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

节点 n2

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 58fe00b8cb: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 5d0b878932。改动的文件：solution/METHOD.md +7 −0、solution/run.py +27 −5

diff --git a/solution/METHOD.md b/solution/METHOD.mdnew file mode 100644index 0000000..8f8ad72--- /dev/null+++ b/solution/METHOD.md@@ -0,0 +1,7 @@+## 改了什么+1. 用 `inputs_by_time(manifest, include_external=False)` 过滤外部输入阶段。父节点在 proxy2 视图上把 Qiu E9.0（不同细胞类型标签、不同基因覆盖）当作最新阶段来抽样和计算 delta，导致 proxy2 分数极低（27.43）。过滤后退化为与 proxy 相同的 copy_last 行为，预期 proxy2 大幅提升。+2. 用按细胞类型分层抽样（`_stratified_sample`）替换均匀随机抽样（`sample_rows`）。保持各类型的原始比例，使输出群体的组成和基因间共变结构更接近输入，预期改善 covariation 和 cell_state 分组得分。+## 用到的知识与出处+- 任务书 CONTRACT 中关于 `inputs_by_time(manifest, include_external=False)` 的说明（proxy2 外部输入阶段处理）。+- 方法卡 k018：proxy 只有一个输入阶段时退化为 copy_last 是预期行为。+- 分层抽样保持组成是标准统计方法，无特定文献。diff --git a/solution/run.py b/solution/run.pyindex f3a0f25..358519d 100644--- a/solution/run.py+++ b/solution/run.py@@ -23,12 +23,32 @@ from src.task1_temporal.view_io import (     load_manifest,     panel_genes,     read_stage,-    sample_rows,     target_n_cells,     write_prediction, )  +def _stratified_sample(labels: np.ndarray, n_target: int, rng: np.random.Generator) -> np.ndarray:+    unique, counts = np.unique(labels, return_counts=True)+    target_counts = np.maximum(1, np.round(counts / labels.shape[0] * n_target).astype(int))+    diff = n_target - int(target_counts.sum())+    step = 1 if diff > 0 else -1+    order = np.argsort(-counts)+    for i in range(abs(diff)):+        target_counts[order[i % len(order)]] += step+    rows = []+    for lab, tc in zip(unique, target_counts):+        idx = np.where(labels == lab)[0]+        tc = int(tc)+        if tc <= len(idx):+            rows.append(rng.choice(idx, size=tc, replace=False))+        else:+            rows.append(rng.choice(idx, size=tc, replace=True))+    out = np.concatenate(rows)+    rng.shuffle(out)+    return out++ def main() -> None:     parser = argparse.ArgumentParser()     parser.add_argument("--data", required=True)@@ -38,16 +58,18 @@ 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)     last = read_stage(args.data, stages[-1], genes)+    labels = labels_of(last)     rng = np.random.default_rng(args.seed)-    rows = sample_rows(last.n_obs, target_n_cells(manifest, last.n_obs), rng)+    n_target = target_n_cells(manifest, last.n_obs)+    rows = _stratified_sample(labels, n_target, 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))+        deltas = type_deltas(prev.X, labels_of(prev), last.X, labels)         del prev-        X = shift_rows(X, labels_of(last)[rows], deltas)+        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 noise addition in round 1 was catastrophic — covariation dropped from 39.48 to 12.91, proxy from 49.40 to 41.11. The fix is to remove the noise entirely and revert to round 0's behavior (pure stratified copy for single-input case).

<<<<<<< SEARCH
def _add_within_type_noise(X, labels, rng, scale=0.15):
    X = X.toarray() if hasattr(X, "toarray") else np.asarray(X)
    X = X.astype(np.float32)
    unique = np.unique(labels)
    for lab in unique:
        mask = labels == lab
        block = X[mask]
        std = block.std(axis=0)
        std[std == 0] = np.median(std[std > 0]) if np.any(std > 0) else 1.0
        noise = rng.normal(0, 1, size=block.shape).astype(np.float32) * (std * scale)
        X[mask] = np.clip(block + noise, 0, None)
    return X
=======
>>>>>>> REPLACE

<<<<<<< SEARCH
    sampled_labels = labels[rows]
    if len(stages) >= 2:
        prev = read_stage(args.data, stages[-2], genes)
        deltas = type_deltas(prev.X, labels_of(prev), last.X, labels)
        del prev
        X = shift_rows(X, sampled_labels, deltas)
    else:
        X = _add_within_type_noise(X, sampled_labels, rng, scale=0.15)
    write_prediction(X, genes, args.out, seed=args.seed)
=======
    sampled_labels = labels[rows]
    if len(stages) >= 2:
        prev = read_stage(args.data, stages[-2], genes)
        deltas = type_deltas(prev.X, labels_of(prev), last.X, labels)
        del prev
        X = shift_rows(X, sampled_labels, deltas)
    write_prediction(X, genes, args.out, seed=args.seed)
>>>>>>> REPLACE

```METHOD
## 改了什么