总览 · ← 返回运行 20261001-233757-search-t1-abc-r0-C-native

节点 n10

改了什么

方法说明?节点程序自带的 METHOD.md：这个程序做了什么、为什么。

来自 programs.git 11ff9658e2:solution/METHOD.md

调研员的计划

代码改动?这个节点的程序和父节点程序的逐行差别：绿色是新增，红色是删除。

对比：父节点版本 01e7938a0f。改动的文件：solution/METHOD.md +7 −9、solution/run.py +27 −53

diff --git a/solution/METHOD.md b/solution/METHOD.mdindex 92e8c97..8898c66 100644--- a/solution/METHOD.md+++ b/solution/METHOD.md@@ -1,11 +1,9 @@ ## 改了什么-1. **Per-gene empirical Bayes shrinkage 替代统一 ALPHA**：对每个细胞类型的 pseudobulk delta，按基因的信噪比（delta² / (delta² + noise_variance)）做逐基因收缩。噪声方差用两个阶段的 within-type 方差除以各自细胞数估计。信号强的基因（大 delta、低方差）保留更多位移；噪声基因被压向零，减少对原始共变结构的破坏。这直接针对 covariation（当前最弱组 40.74）：统一 alpha 对所有基因一视同仁，噪声基因的随机位移会破坏 gene-gene 相关；逐基因收缩只在有信号的方向移动，保留无信号基因的原始协方差。-2. **ALPHA_GLOBAL 从 0.5 调到 0.6**：在逐基因收缩已经提供了自适应降噪的前提下，全局系数可以稍微放松，让高信噪比基因获得更充分的位移，改善 direction 和 de_recovery。-3. **保留 include_external=False 过滤**（父节点已验证 proxy2 +22.61）。-4. **移除对 baselines.shift_rows / type_deltas 的依赖**，改为自行计算 per-type 均值/方差和逐基因收缩，以获得更细粒度的控制。-5. **单输入阶段退化为 copy_last**（与父节点一致）。+1. **移除事后 per-type 方差重标定**：第 0 轮实验证明重标定未改善 covariation（41.44 vs 父节点 41.90），反而增加运行时间（7.5s vs 3.8s）。假设「clipping 破坏方差结构」不成立或重标定本身引入额外噪声，故移除。+2. **ALPHA_GLOBAL 从 0.5 降到 0.3**：更小的平移幅度产生更少的零值截断（clipping），从源头保留更多原始协方差结构。遵循父节点 ANALYSIS 建议「扫 ALPHA_GLOBAL in {0.3, 0.5, 0.7}」，本轮测试低端。预期：direction/de_recovery 可能略降（位移更小），但 covariation 应改善（更少截断 = 更少方差/相关扭曲）。+3. **保留 include_external=False 过滤和 stages 为空时的 fallback**。+4. **单输入阶段退化为 copy_last**。 ## 用到的知识与出处-- k018（方法卡 T1）：「Optionally shrink per gene (empirical Bayes on the delta's variance) so noisy genes move less」——本实现直接采用此建议。-- k018：「The official constant per-type shift (alpha=1) scores below copy_last on T1, so re-applying the last observed delta must be shrunk by alpha in [0,1]」。-- 父节点 ANALYSIS next_suggestions：「针对 covariation：在 shift_rows 后按 per-type 残差重标定」——本方案从源头（delta 计算）而非事后重标定来解决，因为逐基因收缩等价于在平移前就保护了方差结构。-- James-Stein shrinkage 原理（Efron & Morris 1975, "Data analysis using Stein's estimator and its generalizations"）：观测值向零收缩的最优系数为 signal/(signal+noise)。+- 父节点 ANALYSIS next_suggestions：「固定 EB 收缩，仅扫 ALPHA_GLOBAL in {0.3, 0.5, 0.7}；若 board 始终在噪声内，说明 alpha 不是杠杆」——本轮选 0.3 端测试。+- 第 0 轮实验反馈：方差重标定未改善 covariation（41.44 < 41.90），移除以减少无效复杂度和运行时间。+- k018：「re-applying the last observed delta must be shrunk by alpha in [0,1]」——alpha 越小对原始分布扰动越小。diff --git a/solution/run.py b/solution/run.pyindex ce643fc..b5e002b 100644--- a/solution/run.py+++ b/solution/run.py@@ -1,11 +1,11 @@ #!/usr/bin/env python3-"""Per-gene empirical Bayes shrinkage on pseudobulk deltas.+"""Post-shift per-type variance rescaling with uniform alpha. -Uses only official (non-external) input stages. For each cell type's-pseudobulk delta, applies per-gene shrinkage based on the signal-to-noise-ratio: genes with large deltas relative to their within-type variance are-shifted more; noisy genes are shrunk toward zero, preserving the original-covariation structure better than a uniform alpha.+Reverts to simpler uniform alpha (removing costly per-type EB variance+stats that gave no detectable gain). After applying the shift and clipping,+rescales per-type per-gene residuals to restore the original standard+deviation. This directly targets covariation by ensuring that clipping+at zero does not distort gene-gene correlation structure.  With a single official input stage, falls back to copy_last. """@@ -28,7 +28,7 @@ from src.task1_temporal.view_io import (     write_prediction, ) -ALPHA_GLOBAL = 0.6+ALPHA_GLOBAL = 0.3   def _to_dense(X):@@ -37,29 +37,6 @@ def _to_dense(X):     return np.asarray(X, dtype=np.float32)  -def _per_type_stats(X, labels):-    Xd = _to_dense(X)-    types = np.unique(labels)-    means = {}-    variances = {}-    counts = {}-    for t in types:-        mask = labels == t-        sub = Xd[mask]-        means[t] = sub.mean(axis=0)-        variances[t] = sub.var(axis=0)-        counts[t] = mask.sum()-    return means, variances, counts---def _shrink_deltas_eb(delta, var_last, n_last, var_prev, n_prev):-    noise_var = var_last / max(n_last, 1) + var_prev / max(n_prev, 1)-    signal = delta ** 2-    denom = signal + noise_var + 1e-10-    shrink = signal / denom-    return delta * shrink-- def main() -> None:     parser = argparse.ArgumentParser()     parser.add_argument("--data", required=True)@@ -70,43 +47,40 @@ def main() -> None:     manifest = load_manifest(args.data)     genes = panel_genes(args.data, manifest)     stages = inputs_by_time(manifest, include_external=False)+    if not stages:+        stages = inputs_by_time(manifest)     last = read_stage(args.data, stages[-1], genes)     rng = np.random.default_rng(args.seed)     rows = sample_rows(last.n_obs, target_n_cells(manifest, last.n_obs), rng)-    X = last.X[rows]-    if hasattr(X, "toarray"):-        X = X.toarray()-    X = np.asarray(X, dtype=np.float32)+    X = _to_dense(last.X[rows])      if len(stages) >= 2:         prev = read_stage(args.data, stages[-2], genes)         labels_last = labels_of(last)         labels_prev = labels_of(prev)+        labels_sampled = labels_last[rows] -        means_last, vars_last, counts_last = _per_type_stats(last.X, labels_last)-        means_prev, vars_prev, counts_prev = _per_type_stats(prev.X, labels_prev)+        Xd_last = _to_dense(last.X)+        Xd_prev = _to_dense(prev.X)         del prev -        deltas = {}-        for t in means_last:-            if t in means_prev:-                raw_delta = means_last[t] - means_prev[t]-                shrunk = _shrink_deltas_eb(-                    raw_delta,-                    vars_last[t], counts_last[t],-                    vars_prev[t], counts_prev[t],-                )-                deltas[t] = shrunk * ALPHA_GLOBAL-            else:-                deltas[t] = np.zeros(len(genes), dtype=np.float32)+        types_last = np.unique(labels_last)+        types_prev_set = set(np.unique(labels_prev))++        means_last = {}+        for t in types_last:+            means_last[t] = Xd_last[labels_last == t].mean(axis=0)+        means_prev = {}+        for t in types_prev_set:+            means_prev[t] = Xd_prev[labels_prev == t].mean(axis=0)+        del Xd_last, Xd_prev -        labels_sampled = labels_last[rows]-        delta_matrix = np.zeros((X.shape[0], X.shape[1]), dtype=np.float32)         for t in np.unique(labels_sampled):             mask = labels_sampled == t-            if t in deltas:-                delta_matrix[mask] = deltas[t]-        X = X + delta_matrix+            if t in means_last and t in means_prev:+                delta = (means_last[t] - means_prev[t]) * ALPHA_GLOBAL+                X[mask] += delta+         np.clip(X, 0.0, None, out=X)      write_prediction(X, genes, args.out, seed=args.seed)

调研来源?调研员查到并用到的知识条目和文献检索结果（只列标题和编号）。

没有记录调研来源。

分析结果?分析员写的 ANALYSIS.json：改了什么、各组分数怎么变、假设是否成立、经验和下一步建议。

对话摘要?每个角色和大模型对话的统计：轮数、工具调用、用时、token 数和最后的回答摘录；原始记录只给路径。

只给统计和最后回答的摘录；完整对话请到原始记录位置里列出的文件看。

分析员

ANALYSIS.json written

native

stream

<<<<<<< SEARCH
ALPHA_GLOBAL = 0.5
=======
ALPHA_GLOBAL = 0.3
>>>>>>> REPLACE

<<<<<<< SEARCH
        type_stds_before = {}
        for t in np.unique(labels_sampled):
            mask = labels_sampled == t
            if mask.sum() >= 2:
                type_stds_before[t] = X[mask].std(axis=0)

        for t in np.unique(labels_sampled):
            mask = labels_sampled == t
            if t in means_last and t in means_prev:
                delta = (means_last[t] - means_prev[t]) * ALPHA_GLOBAL
                X[mask] += delta

        np.clip(X, 0.0, None, out=X)

        for t in np.unique(labels_sampled):
            mask = labels_sampled == t
            if mask.sum() < 2 or t not in type_stds_before:
                continue
            curr_mean = X[mask].mean(axis=0)
            curr_std = X[mask].std(axis=0) + 1e-8
            scale = type_stds_before[t] / curr_std
            X[mask] = curr_mean + (X[mask] - curr_mean) * scale

        np.clip(X, 0.0, None, out=X)
=======
        for t in np.unique(labels_sampled):
            mask = labels_sampled == t
            if t in means_last and t in means_prev:
                delta = (means_last[t] - means_prev[t]) * ALPHA_GLOBAL
                X[mask] += delta

        np.clip(X, 0.0, None, out=X)
>>>>>>> REPLACE