总览 · ← 返回运行 20261002-202908-search-t1-scr-C

节点 n7

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 82ed7164e0: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 5c87608489。改动的文件：solution/METHOD.md +7 −0、solution/run.py +61 −4

diff --git a/solution/METHOD.md b/solution/METHOD.mdnew file mode 100644index 0000000..a1551e5--- /dev/null+++ b/solution/METHOD.md@@ -0,0 +1,7 @@+## 改了什么+相对父节点（copy_last），改为投影加权位移：当有两个输入阶段时，对同时出现在两个阶段的类型，计算伪批量差值delta，然后对每个细胞计算其在delta方向上的投影（相对类型均值），标准化后按投影值比例沿delta方向位移。关键特性：(1)位移均值为零（投影标准化后均值为0），不改变类型均值，因此不改变类型间均值差异结构；(2)协方差变化是秩1的（仅在delta方向增加方差），对整体基因共变结构影响远小于均匀位移；(3)沿变化方向拉伸分布，有助于direction指标。alpha=0.05，上限0.1。单输入退化为copy_last。第0轮均匀位移alpha=0.1导致covariation从48.44降至34.5（-13.94），原因是均匀位移改变了类型间均值差异，引入全秩协方差变化。本轮用零均值投影位移避免此问题。+## 用到的知识与出处+- 方法卡 k018：常数位移需收缩系数α∈[0,1]+- 方法卡 k012：pseudobulk_shift定义，clip≥0+- 第0轮实验反馈：均匀位移严重破坏covariation（48.44→34.5），需要保持协方差结构的位移方式+- 线性代数：协方差对平移不变，秩1扰动对整体协方差影响小于全秩扰动diff --git a/solution/run.py b/solution/run.pyindex aa4bbf0..2c889b5 100644--- a/solution/run.py+++ b/solution/run.py@@ -1,11 +1,12 @@ #!/usr/bin/env python3-"""copy_last: the latest input stage, subsampled into the board's cell range."""+"""Projection-weighted shift: cells shifted along delta proportional to their position, preserving mean."""  from __future__ import annotations  import argparse  import numpy as np+import scipy.sparse as sp  from src.task1_temporal.view_io import (     inputs_by_time,@@ -17,6 +18,9 @@ from src.task1_temporal.view_io import (     write_prediction, ) +ALPHA = 0.05+MAX_ALPHA = 0.1+  def main() -> None:     parser = argparse.ArgumentParser()@@ -27,10 +31,63 @@ def main() -> None:      manifest = load_manifest(args.data)     genes = panel_genes(args.data, manifest)-    last = read_stage(args.data, inputs_by_time(manifest)[-1], genes)+    entries = inputs_by_time(manifest)+    last = read_stage(args.data, entries[-1], genes)     rng = np.random.default_rng(args.seed)-    rows = sample_rows(last.n_obs, target_n_cells(manifest, last.n_obs), rng)-    write_prediction(last.X[rows], genes, args.out, seed=args.seed)++    if len(entries) < 2:+        rows = sample_rows(last.n_obs, target_n_cells(manifest, last.n_obs), rng)+        write_prediction(last.X[rows], genes, args.out, seed=args.seed)+        return++    prev = read_stage(args.data, entries[-2], genes)++    prev_labels = prev.obs["celltype"].values+    last_labels = last.obs["celltype"].values++    X_last = last.X+    if sp.issparse(X_last):+        X_out = X_last.toarray().astype(np.float32)+    else:+        X_out = np.array(X_last, dtype=np.float32)++    dt_total = entries[-1]["time"] - entries[-2]["time"]+    dt_target = manifest["target"]["time"] - entries[-1]["time"]+    if dt_total > 0:+        alpha = min(ALPHA * (dt_target / dt_total), MAX_ALPHA)+    else:+        alpha = 0.0++    if alpha > 0:+        for ct in set(last_labels) & set(prev_labels):+            mask_prev = prev_labels == ct+            mask_last = last_labels == ct+            n_prev = int(mask_prev.sum())+            n_last = int(mask_last.sum())+            if n_prev < 3 or n_last == 0:+                continue+            X_prev_ct = prev.X[mask_prev]+            if sp.issparse(X_prev_ct):+                mean_prev = np.asarray(X_prev_ct.mean(axis=0)).flatten().astype(np.float32)+            else:+                mean_prev = np.asarray(X_prev_ct, dtype=np.float32).mean(axis=0)+            X_ct = X_out[mask_last]+            mean_last = X_ct.mean(axis=0)+            delta = mean_last - mean_prev+            d_sq = float(np.dot(delta, delta))+            if d_sq < 1e-10:+                continue+            centered = X_ct - mean_last+            proj = centered @ delta / d_sq+            p_std = proj.std()+            if p_std < 1e-10:+                continue+            proj = (proj - proj.mean()) / p_std+            X_out[mask_last] = np.clip(X_ct + alpha * np.outer(proj, delta), 0, None)++    n_target = target_n_cells(manifest, last.n_obs)+    rows = sample_rows(last.n_obs, n_target, rng)+    write_prediction(X_out[rows], genes, args.out, seed=args.seed)   if __name__ == "__main__":

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

Looking at the results, even a uniform shift with alpha=0.1 devastates covariation (48.44→34.5). The issue is likely that shifting creates non-zero values where there were biological zeros, disrupting the sparsity/covariance structure. I'll try an extremely conservative approach: only apply positive deltas (avoiding creation of artificial expression from zeros) with a very small alpha.

<<<<<<< SEARCH
#!/usr/bin/env python3
"""Mild per-type uniform shift: small alpha, no per-gene weighting to preserve covariation."""

from __future__ import annotations

import argparse

import numpy as np
import scipy.sparse as sp

from src.task1_temporal.view_io import (
    inputs_by_time,
    load_manifest,
    panel_genes,
    read_stage,
    sample_rows,
    target_n_cells,
    write_prediction,
)

ALPHA = 0.1
MAX_ALPHA = 0.15


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)
    entries = inputs_by_time(manifest)
    last = read_stage(args.data, entries[-1], genes)
    rng = np.random.default_rng(args.seed)