@barivia/barsom-mcp 0.15.4 → 0.17.1

package/dist/tools/results.js

@@ -4,7 +4,7 @@ import { z } from "zod";
 import { registerAuditedTool } from "../audit.js";
 import { apiCall, apiRawCall, getWorkspaceRootAsync, sandboxPath, pollUntilComplete, tryAttachImage, getResultsImagesToFetch, getCaptionForImage, mimeForFilename, } from "../shared.js";
 export function registerResultsTool(server) {
-    registerAuditedTool(server, "results", `Retrieve, recolor, download, export, or run temporal flow on a completed map job.
+    registerAuditedTool(server, "results", `Retrieve, recolor, download, or export figures and metrics for a completed map job. Presentation and artifact access only — operations on the frozen map (predict, compare, project, impute_column, transition_flow) live in **inference**.
 
 | Action | Use when | Sync/Async |
 |--------|----------|------------|
@@ -12,7 +12,8 @@ export function registerResultsTool(server) {
 | export | Learning curve, raw weights, or per-node stats | instant |
 | download | Saving figures to a local folder | instant |
 | recolor | Changing colormap or output format without retraining | async (~10–30s) |
-| transition_flow | Temporal dynamics on time-ordered data | async (~30–60s) |
+
+Temporal state-transition arrows moved to **inference(action=transition_flow)** (use the new result job_id with results(get/download) to read the figure).
 
 ONLY call this after jobs(action=status) returns "completed".
 ESCALATION: If job not found, verify job_id. If "job not complete", poll with jobs(action=status).
@@ -38,15 +39,10 @@ action=recolor: Change colormap or output format — no retraining. Returns a ne
   AFTER: use results(action=get, job_id=NEW_JOB_ID).
   Colormaps: viridis, plasma, inferno, magma, cividis, turbo, coolwarm, balance, hsv, twilight, RdBu, Spectral. Also: thermal, hot, grays, RdYlBu, delta, curl, phase, rainbow, tol_bright, tol_muted.
 
-action=transition_flow: Temporal state transition arrows on the map grid. Only use when rows are in chronological order. If your data has no temporal ordering, transition_flow will still run but the arrows are meaningless — use for time-ordered data only.
-  - You pass the training job_id; the API returns a new result job_id. Use that new job_id for results(get) or results(download).
-  - To persist the figure: results(action=download, job_id=<returned_job_id>).
-  - lag=1 (default): immediate next-step | lag=N: N-step horizon (e.g. 24 for daily cycles in hourly data).
-  - min_transitions: filter noisy arrows. Increase for large datasets.
-NOT FOR: Jobs that haven't completed. Use jobs(action=status) to check first.`, {
+NOT FOR: Jobs that haven't completed (use jobs(action=status) first), or temporal flow (use inference(action=transition_flow)).`, {
         action: z
-            .enum(["get", "recolor", "download", "export", "transition_flow"])
-            .describe("get: inline results + metrics; recolor: new colormap/format (async); download: save to disk; export: structured data (training_log/weights/nodes); transition_flow: temporal dynamics (async)"),
+            .enum(["get", "recolor", "download", "export"])
+            .describe("get: inline results + metrics; recolor: new colormap/format (async); download: save to disk; export: structured data (training_log/weights/nodes). Temporal flow moved to inference(action=transition_flow)."),
         job_id: z.string().describe("Job ID of a completed job"),
         figures: z
             .union([z.enum(["default", "combined_only", "all", "images", "none"]), z.array(z.string())])
@@ -69,12 +65,12 @@ NOT FOR: Jobs that haven't completed. Use jobs(action=status) to check first.`,
         colormap: z
             .string()
             .optional()
-            .describe("action=recolor: colormap name (default: coolwarm). action=transition_flow: U-matrix background colormap (default: grays). Examples: viridis, plasma, balance, hsv, twilight, RdBu, phase, rainbow."),
+            .describe("action=recolor: colormap name (default: coolwarm). Examples: viridis, plasma, balance, hsv, twilight, RdBu, phase, rainbow."),
         output_format: z
             .enum(["png", "pdf", "svg"])
             .optional()
             .default("png")
-            .describe("action=recolor / transition_flow: output image format (default: png)"),
+            .describe("action=recolor: output image format (default: png)"),
         output_dpi: z
             .enum(["standard", "retina", "print"])
             .optional()
@@ -88,19 +84,7 @@ NOT FOR: Jobs that haven't completed. Use jobs(action=status) to check first.`,
             .enum(["training_log", "weights", "nodes", "clusters", "transition_matrix"])
             .optional()
             .describe("action=export: training_log=learning curve; weights=weight matrix; nodes=per-node stats; clusters=regime boundaries; transition_matrix=Markov transition matrix"),
-        lag: z
-            .number().int().min(1)
-            .optional().default(1)
-            .describe("action=transition_flow: step lag (default 1 = consecutive rows). Use larger for periodic analysis (e.g. 24 for daily in hourly data)."),
-        min_transitions: z
-            .number().int().min(1)
-            .optional()
-            .describe("action=transition_flow: minimum transition count to draw an arrow (default: auto). Increase to filter noise."),
-        top_k: z
-            .number().int().min(1)
-            .optional().default(10)
-            .describe("action=transition_flow: number of top-flow nodes in statistics (default 10)"),
-    }, async ({ action, job_id, figures, include_individual, include_json, folder, colormap, output_format, output_dpi, recolor_figures, export_type: exportType, lag, min_transitions, top_k }) => {
+    }, async ({ action, job_id, figures, include_individual, include_json, folder, colormap, output_format, output_dpi, recolor_figures, export_type: exportType }) => {
         const dpiMap = { standard: 1, retina: 2, print: 4 };
         if (action === "get") {
             const data = (await apiCall("GET", `/v1/results/${job_id}`));
@@ -261,7 +245,10 @@ NOT FOR: Jobs that haven't completed. Use jobs(action=status) to check first.`,
                     `Missing cells: ${imp.n_missing_cells ?? "?"} (${imp.missing_fraction !== undefined ? (Number(imp.missing_fraction) * 100).toFixed(1) + "%" : "?"})`,
                     `Map quality — QE: ${fmt(summary.quantization_error)} | TE: ${fmt(summary.topographic_error)} | EV: ${fmt(summary.explained_variance)}`,
                     `Observed-dimension QE: ${fmt(imp.observed_quantization_error)}`,
-                    ...(siom ? [`SIOM — utilization: ${fmt(siom.utilization)} | dead_fraction: ${fmt(siom.dead_fraction)} | siom_qe: ${fmt(siom.siom_qe)}`] : []),
+                    hitStats.active_node_fraction !== undefined
+                        ? `Grid BMU occupancy: ${(Number(hitStats.active_node_fraction) * 100).toFixed(1)}% active nodes on the map grid`
+                        : "",
+                    ...(siom ? [`SIOM layer coverage (distinct from grid dead nodes) — utilization: ${fmt(siom.utilization)} | dead_fraction: ${fmt(siom.dead_fraction)} | siom_qe: ${fmt(siom.siom_qe)}`] : []),
                     hitStats.grid_suggestion ? `Grid hint: ${String(hitStats.grid_suggestion)}` : "",
                     ...(policy.auto_rules_applied && Array.isArray(policy.auto_rules_applied) && policy.auto_rules_applied.length > 0
                         ? [`Auto policy: ${policy.auto_rules_applied.join(", ")} (${policy.viz_mode ?? "viz"}, metrics=${policy.quality_metrics ?? "?"})`]
@@ -334,15 +321,24 @@ NOT FOR: Jobs that haven't completed. Use jobs(action=status) to check first.`,
                     `  Status:        ${coverageStatus}${isDegenerate ? " *** NON-COMPETITIVE ***" : isWarning ? " * CAUTION *" : ""}`,
                     `  Diagnosis:     ${String(coverage.diagnosis ?? "No plain-language diagnosis available.")}`,
                     `  Utilization:   ${siom.utilization !== undefined ? Number(siom.utilization).toFixed(4) : "N/A"}`,
-                    `  Dead fraction: ${siom.dead_fraction !== undefined ? Number(siom.dead_fraction).toFixed(4) : "N/A"}`,
+                    `  SIOM dead fraction (layer coverage, not grid BMU hits): ${siom.dead_fraction !== undefined ? Number(siom.dead_fraction).toFixed(4) : "N/A"}`,
                     `  Occupied nodes:${hitStats.occupied_nodes !== undefined ? ` ${hitStats.occupied_nodes}` : " N/A"}`,
                     `  Top hit share: ${hitStats.top_hit_share !== undefined ? Number(hitStats.top_hit_share).toFixed(4) : "N/A"}`,
                     `  Gini:          ${siom.gini !== undefined ? Number(siom.gini).toFixed(4) : "N/A"}`,
                     `  Entropy:       ${siom.entropy !== undefined ? Number(siom.entropy).toFixed(4) : "N/A"}`,
+                    `  SIOM gamma:    ${siom.gamma !== undefined ? Number(siom.gamma).toFixed(3) : "N/A"} | decay: ${siom.decay !== undefined ? Number(siom.decay).toFixed(4) : "N/A"}`,
+                    ...(summary.elastic ? (() => {
+                        const el = summary.elastic;
+                        return [`  Elastic:       λ=${el.lambda !== undefined ? Number(el.lambda).toFixed(3) : "0"} μ=${el.mu !== undefined ? Number(el.mu).toFixed(3) : "0"} anchor=${el.anchor !== undefined ? Number(el.anchor).toFixed(3) : "0"}`];
+                    })() : []),
+                    ``,
+                    `Layout tuning: if Voronoi cells look crowded in weight space (high top_hit_share), reduce max_nodes and/or raise gamma/siom_decay before growth_interval. See training_guidance keys floop_voronoi_crowding, floop_siom_params, floop_elastic_params.`,
                     ``,
                     `Quality Metrics:`,
                     `  Quantization Error: ${summary.quantization_error !== undefined ? Number(summary.quantization_error).toFixed(4) : "N/A"}${qeNote}`,
                     `  Topographic Error:  ${summary.topographic_error !== undefined ? Number(summary.topographic_error).toFixed(4) : "N/A"}${teNote}`,
+                    "",
+                    "Choosing a map for presentation: Lower QE/TE often favors larger, denser graphs. For slides and interpretation, sparser graphs (fewer occupied nodes, lower top_hit_share) are often easier to read even when metrics rank them lower. Compare figures side-by-side before picking a winner from metrics alone.",
                     ...(Array.isArray(coverage.recommended_actions) && coverage.recommended_actions.length > 0
                         ? ["", "Recommended next steps:", ...coverage.recommended_actions.map((step) => `  - ${String(step)}`)]
                         : []),
@@ -371,7 +367,7 @@ NOT FOR: Jobs that haven't completed. Use jobs(action=status) to check first.`,
                 const nNodes = grid[0] * grid[1];
                 const hitStats = summary.hit_stats;
                 const hitLine = hitStats
-                    ? `  Dead nodes: ${hitStats.dead_nodes ?? "N/A"} of ${nNodes} (${nNodes ? ((Number(hitStats.dead_nodes) / nNodes) * 100).toFixed(1) : "0"}%) | Max hits: ${hitStats.max_hits ?? "N/A"} | Median: ${hitStats.median_hits ?? "N/A"}`
+                    ? `  Grid BMU occupancy: ${hitStats.dead_nodes ?? "N/A"} dead of ${nNodes} (${nNodes ? ((Number(hitStats.dead_nodes) / nNodes) * 100).toFixed(1) : "0"}% dead${hitStats.active_node_fraction !== undefined ? `, ${(Number(hitStats.active_node_fraction) * 100).toFixed(1)}% active` : ""}) | Max hits: ${hitStats.max_hits ?? "N/A"} | Median: ${hitStats.median_hits ?? "N/A"}`
                     : "";
                 const uRange = summary.u_matrix_range;
                 const uRangeLine = uRange && uRange.length >= 2 ? `  U-matrix range: ${fmt(uRange[0])} — ${fmt(uRange[1])}` : "";
@@ -399,6 +395,12 @@ NOT FOR: Jobs that haven't completed. Use jobs(action=status) to check first.`,
                     `Grid: ${grid[0]}×${grid[1]} | Features: ${summary.n_features ?? 0} | Samples: ${summary.n_samples ?? 0}`,
                     `Model: ${summary.model ?? "SOM"} | Epochs: ${epochStr}`,
                     `Periodic: ${summary.periodic ?? true} | Normalize: ${summary.normalize ?? "auto"}`,
+                    (() => {
+                        const audit = summary.normalization_audit;
+                        if (!audit?.length)
+                            return "";
+                        return `Normalization audit: ${audit.map((a) => `${String(a.feature)}: requested ${String(a.requested)}, applied ${String(a.applied)} (${String(a.reason)})`).join("; ")}`;
+                    })(),
                     summary.sigma_f !== undefined ? `Sigma_f: ${summary.sigma_f}` : "",
                     duration !== undefined ? `Training duration: ${duration}s` : "",
                     ``, `Quality Metrics:`,
@@ -409,14 +411,14 @@ NOT FOR: Jobs that haven't completed. Use jobs(action=status) to check first.`,
                     `  Davies-Bouldin:     ${fmt(summary.davies_bouldin)}      (lower is better)`,
                     `  Calinski-Harabasz:  ${fmt(summary.calinski_harabasz)}   (higher is better)`,
                     ordErrors && ordErrors.length > 0 ? `  Final ordering QE: ${ordErrors.at(-1)?.toFixed(4)} (use results(action=export, export_type=training_log) for full curve)` : "",
-                    ...(hitLine ? ["", "Hit / U-matrix:", hitLine] : []),
+                    ...(hitLine ? ["", "Hit / U-matrix (grid BMU occupancy):", hitLine] : []),
                     ...(uRangeLine ? [uRangeLine] : []),
                     ...(compLines ? [`Component value ranges:\n${compLines}`] : []),
                     ...(clusterLines ? ["", "Cluster map (defining features):", clusterLines] : []),
                     ...(topCorrLine ? ["", topCorrLine] : []),
-                    ...(siom ? ["", "SIOM coverage metrics:",
+                    ...(siom ? ["", "SIOM layer coverage (distinct from grid dead nodes):",
                         `  Utilization: ${siom.utilization !== undefined ? Number(siom.utilization).toFixed(4) : "N/A"}`,
-                        `  Dead fraction: ${siom.dead_fraction !== undefined ? Number(siom.dead_fraction).toFixed(4) : "N/A"}`,
+                        `  SIOM dead fraction: ${siom.dead_fraction !== undefined ? Number(siom.dead_fraction).toFixed(4) : "N/A"}`,
                         `  Gini: ${siom.gini !== undefined ? Number(siom.gini).toFixed(4) : "N/A"}`,
                         `  Entropy: ${siom.entropy !== undefined ? Number(siom.entropy).toFixed(4) : "N/A"}`] : []),
                     ``, `Features: ${features.join(", ")}`,
@@ -432,7 +434,7 @@ NOT FOR: Jobs that haven't completed. Use jobs(action=status) to check first.`,
                             return "";
                         return `Columns not used in training: ${excluded.join(", ")}. You can project them onto this map with inference(action=project_columns, job_id=${job_id}, dataset_id=<dataset_id>, columns=[${excluded.map((c) => `"${c}"`).join(", ")}]) to see how they distribute across the topology. If you ran datasets(action=analyze) before training, any columns it recommended as "project later" are especially good candidates.`;
                     })(),
-                    ...((jobType === "train_som" || jobType === "train_siom" || jobType === "train_impute") ? ["", `Next: results(action=export, export_type=training_log) for learning curve; results(action=download) to save figures${jobType === "train_impute" ? " (includes imputed.csv)" : ""}; jobs(action=compare, job_ids=[...]) to compare runs; inference(action=predict) to score new data; inference(action=project_columns) to project other variables onto the map.`] : []),
+                    ...((jobType === "train_som" || jobType === "train_siom" || jobType === "train_impute") ? ["", `Next: results_explorer(job_id="${job_id}") to browse all figures interactively; results(action=export, export_type=training_log) for learning curve; results(action=download) to save figures${jobType === "train_impute" ? " (includes imputed.csv)" : ""}; jobs(action=compare, job_ids=[...]) to compare runs; inference(action=predict) to score new data; inference(action=project_columns) to project other variables onto the map.`] : []),
                 ].filter((l) => l !== "").join("\n");
                 content.push({ type: "text", text: textSummary });
                 const imagesToFetch = getResultsImagesToFetch(jobType, summary, figures, include_individual);
@@ -663,36 +665,6 @@ NOT FOR: Jobs that haven't completed. Use jobs(action=status) to check first.`,
             }
             return { content: [{ type: "text", text: `Recolor job ${newJobId} submitted. Poll with jobs(action=status, job_id="${newJobId}"), then results(action=get, job_id="${newJobId}").` }] };
         }
-        if (action === "transition_flow") {
-            const body = { lag: lag ?? 1, output_format: output_format ?? "png" };
-            if (min_transitions !== undefined)
-                body.min_transitions = min_transitions;
-            if (top_k !== undefined)
-                body.top_k = top_k;
-            if (colormap !== undefined)
-                body.colormap = colormap;
-            if (output_dpi && output_dpi !== "retina")
-                body.output_dpi = dpiMap[output_dpi] ?? 2;
-            const data = (await apiCall("POST", `/v1/results/${job_id}/transition-flow`, body));
-            const flowJobId = data.id;
-            const poll = await pollUntilComplete(flowJobId, 120_000);
-            if (poll.status === "completed") {
-                const res = (await apiCall("GET", `/v1/results/${flowJobId}`));
-                const summary = (res.summary ?? {});
-                const stats = (summary.flow_stats ?? {});
-                const content = [{ type: "text", text: [
-                            `Transition Flow (job: ${flowJobId}) | Parent map: ${job_id} | Lag: ${lag ?? 1}`,
-                            `Active flow nodes: ${stats.active_flow_nodes ?? "N/A"} | Total transitions: ${stats.total_transitions ?? "N/A"}`,
-                            `Mean magnitude: ${stats.mean_magnitude !== undefined ? Number(stats.mean_magnitude).toFixed(4) : "N/A"}`,
-                        ].join("\n") }];
-                await tryAttachImage(content, flowJobId, `transition_flow_lag${lag ?? 1}.${output_format ?? "png"}`);
-                return { content };
-            }
-            if (poll.status === "failed") {
-                return { content: [{ type: "text", text: `Transition flow job ${flowJobId} failed: ${poll.error ?? "unknown error"}` }] };
-            }
-            return { content: [{ type: "text", text: `Transition flow job ${flowJobId} submitted. Poll with jobs(action=status, job_id="${flowJobId}"), retrieve with results(action=get, job_id="${flowJobId}").` }] };
-        }
         throw new Error("Invalid action");
     });
 }

package/dist/tools/train.js

@@ -2,53 +2,58 @@ import { z } from "zod";
 import { registerAuditedTool } from "../audit.js";
 import { apiCall, textResult } from "../shared.js";
 import { buildTrainSubmitExtras } from "../train_submit_message.js";
-import { TRAINING_INPUT_SCHEMA, buildTrainMapParams, fetchTrainingPresets, } from "./training_core.js";
+import { TRAINING_INPUT_SCHEMA, buildTrainMapParams, buildFloopParams, fetchTrainingPresets, } from "./training_core.js";
 export const TRAIN_DESCRIPTION = `Submit a self-organizing map training job. Returns a job_id; poll with jobs(action=status) and then results(action=get). All actions are async.
 
 | Action | Use when |
 |--------|----------|
 | map | Standard SOM on a fixed hex grid — complete-case numeric data (no NaNs in training columns). |
 | siom_map | Self-interacting map — same grid flow plus SIOM coverage control (gamma, siom_decay, penalty) to avoid dead nodes. |
-| impute | Sparse training data: trains a missing-tolerant map (accelerated missSOM) AND returns dense imputed.csv in one job. Plain numeric columns only. |
+| impute | Sparse training data: trains a missing-tolerant map (accelerated missSOM; SIOM missSOM when model=auto on som_siom+ or model=SIOM) AND returns dense imputed.csv in one job. Plain numeric columns only. Optional SIOM knobs: gamma, siom_decay, siom_penalty, penalty_alpha, reset_per_epoch. |
 | floop_siom | FLooP-SIOM — a growing node-budget manifold instead of a fixed grid; default topology=free (CHL). Requires a plan with all_algorithms (Premium/Enterprise). |
-| baseline_study | One-call heuristic baseline SOM (grid ~5·√n, MAD normalization) for a quick first map. |
+
+First quick map (no separate baseline action): use \`train(action=map, preset=quick)\` (or \`normalize=mad\` with a grid sized from the preview row count); see the \`baseline_explore\` preset in **training_guidance**.
 
 Parameter details (grid, epochs, batch, model, presets, normalize, categorical_features, backend, FLooP max_nodes/effort, SIOM geometry) live in the **training_guidance** tool / **prepare_training** prompt — call them instead of guessing. Backend strings are "cpu" | "gpu" | "gpu_graphs" (no colon); on CPU-only hosts pass backend=cpu.
 
 ASYNC: every action returns a job_id immediately. Poll jobs(action=status, job_id=...) every 10–15s (do NOT poll faster). Map training typical times: 10×10 ~30s | 20×20 ~3–5 min | 40×40 ~15–30 min — do not assume failure before 3 minutes on CPU. When status is completed, call results(action=get, job_id=...).
 
-action=impute: use when many cells are missing across training columns; use inference(impute_column) instead when you already have a complete-case map and need to fill one held-out column. Defaults model=auto, cv_folds=5 → quality.csv with two labeled numbers per column (kfold_holdout_pool = honest held-out; resubstitution_optimistic = in-sample, optimistic). Artifacts: imputed.csv, imputation_mask.csv, optional imputation_uncertainty.csv. A completed impute job_id is a valid parent for inference(impute_column).
+action=impute: use when many cells are missing across training columns; use inference(impute_column) instead when you already have a complete-case map and need to fill one held-out column. Supports transforms (log/sqrt/etc., not rank), temporal_features, cyclic_features, and normalization_methods (zscore/mad/sigmoidal/none — not sepd). Does not support auto_log_transforms, rank transforms, or normalize=sepd. Defaults model=auto (SIOM missSOM on som_siom+ plans, else plain missSOM), cv_folds=5 → quality.csv with two labeled numbers per column (kfold_holdout_pool = honest held-out; resubstitution_optimistic = in-sample, optimistic). Optional SIOM coverage knobs: gamma, gamma_f, siom_decay, siom_penalty, penalty_alpha, reset_per_epoch (same names as siom_map). Artifacts: imputed.csv, imputation_mask.csv, optional imputation_uncertainty.csv, summary.siom when SIOM path runs. A completed impute job_id is a valid parent for inference(impute_column).
 action=floop_siom: max_nodes is a TOTAL node budget (not a grid width/area); if coverage collapses, reduce max_nodes before increasing effort.
-NOT FOR: lifecycle (status/list/compare/cancel/delete) → use jobs; scoring data → use inference.`;
+NOT FOR: lifecycle (status/list/compare/cancel/delete) → use jobs; scoring data → use inference.
+
+Rank transform (map/siom/floop only): training-only — batch predict, project_columns, and impute_column cannot score rank-transformed columns.`;
+function assertImputeTrainingArgs(args) {
+    const normalize = args.normalize;
+    if (normalize === "sepd") {
+        throw new Error("train(impute) does not support normalize=sepd");
+    }
+    const nm = args.normalization_methods;
+    if (nm) {
+        for (const [col, method] of Object.entries(nm)) {
+            if (method === "sepd") {
+                throw new Error(`train(impute) does not support normalization_methods.${col}=sepd`);
+            }
+        }
+    }
+    const transforms = args.transforms;
+    if (transforms) {
+        for (const [col, t] of Object.entries(transforms)) {
+            if (t === "rank") {
+                throw new Error("train(impute) rank transform is not supported");
+            }
+        }
+    }
+    if (args.auto_log_transforms === true) {
+        throw new Error("train(impute) does not support auto_log_transforms yet.");
+    }
+}
 /**
  * Core training dispatcher. Shared by the `train` tool and the deprecated
  * `jobs(train_*)` wrappers. Every branch posts to the same /v1/jobs route.
  */
 export async function runTrain(action, args) {
     const dataset_id = args.dataset_id;
-    if (action === "baseline_study") {
-        if (!dataset_id)
-            throw new Error("train(baseline_study) requires dataset_id");
-        const dsInfo = (await apiCall("GET", `/v1/datasets/${dataset_id}/preview?n_rows=1`));
-        const nSamples = dsInfo.total_rows || 1000;
-        const totalNodes = Math.round(5 * Math.sqrt(nSamples));
-        const side = Math.max(5, Math.round(Math.sqrt(totalNodes)));
-        const params = {
-            model: "SOM",
-            periodic: false,
-            normalize: "mad",
-            grid: [side, side],
-            epochs: [10, 20],
-            batch_size: 32,
-            quality_metrics: "standard",
-            output_format: "png",
-            output_dpi: 2,
-            colormap: "coolwarm",
-        };
-        const data = (await apiCall("POST", "/v1/jobs", { dataset_id, params }));
-        const jid = String(data.id ?? "");
-        return { content: [{ type: "text", text: `Baseline study submitted. Job ID: ${jid}\nGrid size: ${side}x${side}\nNormalization: MAD\n\nPoll with jobs(action=status, job_id="${jid}") until complete, then retrieve with results(action=get, job_id="${jid}"). Optional: training_monitor(job_id="${jid}") for a visual panel—not required.` }] };
-    }
     if (action === "map" || action === "siom_map" || action === "impute") {
         const { preset, grid_x, grid_y, epochs, model, periodic, columns, cyclic_features, temporal_features, feature_weights, transforms, auto_log_transforms, time_delay_embeddings, categorical_features, normalize, sigma_f, learning_rate, batch_size, quality_metrics, backend, output_format, output_dpi, colormap, row_range, gamma, gamma_f, siom_decay, siom_penalty, penalty_alpha, reset_per_epoch, siom_feature_geometry, siom_qe_backend, siom_qe_batch_size, label, cv_folds, viz_mode, viz_top_components, emit_cell_uncertainty, } = args;
         let PRESETS = {};
@@ -64,6 +69,9 @@ export async function runTrain(action, args) {
         }
         if (!dataset_id)
             throw new Error(`train(${action}) requires dataset_id`);
+        if (action === "impute") {
+            assertImputeTrainingArgs(args);
+        }
         const { params, paramSummary, effectiveGrid } = buildTrainMapParams({
             preset, grid_x, grid_y, epochs, model, periodic, columns, cyclic_features,
             temporal_features, feature_weights, transforms, auto_log_transforms,
@@ -81,34 +89,44 @@ export async function runTrain(action, args) {
                 params.viz_top_components = viz_top_components;
             if (emit_cell_uncertainty !== undefined)
                 params.emit_cell_uncertainty = emit_cell_uncertainty;
-            // Plain numeric path only — strip unsupported keys if caller passed them
-            delete params.cyclic_features;
-            delete params.temporal_features;
+            // Strip keys still rejected on impute (transforms/temporal/cyclic/normalization_methods are kept)
             delete params.categorical_features;
-            delete params.transforms;
             delete params.auto_log_transforms;
             delete params.time_delay_embeddings;
         }
         if (action === "siom_map") {
             params._job_type = "train_siom";
-            if (gamma !== undefined)
-                params.gamma = gamma;
-            if (gamma_f !== undefined)
-                params.gamma_f = gamma_f;
-            if (siom_decay !== undefined)
-                params.siom_decay = siom_decay;
-            if (siom_penalty !== undefined)
-                params.siom_penalty = siom_penalty;
-            if (penalty_alpha !== undefined)
-                params.penalty_alpha = penalty_alpha;
-            if (reset_per_epoch !== undefined)
-                params.reset_per_epoch = reset_per_epoch;
-            if (siom_feature_geometry !== undefined)
-                params.siom_feature_geometry = siom_feature_geometry;
-            if (siom_qe_backend !== undefined)
-                params.siom_qe_backend = siom_qe_backend;
-            if (siom_qe_batch_size !== undefined)
-                params.siom_qe_batch_size = siom_qe_batch_size;
+        }
+        if (action === "siom_map" || action === "impute") {
+            const imputeModel = action === "impute" ? String(params.model ?? "auto") : "";
+            const siomKnobSet = gamma !== undefined ||
+                gamma_f !== undefined ||
+                siom_decay !== undefined ||
+                siom_penalty !== undefined ||
+                penalty_alpha !== undefined ||
+                reset_per_epoch !== undefined;
+            if (action === "siom_map" || imputeModel.toUpperCase() === "SIOM" || siomKnobSet) {
+                if (gamma !== undefined)
+                    params.gamma = gamma;
+                if (gamma_f !== undefined)
+                    params.gamma_f = gamma_f;
+                if (siom_decay !== undefined)
+                    params.siom_decay = siom_decay;
+                if (siom_penalty !== undefined)
+                    params.siom_penalty = siom_penalty;
+                if (penalty_alpha !== undefined)
+                    params.penalty_alpha = penalty_alpha;
+                if (reset_per_epoch !== undefined)
+                    params.reset_per_epoch = reset_per_epoch;
+            }
+            if (action === "siom_map") {
+                if (siom_feature_geometry !== undefined)
+                    params.siom_feature_geometry = siom_feature_geometry;
+                if (siom_qe_backend !== undefined)
+                    params.siom_qe_backend = siom_qe_backend;
+                if (siom_qe_batch_size !== undefined)
+                    params.siom_qe_batch_size = siom_qe_batch_size;
+            }
         }
         let totalRows = 0;
         try {
@@ -125,8 +143,7 @@ export async function runTrain(action, args) {
             : action === "impute" ? "variant=train_impute"
                 : "variant=som";
         data.effective_params = `${variantPrefix}, ${paramSummary}`;
-        const hasTransforms = action !== "impute" &&
-            transforms != null &&
+        const hasTransforms = transforms != null &&
             typeof transforms === "object" &&
             Object.keys(transforms).length > 0;
         const extras = buildTrainSubmitExtras({
@@ -164,79 +181,10 @@ export async function runTrain(action, args) {
         return textResult(data);
     }
     if (action === "floop_siom") {
-        const { columns, cyclic_features, temporal_features, feature_weights, transforms, auto_log_transforms, normalize, row_range, output_format, output_dpi, colormap, topology, max_nodes, effort, n_initial, n_passes, growth_interval, neighborhood_size, edge_max_age, max_degree, gamma, siom_decay, siom_penalty, penalty_alpha, error_threshold, error_decay, ring_close_threshold, sigma_0, sigma_f, eta_0, eta_f, elastic_lambda, elastic_mu, elastic_anchor, anchor_percentile, label, } = args;
         if (!dataset_id)
             throw new Error("train(floop_siom) requires dataset_id");
-        const params = {
-            _job_type: "train_floop_siom",
-            topology,
-            effort,
-            normalize,
-            output_format: output_format ?? "png",
-        };
-        if (max_nodes !== undefined)
-            params.max_nodes = max_nodes;
-        if (columns?.length)
-            params.columns = columns;
-        if (cyclic_features?.length)
-            params.cyclic_features = cyclic_features;
-        if (temporal_features?.length)
-            params.temporal_features = temporal_features;
-        if (feature_weights && Object.keys(feature_weights).length > 0)
-            params.feature_weights = feature_weights;
-        if (transforms && Object.keys(transforms).length > 0)
-            params.transforms = transforms;
-        if (auto_log_transforms)
-            params.auto_log_transforms = auto_log_transforms;
-        if (row_range && row_range.length >= 2 && row_range[0] <= row_range[1])
-            params.row_range = row_range;
-        const dpiMap = { standard: 1, retina: 2, print: 4 };
-        if (output_dpi && output_dpi !== "retina")
-            params.output_dpi = dpiMap[output_dpi] ?? 2;
-        if (colormap)
-            params.colormap = colormap;
-        if (n_initial !== undefined)
-            params.n_initial = n_initial;
-        if (n_passes !== undefined)
-            params.n_passes = n_passes;
-        if (growth_interval !== undefined)
-            params.growth_interval = growth_interval;
-        if (neighborhood_size !== undefined)
-            params.neighborhood_size = neighborhood_size;
-        if (edge_max_age !== undefined)
-            params.edge_max_age = edge_max_age;
-        if (max_degree !== undefined)
-            params.max_degree = max_degree;
-        if (gamma !== undefined)
-            params.gamma = gamma;
-        if (siom_decay !== undefined)
-            params.siom_decay = siom_decay;
-        if (siom_penalty !== undefined)
-            params.siom_penalty = siom_penalty;
-        if (penalty_alpha !== undefined)
-            params.penalty_alpha = penalty_alpha;
-        if (error_threshold !== undefined)
-            params.error_threshold = error_threshold;
-        if (error_decay !== undefined)
-            params.error_decay = error_decay;
-        if (ring_close_threshold !== undefined)
-            params.ring_close_threshold = ring_close_threshold;
-        if (sigma_0 !== undefined)
-            params.sigma_0 = sigma_0;
-        if (sigma_f !== undefined)
-            params.sigma_f = sigma_f;
-        if (eta_0 !== undefined)
-            params.eta_0 = eta_0;
-        if (eta_f !== undefined)
-            params.eta_f = eta_f;
-        if (elastic_lambda !== undefined)
-            params.elastic_lambda = elastic_lambda;
-        if (elastic_mu !== undefined)
-            params.elastic_mu = elastic_mu;
-        if (elastic_anchor !== undefined)
-            params.elastic_anchor = elastic_anchor;
-        if (anchor_percentile !== undefined)
-            params.anchor_percentile = anchor_percentile;
+        const { transforms, label } = args;
+        const { params, paramSummary } = buildFloopParams(args);
         const submitBody = { dataset_id, params };
         if (label && label.trim() !== "")
             submitBody.label = label;
@@ -249,14 +197,6 @@ export async function runTrain(action, args) {
         }
         catch { /* ignore */ }
         const hasTransforms = transforms != null && typeof transforms === "object" && Object.keys(transforms).length > 0;
-        const maxNodeSummary = max_nodes === undefined ? "max_nodes=auto(~2*sqrt(n_samples))" : `max_nodes=${max_nodes}`;
-        const paramSummary = [
-            "variant=floop",
-            `topology=${topology ?? "free"}`,
-            maxNodeSummary,
-            `effort=${effort ?? "standard"}`,
-            gamma !== undefined ? `gamma=${gamma}` : "",
-        ].filter(Boolean).join(", ");
         data.effective_params = paramSummary;
         const extras = buildTrainSubmitExtras({
             newJobId,
@@ -276,8 +216,8 @@ export async function runTrain(action, args) {
 export function registerTrainTool(server) {
     registerAuditedTool(server, "train", TRAIN_DESCRIPTION, {
         action: z
-            .enum(["map", "siom_map", "impute", "floop_siom", "baseline_study"])
-            .describe("map: standard SOM; siom_map: self-interacting map; impute: missing-tolerant map + imputed.csv; floop_siom: growing manifold (Premium/Enterprise); baseline_study: one-call heuristic baseline SOM"),
+            .enum(["map", "siom_map", "impute", "floop_siom"])
+            .describe("map: standard SOM; siom_map: self-interacting map; impute: missing-tolerant map + imputed.csv; floop_siom: growing manifold (Premium/Enterprise). For a quick first map use action=map with preset=quick."),
         dataset_id: z.string().optional().describe("Dataset ID to train on (required for all actions)."),
         label: z
             .string()