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

节点 n73

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 5c6b691c30: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：这个提交的上一版（种子程序：相对空仓库）。改动的文件：solution/EXECUTION.json +1 −0、solution/METHOD.md +15 −0、solution/README.md +4 −0、solution/g37_common.py +101 −0、solution/run.py +176 −0

diff --git a/solution/EXECUTION.json b/solution/EXECUTION.jsonnew file mode 100644index 0000000..9d5125c--- /dev/null+++ b/solution/EXECUTION.json@@ -0,0 +1 @@+{"gpu": false}diff --git a/solution/METHOD.md b/solution/METHOD.mdnew file mode 100644index 0000000..09b15c8--- /dev/null+++ b/solution/METHOD.md@@ -0,0 +1,15 @@+## 改了什么+相对父节点（ot_moscot 种子）的全部改动：+1. 保留原始重心位移步长（step_bary）作为主体，保持其在 covariation 和 de_recovery 上的优势。+2. 新增 OT-CFM 速度场（2 隐层 64 单元 MLP，1000 步训练），在 PCA 空间积分得到每细胞的位移，解码回基因空间（仅 HVG 基因），作为修正项。+3. 最终步长 = step_bary + 0.25 × step_cfm。混合权重 0.25 足够小，不会破坏基因间共变结构，但能提供状态相关的方向修正。+4. 第 0 轮纯 CFM 解码（无 barycentric 基底）导致 covariation/de_recovery 暴跌（只改 HVG）；第 1 轮用 CFM 模长比值做缩放太噪声（47.79）。本轮改为加性混合：barycentric 提供全基因结构，CFM 提供方向修正。+5. 模型简化为 2 隐层（减少过拟合），FM_SIGMA=0（无噪声路径），积分 5 步。++## 用到的知识与出处+- Tong et al., OT-CFM, arXiv:2302.00482：minibatch OT 配对训练条件流匹配。+- Lipman et al., Flow Matching, arXiv:2210.02747：流匹配框架。+- torchcfm 1.0.7（MIT）：ExactOptimalTransportConditionalFlowMatcher。+- 知识条目 k034、k009：PCA 空间工作、解码保持残差、两时间点外推需阻尼。+- Schiebinger et al., Cell 2019（WOT）：非平衡熵 OT 耦合、生长先验。+- 知识条目 k031：moscot TemporalProblem。diff --git a/solution/README.md b/solution/README.mdnew file mode 100644index 0000000..dc1cbaf--- /dev/null+++ b/solution/README.md@@ -0,0 +1,4 @@+# ot_moscot++Waddington-OT / moscot：两个最新输入阶段在联合 PCA 上做非平衡熵 OT 耦合（WOT 增殖/凋亡先验生长），最新阶段按 g^Δt 重抽样，每个细胞沿“kNN 平滑位置 − 耦合祖先均值”再走一个等比例步长，只加在非零基因上，夹到 ≥0。+来自 G37 候选 modeling/candidates/T1/ot_moscot；addnz 解码是在 X3 上选的（详见 METHOD.md）。纯 CPU，final 约 70 s、6.7 GB。diff --git a/solution/g37_common.py b/solution/g37_common.pynew file mode 100644index 0000000..8c991b3--- /dev/null+++ b/solution/g37_common.py@@ -0,0 +1,101 @@+"""Helpers of the ot_moscot seed (copy of modeling/candidates/T1/ot_moscot/g37_common.py, G37; stage_pair reads every+input stage of the view).++- ``stage_pair``: the two latest input stages (official, or external in proxy2 / test views) and the gene mask on+  which both are really measured (external stages lack part of the panel; filled columns must not drive dynamics);+- ``embed``: HVG -> z-score (clip 10) -> PCA, fitted on the input stages only (scanpy-style preprocessing);+- ``knn_mean``: per-cell kNN average in the embedding (smooths single-cell noise before taking displacements);+- ``growth_rates``: Waddington-OT / moscot prior growth from proliferation and apoptosis gene scores.+"""++from __future__ import annotations++import numpy as np+from scipy import sparse++from src.task1_temporal.view_io import covered_mask, inputs_by_time, read_stage+++def stage_pair(view, manifest: dict, genes: list[str]):+    """(prev, last, mask, dt_in) for the two latest inputs; prev None with a single input stage."""+    stages = inputs_by_time(manifest, include_external=True)  # every input stage, whatever its source+    last_e = stages[-1]+    last = read_stage(view, last_e, genes)+    mask = covered_mask(view, last_e, genes)+    if len(stages) < 2:+        return None, last, mask, None, last_e+    prev_e = stages[-2]+    prev = read_stage(view, prev_e, genes)+    mask &= covered_mask(view, prev_e, genes)+    return prev, last, mask, float(last_e["time"] - prev_e["time"]), last_e+++def _gene_moments(X: sparse.csr_matrix):+    n = X.shape[0]+    mean = np.asarray(X.mean(axis=0)).ravel()+    sq = np.asarray(X.multiply(X).sum(axis=0)).ravel() / n+    return mean, np.maximum(sq - mean ** 2, 0.0)+++def embed(mats: list, mask: np.ndarray, n_hvg: int, n_pcs: int, seed: int):+    """PCA coordinates of each matrix in ``mats`` (same order), fitted on all of them together.++    Returns (list of Z, info) with info = {hvg, mean, std, components} so displacements can be decoded."""+    from sklearn.decomposition import PCA++    allX = sparse.vstack(mats).tocsr()+    mean, var = _gene_moments(allX)+    var = np.where(mask, var, -1.0)+    hvg = np.sort(np.argsort(var)[::-1][:n_hvg])+    sub = allX[:, hvg].toarray().astype(np.float32)+    mu, sd = mean[hvg].astype(np.float32), np.sqrt(var[hvg]).astype(np.float32)+    sd[sd == 0] = 1.0+    sub -= mu+    sub /= sd+    np.clip(sub, -10, 10, out=sub)+    pca = PCA(n_components=n_pcs, svd_solver="randomized", random_state=seed)+    Z = pca.fit_transform(sub).astype(np.float32)+    out, start = [], 0+    for M in mats:+        out.append(Z[start:start + M.shape[0]])+        start += M.shape[0]+    return out, {"hvg": hvg, "mu": mu, "sd": sd, "components": pca.components_.astype(np.float32)}+++def knn_mean(Z: np.ndarray, X: sparse.csr_matrix, rows: np.ndarray, k: int) -> np.ndarray:+    """Dense (len(rows), n_genes): mean expression of the k nearest cells (in Z, the cell itself included)."""+    from sklearn.neighbors import NearestNeighbors++    nn = NearestNeighbors(n_neighbors=k).fit(Z)+    idx = nn.kneighbors(Z[rows], return_distance=False)+    W = sparse.csr_matrix((np.full(idx.size, 1.0 / k, dtype=np.float32), idx.ravel(),+                           np.arange(0, idx.size + 1, k)), shape=(len(rows), Z.shape[0]))+    return np.asarray((W @ X).todense(), dtype=np.float32)+++# Waddington-OT prior growth (Schiebinger et al. 2019, Cell; as implemented in moscot.base.problems.birth_death):+# birth = generalised logistic of the proliferation score, death = of the apoptosis score, g = exp(birth - death) / day.+def _gen_logistic(p, sup, inf, center, width):+    return inf + (sup - inf) / (1 + np.exp(-(p - center) / width))+++def growth_rates(adata_list: list, genes: list[str], mask: np.ndarray, seed: int) -> list[np.ndarray]:+    """Per-day growth rate of every cell of each AnnData (scores computed on all of them jointly)."""+    import anndata as ad+    import scanpy as sc+    from moscot.utils.data import apoptosis_markers, proliferation_markers++    ok = set(np.asarray(genes)[mask])+    joint = ad.concat(adata_list, join="inner")+    prolif = [g for g in proliferation_markers("mouse") if g in ok]+    apopt = [g for g in apoptosis_markers("mouse") if g in ok]+    sc.tl.score_genes(joint, prolif, score_name="proliferation", random_state=seed)+    sc.tl.score_genes(joint, apopt, score_name="apoptosis", random_state=seed)+    birth = _gen_logistic(joint.obs["proliferation"].to_numpy(float), 1.7, 0.3, 0.25, 0.5)+    death = _gen_logistic(joint.obs["apoptosis"].to_numpy(float), 1.7, 0.3, 0.1, 0.2)+    g = np.exp(birth - death)+    out, start = [], 0+    for a in adata_list:+        out.append(g[start:start + a.n_obs])+        start += a.n_obs+    return outdiff --git a/solution/run.py b/solution/run.pynew file mode 100644index 0000000..21c746a--- /dev/null+++ b/solution/run.py@@ -0,0 +1,176 @@+#!/usr/bin/env python3+"""ot_moscot: Waddington-OT / moscot TemporalProblem coupling, extrapolated one step past the latest input stage.++Two input stages (final: E8.5, E9.5):+  1. joint PCA of both stages (HVG, z-score, 30 PCs; fitted on the inputs only);+  2. moscot TemporalProblem prev -> last on the PCA, source marginals from Waddington-OT prior growth+     (proliferation / apoptosis gene scores), entropic unbalanced Sinkhorn (epsilon 1e-3, tau_a 0.95);+  3. output cells = latest-stage cells resampled with weights g^dt_out (prior growth rate continued for the target+     interval, Waddington-OT birth-death model);+  4. each output cell j moves by LAMBDA * dt_out / dt_in * (kNN-mean(x_j) - ancestor_mean_j) in gene space, where+     ancestor_mean_j is the coupling-weighted (barycentric) mean of its ancestors in the earlier stage: the last+     observed displacement continued for the target interval; the cell keeps its own residual. The step is applied+     to the cell's non-zero entries only (DECODE "addnz": a dense step turns every zero into a small positive value+     and wrecks cell_state / covariation, see METHOD.md). Clipped at 0.+One input stage (proxy: E8.5 only): steps 1, 2, 4 need two stages; only the growth resampling (3) runs, i.e.+a growth-weighted copy of the latest stage.++Seed version (agent/seeds/T1__val/ot_moscot, 2026-10-02) of modeling/candidates/T1/ot_moscot (G37): same method and+hyper-parameters; the dev-only environment overrides (G37_LAMBDA / G37_DECODE / G37_GROWTH / G37_JAX_GPU) and the+`knn` decode branch are removed; JAX and torch on CPU, fixed thread count (EXECUTION.json gpu false).+Parameter provenance (METHOD.md): DECODE = addnz was chosen on the X3 ruler (G37); LAMBDA = 1 a priori, also checked+on X3 (0.5 vs 1 within 0.2).+"""++from __future__ import annotations++import argparse+import os+import sys+from pathlib import Path++os.environ.setdefault("XLA_PYTHON_CLIENT_PREALLOCATE", "false")+os.environ.setdefault("XLA_PYTHON_CLIENT_MEM_FRACTION", "0.1")+os.environ["JAX_PLATFORMS"] = "cpu"  # CPU only: deterministic, no GPU slot++import numpy as np++sys.path.insert(0, str(Path(__file__).resolve().parent))+from g37_common import embed, growth_rates, knn_mean, stage_pair  # noqa: E402+from torchcfm.conditional_flow_matching import ExactOptimalTransportConditionalFlowMatcher  # noqa: E402++from src.task1_temporal.view_io import load_manifest, panel_genes, target_n_cells, write_prediction  # noqa: E402++N_HVG = 2000+N_PCS = 30+EPSILON = 1e-3+TAU_A = 0.95+TAU_B = 1.0+K_SMOOTH = 30+LAMBDA = 1.0+GROWTH = True+N_THREADS = 8+FM_HIDDEN = 64+FM_STEPS = 1000+FM_LR = 1e-3+FM_INTEGRATE_STEPS = 5+FM_SIGMA = 0.0+FM_BLEND = 0.25+++def weighted_rows(w: np.ndarray, n: int, rng: np.random.Generator) -> np.ndarray:+    p = w / w.sum()+    return np.sort(rng.choice(len(w), size=n, replace=n > len(w), p=p))+++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)+    prev, last, mask, dt_in, last_e = stage_pair(args.data, manifest, genes)+    dt_out = float(manifest["target"]["time"] - last_e["time"])+    rng = np.random.default_rng(args.seed)+    n = target_n_cells(manifest, last.n_obs)++    if prev is None:+        (g_last,) = growth_rates([last], genes, mask, args.seed)+        rows = weighted_rows(g_last ** dt_out if GROWTH else np.ones(last.n_obs), n, rng)+        write_prediction(last.X[rows], genes, args.out, seed=args.seed)+        return++    import anndata as ad+    import pandas as pd+    import torch+    from moscot.problems.time import TemporalProblem++    (Zp, Zl), info = embed([prev.X, last.X], mask, N_HVG, N_PCS, args.seed)+    g_prev, g_last = growth_rates([prev, last], genes, mask, args.seed)++    obs = pd.DataFrame({"time": np.r_[np.zeros(prev.n_obs), np.ones(last.n_obs)],+                        "growth": np.r_[g_prev ** dt_in, g_last ** dt_in]},+                       index=[f"p{i}" for i in range(prev.n_obs)] + [f"l{i}" for i in range(last.n_obs)])+    obs["time"] = obs["time"].astype(float)+    small = ad.AnnData(X=np.zeros((len(obs), 1), dtype=np.float32), obs=obs)+    small.obsm["X_pca"] = np.vstack([Zp, Zl])+    tp = TemporalProblem(small)+    # source marginals = WOT prior growth over the input interval (normalised by moscot); target uniform+    tp = tp.prepare(time_key="time", joint_attr="X_pca", a="growth")+    tp = tp.solve(epsilon=EPSILON, tau_a=TAU_A, tau_b=TAU_B, scale_cost="mean")+    sol = tp[(0.0, 1.0)].solution+    print(f"ot: converged={getattr(sol, 'converged', None)} cost={getattr(sol, 'cost', None)}", file=sys.stderr)+    P = np.asarray(sol.transport_matrix, dtype=np.float32)  # (n_prev, n_last)++    rows = weighted_rows(g_last ** dt_out if GROWTH else np.ones(last.n_obs), n, rng)+    print(f"diag: growth last min/median/max {g_last.min():.3f}/{np.median(g_last):.3f}/{g_last.max():.3f}", file=sys.stderr)+    Pc = P[:, rows]+    del P+    Pc /= np.maximum(Pc.sum(axis=0, keepdims=True), 1e-30)+    factor = LAMBDA * dt_out / dt_in+    torch.set_num_threads(N_THREADS)+    torch.manual_seed(args.seed)+    Xp = torch.from_numpy(prev.X.toarray())+    anc = (torch.from_numpy(Pc).T @ Xp).numpy()+    del Xp, Pc+    smooth = knn_mean(Zl, last.X, rows, K_SMOOTH)+    step_bary = (smooth - anc) * factor+    del smooth, anc+    step_bary[:, ~mask] = 0.0++    fm = ExactOptimalTransportConditionalFlowMatcher(sigma=FM_SIGMA)+    model = torch.nn.Sequential(+        torch.nn.Linear(N_PCS + 1, FM_HIDDEN),+        torch.nn.SiLU(),+        torch.nn.Linear(FM_HIDDEN, FM_HIDDEN),+        torch.nn.SiLU(),+        torch.nn.Linear(FM_HIDDEN, N_PCS),+    )+    opt = torch.optim.Adam(model.parameters(), lr=FM_LR)++    Zp_t = torch.from_numpy(Zp)+    Zl_t = torch.from_numpy(Zl)+    bs = min(256, Zp_t.shape[0], Zl_t.shape[0])+    for _ in range(FM_STEPS):+        idx_p = torch.randint(0, Zp_t.shape[0], (bs,))+        idx_l = torch.randint(0, Zl_t.shape[0], (bs,))+        x0 = Zp_t[idx_p]+        x1 = Zl_t[idx_l]+        t, xt, ut = fm.sample_location_and_conditional_flow(x0, x1)+        vt = model(torch.cat([xt, t.unsqueeze(1)], dim=1))+        loss = torch.mean((vt - ut) ** 2)+        opt.zero_grad()+        loss.backward()+        opt.step()++    with torch.no_grad():+        z_start = torch.from_numpy(Zl[rows])+        z_cur = z_start.clone()+        dt_step = (dt_out / dt_in) / FM_INTEGRATE_STEPS+        for i in range(FM_INTEGRATE_STEPS):+            t_val = torch.full((z_cur.shape[0], 1), 1.0 + i * dt_step)+            v = model(torch.cat([z_cur, t_val], dim=1))+            z_cur = z_cur + dt_step * v+    z_delta_cfm = (z_cur - z_start).numpy()++    components = info["components"]+    sd = info["sd"]+    hvg = info["hvg"]+    delta_hvg = (z_delta_cfm @ components) * sd+    step_cfm = np.zeros((len(rows), len(genes)), dtype=np.float32)+    step_cfm[:, hvg] = delta_hvg+    step_cfm[:, ~mask] = 0.0++    step = step_bary + FM_BLEND * step_cfm+    X = last.X[rows].toarray()+    step *= X > 0+    X += step+    np.maximum(X, 0.0, out=X)+    write_prediction(X, genes, args.out, seed=args.seed)+++if __name__ == "__main__":+    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:
- Original barycentric: 50.65 (best covariation 52.52, de_recovery 51.38)
- Round 0 pure CFM: 49.71 (best cell_state 56.15)
- Round 1 scale modulation: 47.79 (worse than both)

The scale modulation approach failed because the norm ratio is too noisy. Instead, I'll **blend** the barycentric step with a small CFM-decoded correction. This preserves the gene-space structure while adding state-dependent directionality.

<<<<<<< SEARCH
N_HVG = 2000
N_PCS = 30
EPSILON = 1e-3
TAU_A = 0.95
TAU_B = 1.0
K_SMOOTH = 30
LAMBDA = 1.0     # 1 = continue the observed displacement at full rate
GROWTH = True    # resample output cells by g^dt_out (WOT birth-death model)
N_THREADS = 8
FM_HIDDEN = 128
FM_STEPS = 1500
FM_LR = 1e-3
FM_INTEGRATE_STEPS = 5
FM_SIGMA = 0.0
FM_SCALE_CLIP = (0.4, 2.0)
=======
N_HVG = 2000
N_PCS = 30
EPSILON = 1e-3
TAU_A = 0.95
TAU_B = 1.0
K_SMOOTH = 30
LAMBDA = 1.0
GROWTH = True
N_THREADS = 8
FM_HIDDEN = 64
FM_STEPS = 1000
FM_LR = 1e-3
FM_INTEGRATE_STEPS = 5
FM_SIGMA = 0.0
FM_BLEND = 0.25
>>>>>>> REPLACE