Code Execution (Meta Skill)

Execute scientific Python code for computation, analysis, simulation, and verification of results.

Common Imports

import numpy as np
import pandas as pd
from scipy import stats, optimize, integrate, signal
import matplotlib; matplotlib.use('Agg')
import matplotlib.pyplot as plt
import json, csv, sys
from collections import Counter, defaultdict

Pattern 1: Statistical Analysis

import numpy as np
from scipy import stats

data_a, data_b = np.array([...]), np.array([...])
print(f"Group A: mean={np.mean(data_a):.4f}, std={np.std(data_a, ddof=1):.4f}, n={len(data_a)}")
print(f"Group B: mean={np.mean(data_b):.4f}, std={np.std(data_b, ddof=1):.4f}, n={len(data_b)}")

t_stat, p_value = stats.ttest_ind(data_a, data_b, equal_var=False)
print(f"Welch's t-test: t={t_stat:.4f}, p={p_value:.6f}")

# Effect size (Cohen's d)
pooled_std = np.sqrt((np.std(data_a, ddof=1)**2 + np.std(data_b, ddof=1)**2) / 2)
print(f"Cohen's d: {(np.mean(data_a) - np.mean(data_b)) / pooled_std:.4f}")

Pattern 2: Numerical Computation

from scipy import integrate, optimize

result, error = integrate.quad(lambda x: np.exp(-x**2), -np.inf, np.inf)
print(f"Integral result: {result:.6f} (error: {error:.2e})")

solution = optimize.fsolve(lambda v: [v[0]**2+v[1]**2-4, v[0]-v[1]-1], [1, 0])
print(f"Solution: x={solution[0]:.4f}, y={solution[1]:.4f}")

Pattern 3: Data Processing

import pandas as pd
from io import StringIO

df = pd.read_csv(StringIO("col1,col2\n1,2\n3,4"))
df = df.dropna()
df['computed'] = df['col1'] * df['col2']
print(df.groupby('col1').agg({'col2': ['mean', 'std', 'count']}).round(4).to_string())

Pattern 4: Monte Carlo Simulation

np.random.seed(42)
n = 100000
x, y = np.random.uniform(-1, 1, n), np.random.uniform(-1, 1, n)
pi_est = 4 * np.sum(x**2 + y**2 <= 1) / n
print(f"Pi estimate: {pi_est:.6f} (error: {abs(pi_est - np.pi):.6f})")

Pattern 5: Verification