=[f"f{i}" for i in range(15)])
    df["target"] = y
    return df

df1 = make_reg_df(seed=42, noise=5.0)   # Low noise
df2 = make_reg_df(seed=99, noise=30.0)  # High noise

evolver = VeoxEvolver("regression")
evolver.fit(
    data=[df1, df2],
    target_column="target",
    max_generations=5,
)
print(f"Best R² (averaged): {evolver.best_fitness_:.4f}")

Tips

• R² = 1.0 means perfect fit; R² = 0 means predicting the mean.
• Multiple datasets: Pass a list to test generalization: data=[df1, df2].
• File paths: Pass CSV paths directly: data=["train_a.csv", "train_b.csv"].
• Noise matters: Higher noise → more generations needed to evolve robust regressors.