Logistic Regression

题面

在 GPU 上求解逻辑回归：给定 X∈R^{n_samples×n_features} 与二元标签 y∈{0,1}^{n_samples}，最大化对数似然 Σ [ y_i log σ(X_i^T β) + (1-y_i) log (1-σ(X_i^T β)) ]，σ 为 Sigmoid。

Implementation Requirements

• External libraries are not permitted
• The solve function signature must remain unchanged
• 系数写入向量 beta；y 仅包含 0/1

Examples

见页面给出的样例 X、y 与 β。

Constraints

• 1 ≤ n_samples ≤ 100,000；1 ≤ n_features ≤ 1,000；n_samples ≥ n_features
• -10.0 ≤ X ≤ 10.0；y ∈
• 评测容差：abs/rel 1e-2；Performance: n_features=8, n_samples=16