总览 · ← 返回运行 20261002-034201-search-t1-abc-r1-C-native

节点 n20 在终选来历上

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 6f4818572b: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 34254aa790。改动的文件：solution/METHOD.md +2 −2、solution/run.py +82 −10

diff --git a/solution/METHOD.md b/solution/METHOD.mdindex 58fac39..f58a678 100644--- a/solution/METHOD.md+++ b/solution/METHOD.md@@ -1,4 +1,4 @@ ## 改了什么-从 pseudobulk_shift 改为组成重加权（T1-01 方向）。核心变化：(1) 使用 include_external=False 避免 proxy2 中 Qiu E9.0 标签不匹配导致的崩溃（原程序 proxy2 仅 27.43）；(2) 按正则模式将细胞类型分为 heart（×1.6）、reduced（×0.15，含 surface ectoderm/EXEM/paraxial/neural）、gut（×0.9）、neutral（×1.0）四族，用加权抽样替代均匀抽样，不改表达值；(3) 对 X3 等无官方输入的视图有退路（回退到全部输入）。权重规则按名称模式定义，不写死特定阶段的名字。+相对父节点（组成重加权 + replace=True 加权抽样）：(1) 将加权有放回抽样改为分层无放回抽样：按家族权重调整各类型的目标比例，然后在每个类型内无放回抽样，保留类型内部的细胞分布和协方差结构（父节点 replace=True 会重复抽样导致协方差塌缩，covariation 从 51.96 降至 49.73）；(2) 保留逐基因 EB 收缩位移（alpha=0.03），在有两个官方输入阶段时施加；(3) 保留 reduced 族权重 0.15。核心思路：分层抽样保协方差（节点 14 验证），逐基因收缩保 de_recovery。 ## 用到的知识与出处-方向库 T1-01（种子 heart_jcf_peri：心脏类 ×1.6，表面外胚层/EXEM/轴旁中胚层 ×0.25，丢 Neural Tube；proxy 55.97）；方法卡 T1 标签节（E8.5 独有名应降权或丢弃）；k012（cell_state 权重 30%，组成匹配是关键得分项）。+实验表节点 14（分层抽样 + EB 收缩，covariation 50.79、de_recovery 51.33，综合 52.24 为全树最佳）；父节点 ANALYSIS（replace=True 加权抽样导致 covariation 下降）；k018（EB 收缩：按基因信噪比逐基因收缩）；方向库 T1-01（组成重加权）+ T1-02（伪批量位移 + 收缩）；方法卡（alpha 不能默认 1）。diff --git a/solution/run.py b/solution/run.pyindex 4710196..ca1af32 100644--- a/solution/run.py+++ b/solution/run.py@@ -15,7 +15,9 @@ import argparse import re  import numpy as np+from scipy import sparse as sp +from src.task1_temporal.baselines import shift_rows from src.task1_temporal.view_io import (     inputs_by_time,     labels_of,@@ -26,6 +28,8 @@ from src.task1_temporal.view_io import (     write_prediction, ) +ALPHA = 0.03+ _HEART_PAT = re.compile(     r"cm\b|cardi|heart|shf|endocard|oft\b|avc|ift\b|sv[- ]cm|v[- ]cm|"     r"proepicard|pericard|epicard|blood",@@ -51,6 +55,47 @@ def family_weight(name: str) -> float:     return 1.0  +def eb_shrunk_deltas(prev_X, prev_labels, last_X, last_labels, alpha):+    prev_types = set(np.unique(prev_labels))+    last_types = np.unique(last_labels)+    deltas = {}+    for t in last_types:+        last_mask = last_labels == t+        n_last = int(last_mask.sum())+        if n_last == 0:+            continue+        last_sub = last_X[last_mask]+        if sp.issparse(last_sub):+            mean_last = np.asarray(last_sub.mean(axis=0)).ravel()+            var_last = np.asarray(last_sub.multiply(last_sub).mean(axis=0)).ravel() - mean_last**2+        else:+            arr = np.asarray(last_sub)+            mean_last = arr.mean(axis=0).ravel()+            var_last = arr.var(axis=0).ravel()+        var_last = np.maximum(var_last, 0.0)+        if t in prev_types:+            prev_mask = prev_labels == t+            n_prev = int(prev_mask.sum())+            prev_sub = prev_X[prev_mask]+            if sp.issparse(prev_sub):+                mean_prev = np.asarray(prev_sub.mean(axis=0)).ravel()+                var_prev = np.asarray(prev_sub.multiply(prev_sub).mean(axis=0)).ravel() - mean_prev**2+            else:+                arr_p = np.asarray(prev_sub)+                mean_prev = arr_p.mean(axis=0).ravel()+                var_prev = arr_p.var(axis=0).ravel()+            var_prev = np.maximum(var_prev, 0.0)+            delta = mean_last - mean_prev+            se2 = var_prev / max(n_prev, 1) + var_last / max(n_last, 1)+            n_eff = min(n_prev, n_last)+            size_factor = n_eff / (n_eff + 50.0)+            shrink = delta**2 / (delta**2 + se2 + 1e-10)+            deltas[t] = alpha * size_factor * shrink * delta+        else:+            deltas[t] = np.zeros(last_X.shape[1])+    return deltas++ def main() -> None:     parser = argparse.ArgumentParser()     parser.add_argument("--data", required=True)@@ -61,22 +106,49 @@ 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)+    stages = inputs_by_time(manifest)+    official = [s for s in stages if s.get("source") != "external"]+    if not official:+        official = stages -    last = read_stage(args.data, stages[-1], genes)+    last = read_stage(args.data, official[-1], genes)     labels = labels_of(last)     rng = np.random.default_rng(args.seed)     n_target = target_n_cells(manifest, last.n_obs) -    weights = np.array([family_weight(str(l)) for l in labels], dtype=np.float64)-    if weights.sum() == 0:-        weights[:] = 1.0-    weights /= weights.sum()--    rows = rng.choice(last.n_obs, size=n_target, replace=True, p=weights)+    unique_labels, counts = np.unique(labels, return_counts=True)+    n_total = len(labels)+    raw_w = np.array([family_weight(str(u)) for u in unique_labels], dtype=np.float64)+    target_props = (counts / n_total) * raw_w+    target_props /= target_props.sum()+    n_per_type = np.floor(target_props * n_target).astype(int)+    remainder = n_target - n_per_type.sum()+    frac = (target_props * n_target) - n_per_type+    top_idx = np.argsort(-frac)[:remainder]+    n_per_type[top_idx] += 1+    n_per_type = np.minimum(n_per_type, counts)+    indices = []+    for u, n_take in zip(unique_labels, n_per_type):+        if n_take <= 0:+            continue+        type_idx = np.where(labels == u)[0]+        if n_take >= len(type_idx):+            indices.append(type_idx)+        else:+            indices.append(rng.choice(type_idx, size=n_take, replace=False))+    rows = np.concatenate(indices)+    if len(rows) < n_target:+        remaining = np.setdiff1d(np.arange(n_total), rows)+        extra = rng.choice(remaining, size=n_target - len(rows), replace=False)+        rows = np.concatenate([rows, extra])     X = last.X[rows]++    if len(official) >= 2:+        prev = read_stage(args.data, official[-2], genes)+        deltas = eb_shrunk_deltas(prev.X, labels_of(prev), last.X, labels, ALPHA)+        del prev+        X = shift_rows(X, labels[rows], deltas)+     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

The main issue is that `rng.choice(..., replace=True)` with weighted probabilities distorts within-type covariance structure. Node 14 (best, 52.24) uses stratified sampling that preserves within-type distributions while still applying family weights to composition. I'll switch to stratified sampling with family-weighted target proportions, keeping the per-gene EB shrink.

<<<<<<< SEARCH
    weights = np.array([family_weight(str(l)) for l in labels], dtype=np.float64)
    if weights.sum() == 0:
        weights[:] = 1.0
    weights /= weights.sum()

    rows = rng.choice(last.n_obs, size=n_target, replace=True, p=weights)
    X = last.X[rows]
=======
    unique_labels, counts = np.unique(labels, return_counts=True)
    n_total = len(labels)
    raw_w = np.array([family_weight(str(u)) for u in unique_labels], dtype=np.float64)
    target_props = (counts / n_total) * raw_w
    target_props /= target_props.sum()
    n_per_type = np.floor(target_props * n_target).astype(int)
    remainder = n_target - n_per_type.sum()
    frac = (target_props * n_target) - n_per_type
    top_idx = np.argsort(-frac)[:remainder]
    n_per_type[top_idx] += 1
    n_per_type = np.minimum(n_per_type, counts)
    indices = []
    for u, n_take in zip(unique_labels, n_per_type):
        if n_take <= 0:
            continue
        type_idx = np.where(labels == u)[0]
        if n_take >= len(type_idx):
            indices.append(type_idx)
        else:
            indices.append(rng.choice(type_idx, size=n_take, replace=False))
    rows = np.concatenate(indices)
    if len(rows) < n_target:
        remaining = np.setdiff1d(np.arange(n_total), rows)
        extra = rng.choice(remaining, size=n_target - len(rows), replace=False)
        rows = np.concatenate([rows, extra])
    X = last.X[rows]
>>>>>>> REPLACE

```METHOD