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

节点 n3

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 17f3e11784: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 34ea38b45b。改动的文件：solution/METHOD.md +4 −0、solution/run.py +12 −9

diff --git a/solution/METHOD.md b/solution/METHOD.mdnew file mode 100644index 0000000..e9f77d1--- /dev/null+++ b/solution/METHOD.md@@ -0,0 +1,4 @@+## 改了什么+用 `inputs_by_time(manifest, include_external=False)` 过滤，只用官方阶段做抽样和差值计算的基底。当官方阶段为空时（如 X3 测试题全是外部输入）退回使用全部阶段。这修复了 proxy2 上的严重问题：父节点把 Qiu E9.0（心脏谱系、不同标签体系）当作最新阶段直接抽样输出，导致细胞类型和表达谱与官方 E9.5 真值严重不匹配，proxy2 只得 27.43。修复后 proxy2 退化为 copy_last（E8.5），预期接近 proxy 的 50 分水平。+## 用到的知识与出处+CONTRACT 中"外部输入阶段"一节明确说明 `inputs_by_time(manifest, include_external=False)` 可获取仅官方阶段；方法卡 k018 指出单输入时退化为 copy_last 是预期行为。diff --git a/solution/run.py b/solution/run.pyindex f3a0f25..dabccbd 100644--- a/solution/run.py+++ b/solution/run.py@@ -1,13 +1,10 @@ #!/usr/bin/env python3-"""pseudobulk_shift: latest stage + per-cell-type pseudobulk delta of the last step.+"""pseudobulk_shift with official-stage filtering. -The delta is mean(last|type) - mean(prev|type) over the two latest inputs,-computed on the full stages and added once to a subsample of the latest stage-(clipped at 0). Types missing from the earlier stage are copied unchanged.--With a single input stage (T1 proxy: E8.5 only) there is no step to take a-delta from, so this falls back to copy_last with the same sampling. The proxy-therefore cannot tell this seed from copy_last; that gap is expected.+Uses only official (non-external) stages for sampling and delta computation.+This fixes proxy2 where the latest input is Qiu E9.0 (different cell type+labels, heart-only subset) which must not be used as the prediction base.+Falls back to all stages when no official stages exist (e.g. X3 test). """  from __future__ import annotations@@ -38,16 +35,22 @@ def main() -> None:      manifest = load_manifest(args.data)     genes = panel_genes(args.data, manifest)-    stages = inputs_by_time(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 len(stages) >= 2:         prev = read_stage(args.data, stages[-2], genes)         deltas = type_deltas(prev.X, labels_of(prev), last.X, labels_of(last))         del prev         X = shift_rows(X, labels_of(last)[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

审查员

REVIEW.json written

stream

<<<<<<< SEARCH
def main() -> None:
    parser = argparse.ArgumentParser()
    parser.add_argument("--data", required=True)
    parser.add_argument("--out", required=True)
    parser.add_argument("--seed", type=int, default=0)
    args = parser.parse_args()

    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 len(stages) >= 2:
        prev = read_stage(args.data, stages[-2], genes)
        deltas = type_deltas(prev.X, labels_of(prev), last.X, labels_of(last))
        del prev
        X = shift_rows(X, labels_of(last)[rows], deltas)

    write_prediction(X, genes, args.out, seed=args.seed)
=======
def stratified_sample(labels: np.ndarray, n_target: int, rng: np.random.Generator) -> np.ndarray:
    unique, counts = np.unique(labels, return_counts=True)
    proportions = counts / counts.sum()
    alloc = np.floor(proportions * n_target).astype(int)
    remainder = n_target - alloc.sum()
    if remainder > 0:
        frac = (proportions * n_target) - alloc
        top = np.argsort(-frac)[:remainder]
        alloc[top] += 1
    rows = []
    for ct, n_ct in zip(unique, alloc):
        idx = np.where(labels == ct)[0]