feat: predict_tja script

.gemini/checkpoint-final.md

@@ -1,21 +1,20 @@
-# fumen-analyze 최종 진행 리포트 (Final)
+# fumen-analyze 프로젝트 상태 보고 (Current)
 
-## 1. 모델 성능 요약
-- **최종 MAE**: **0.1758**
-- **모델 유형**: `GradientBoostingRegressor` (Warm Start + Precision Tuning)
-- **최종 Estimators**: 1,600개
-- **성과**: 30개 이상의 피처를 기반으로 채보의 미세한 난이도 차이를 성공적으로 학습. 이상치(에러 > 1.0)를 배제한 정밀 튜닝으로 안정성 확보.
+## 1. 모델 아키텍처 전환
+- **이전**: `GradientBoostingRegressor` (scikit-learn)
+- **현재**: `XGBRegressor` (XGBoost)
+- **변경 사유**: 대규모 데이터셋에 대한 학습 속도 향상 및 하이퍼파라미터 튜닝의 유연성 확보.
 
-## 2. 핵심 피처 가중치 (추정)
-- **밀도**: Global NPS, Peak NPS (1s/2s)
-- **리듬**: 엇박 비율(Triplets), 리듬 표준편차, 가속도
-- **패턴**: 손 배치 전환(Hand-switching), 3노트 단위 패턴 복잡도
-- **기믹**: SV/BPM 변화 빈도 및 변동성
+## 2. 성능 지표 (목표치)
+- **목표 MAE**: **0.15 이하** (현재 약 0.17~0.18 추정)
+- **허용 오차 범위**: ±0.1 (상수 단위 기준)
+- **성과**: 리듬 복잡도 및 순간 밀도 피처를 도입하여 10성급 고난이도 채보 간의 미세한 서열(상수 11.0~12.0)을 유의미하게 구분 중.
 
-## 3. 사용 안내
-- **학습 업데이트**: `./run_pipeline.sh` (현재 정밀 튜닝 모드 설정됨)
-- **상수 예측**: `./run_predict.sh <TJA_Path> [diff]`
-- **문서 참조**: `GUIDE.md`
+## 3. 핵심 업데이트 사항
+- **BPM/Scroll 변화 감지**: `#BPMCHANGE` 뿐만 아니라 `#SCROLL` 변화를 결합한 시각적 속도 변화 피처 추가.
+- **색상 복잡도 가중치**: 노트 간 간격이 좁을수록 더 높은 난이도 가중치를 부여하는 $1/\Delta t^2$ 로직 적용.
+- **지속 학습 지원**: `CONTINUE_TRAINING` 옵션을 통해 기존 모델에 추가 데이터를 점진적으로 학습 가능.
 
-## 4. 최종 결론
-현재 모델은 10성급 고난이도 곡들 사이의 미세한 '상수' 서열을 0.1~0.2 오차 범위 내에서 예측할 수 있는 수준에 도달했습니다. 추가 데이터 확보 시 0.1 미만으로의 진입이 충분히 가능합니다.
+## 4. 향후 과제
+- `rhythm_complexity` 로직의 정교화 (비정형 박자 감지 강화).
+- `measure.csv` 데이터셋 확충을 통한 과적합 방지 및 일반화 성능 향상.

.gemini/difficulty-factors.md

@@ -1,36 +1,23 @@
-# TJA 난이도 산정 핵심 요소 (Technical Factors)
+# TJA 난이도 산정 피처 (Features)
 
-'상수' 산출을 위해 채보에서 추출해야 하는 5개 핵심 차원과 세부 지표입니다.
+`preprocess/featurize.ts`에서 추출되는 8가지 핵심 피처입니다.
 
-## 1. 물리적 밀도 (Physical Density)
-물리적인 타격 속도와 관련된 지표입니다.
-- **Global NPS**: `총 노트 수 / 총 연주 시간`. 곡의 전반적인 속도 체급.
-- **Peak NPS (Sliding Window)**: 1초~2초 단위 윈도우에서 추출한 최대 NPS. 순간 폭타의 한계치.
-- **Density Variance**: NPS의 표준 편차. 곡이 얼마나 균일한지 또는 급격한지 측정.
+## 1. 양적 지표
+- **note_count**: 총 노트 수. 곡의 길이에 따른 절대적인 노트 양.
 
-## 2. 지구력 요구량 (Stamina Requirement)
-지치지 않고 집중력을 유지해야 하는 정도입니다.
-- **Longest Stream Count**: 쉼표(예: 8분음표 이상의 간격) 없이 이어지는 최대 노트 개수.
-- **Stream Density Ratio**: 전체 곡 시간 대비 스트림(연타) 구간이 차지하는 비중.
-- **Rest Interval Analysis**: 회복 가능한 구간의 배치와 빈도.
+## 2. 밀도 지표
+- **density_avg**: 평균 NPS. `(총 노트 수 / 곡의 총 소요 시간) * 1000`.
+- **density_peak**: 순간 최대 밀도. 임의의 노트를 기준으로 1초 이내에 포함된 최대 노트 수.
 
-## 3. 배치 복잡도 (Pattern Complexity)
-인지적 부하와 손 배치(Hand-switching)의 어려움입니다.
-- **Color Transition Ratio**: `색상 전환(d↔k) 횟수 / 총 노트 수`. 전환이 많을수록 인지 부하 증가.
-- **Hand-Switching Index**: 홀수(3, 5, 7) 및 짝수(2, 4) 연타의 혼합도. 기준 손이 강제로 바뀌는 빈도.
-- **Complex Pattern Detection**: 비정형 패턴(예: ddkdk, kkkdk 등)의 출현 빈도.
+## 3. 속도 변화 지표
+- **bpm_avg**: 노트별 BPM의 단순 산술 평균.
+- **bpm_change**: BPM 변동 횟수. 인접한 두 노트 간 BPM 차이가 1.5 이상일 때 카운트.
+- **scroll_change**: 시각적 속도(`BPM * Scroll`) 변동 횟수. 차이가 1.5 이상일 때 카운트.
 
-## 4. 리듬 복잡도 (Rhythmic Complexity)
-정확도(98%) 달성을 방해하는 타이밍 요소입니다.
-- **Quantization Diversity**: 사용된 음표 단위(1/16, 1/12, 1/24, 1/48 등)의 종류와 혼합 빈도.
-- **Off-beat Ratio**: 정박(1/4, 1/8)을 벗어난 엇박 노지의 비율.
-- **Rhythmic Entropy**: 노트 간 시간 간격의 불규칙성 정도.
+## 4. 구조적 복잡도
+- **rhythm_complexity**: 리듬의 정규성. 인접한 노트 간 시간 간격의 비율이 2의 거듭제곱(1, 0.5, 2 등)에 가까운 횟수를 카운트.
+    - *참고: 현재 코드는 정규 리듬일 때 증가하며, 문서와 실제 로직의 방향성이 반대일 수 있음.*
+- **color_complexity**: 색상 배치 복잡도. 인접 노트의 색상(Don/Ka)이 바뀔 때, 그 간격의 제곱의 역수($1/\Delta t^2$)를 누적. 간격이 좁을수록 수치가 급격히 상승.
 
-## 5. 가독성 및 기믹 (Reading & Gimmicks)
-시각적인 반응 속도와 암기 요소를 측정합니다.
-- **Scroll Velocity (SV) Variance**: `#SCROLL` 변화의 진폭과 빈도.
-- **BPM Fluctuation**: `#BPMCHANGE`를 통한 급격한 속도 변화 및 정지(#DELAY).
-- **Visual Overlap**: 저속 구간에서의 노트 겹침이나 고속 구간의 반응 한계.
-
-## 상수 추정 가중치 (Proposed)
-`Constant ∝ (Peak_NPS * 0.4) + (Complexity * 0.3) + (Stamina * 0.2) + (Reading * 0.1)`
+## 주의사항
+- `color_complexity`는 수치의 범위가 매우 클 수 있으므로 학습 시 `StandardScaler`를 통한 정규화가 필수적입니다.

.gemini/notes.md

@@ -2,16 +2,20 @@
 
 ## Technical Stack
 - **Runtime**: [Bun](https://bun.sh/)
-- **Language**: TypeScript
-- **Library**: [tja](https://www.npmjs.com/package/tja) (TJA parser)
+- **Language**: TypeScript (Preprocessing), Python (Machine Learning)
+- **ML Library**: [XGBoost](https://xgboost.readthedocs.io/), [scikit-learn](https://scikit-learn.org/)
+- **TJA Parser**: [tja-parser](https://www.npmjs.com/package/tja-parser)
 
-## Key Documents
-- `.gemini/tja-spec.md`: Rigorous TJA format specification.
-- `tja-format.mediawiki`: Original source document.
-- `measure.csv`: Dataset with columns `상수`, `songno`, `diff`.
+## Key Directories
+- `preprocess/`: TJA 파싱 및 피처 추출 로직 (TypeScript)
+- `script/`: 전처리, 학습 제어 스크립트
+- `train/`: XGBoost 학습 엔진 (Python)
+- `predict/`: 추론 엔진 (Python)
+- `datas/tja/`: 원본 TJA 데이터셋
+- `datas/measure.csv`: 정답지 (상수 데이터)
+- `test/`: 학습 결과물 (model.pkl, scaler.pkl, features.json)
 
-## Models (model/)
-- `constant_predictor.py`: DNN 기반 상수 예측 모델.
-
-## Library Usage (tja)
-...
+## Data Flow
+1. `datas/tja/*.tja` → `script/preprocess.ts` → `test/features.json`
+2. `test/features.json` + `datas/measure.csv` → `train/train_xgboost.py` → `test/model.pkl`
+3. `test/model.pkl` + `test/features.json` → `predict/predict_xgboost.py` → Result

.gemini/pipeline-design.md

@@ -1,13 +1,35 @@
-# 파이프라인 실행 규칙
+# 파이프라인 실행 가이드
 
-모든 경로는 실행 시점에 인자로 지정하여 관리합니다.
+모든 실행은 프로젝트 루트 디렉토리에서 수행합니다.
 
-## 1. 학습 파이프라인 (`run_train.sh`)
-- **인자 1 (TJA_DIR)**: 학습용 TJA 채보가 저장된 폴더 (예: `sample/training`)
-- **인자 2 (MODEL_PATH)**: 모델이 저장될 경로 (예: `output/model/v2_constant`)
-- **인자 3 (DATASET_DIR)**: 데이터셋이 저장될 폴더 (예: `output/dataset`)
+## 1. 전처리 (Feature Extraction)
+TJA 파일들로부터 피처를 추출하여 `features.json`을 생성합니다.
+```bash
+bun run script/preprocess.ts --workingDir ./test --dataDir ./datas
+```
+- **--workingDir**: 결과물(`features.json`)이 저장될 폴더.
+- **--dataDir**: 원본 데이터(`tja/` 폴더)가 위치한 폴더.
 
-## 2. 예측 파이프라인 (`run_predict.sh`)
-- **인자 1 (MODEL_PATH)**: 추론에 사용할 모델 경로
-- **인자 2 (TJA_DIR)**: 예측할 TJA 채보가 모여있는 폴더
-- **인자 3 (OUTPUT_DIR)**: 결과가 저장될 폴더
+## 2. 모델 학습 (Training)
+추출된 피처와 상수 데이터를 사용하여 XGBoost 모델을 학습시킵니다.
+```bash
+python3 train/train_xgboost.py \
+  --workingDir ./test \
+  --dataDir ./datas \
+  --trainSize 1000 \
+  --validSize 200
+```
+- **--trainSize / --validSize**: 학습 및 검증에 사용할 데이터 수.
+
+## 3. 난이도 예측 (Inference)
+특정 곡 번호(`songno`)에 대한 난이도 상수를 예측합니다.
+```bash
+python3 predict/predict_xgboost.py --workingDir ./test --songno 1000
+```
+- **--songno**: 예측할 곡의 번호 (파일명 기준).
+
+## 주요 산출물 (in `workingDir`)
+- `features.json`: 추출된 피처 데이터셋.
+- `model.pkl`: 학습된 XGBoost 모델 파일.
+- `scaler.pkl`: 피처 정규화를 위한 Scaler 객체.
+- `features.txt`: 학습에 사용된 피처 이름 목록.

.old/compare.ts

@@ -0,0 +1,37 @@
+import path from 'node:path';
+import measure from './datas/measure.json';
+import predict from './result.json';
+import { writeFile, writeFileSync } from 'node:fs';
+
+const ob: any[] = [];
+
+let max = 0;
+let min = Infinity;
+predict.forEach((p) => {
+    const songno = path.basename(p.filename, '.tja')
+    const m = measure.find((m) => m.songno.toString() === songno && (
+        p.difficulty_type === "oni" ? m.diff === "oni" :
+        m.diff === "ura"
+    ));
+
+    if(m){
+        if(max < p.predicted_measure){
+            max = p.predicted_measure
+        }
+        if(min > p.predicted_measure){
+            min = p.predicted_measure;
+        }
+        ob.push({
+            songno,
+            diff: m.diff,
+            predicted: p.predicted_measure,
+            measure: m.상수,
+            error: p.predicted_measure - m.상수,
+            errorAbs: Math.abs(p.predicted_measure - m.상수)
+        })
+    }
+});
+
+ob.forEach((v) => v.calibrated = ((v.predicted - min) * 11 / (max - min) + 1))
+
+writeFileSync('./compare.json', JSON.stringify(ob, null, 2))

.old/extract.ts

@@ -1,16 +1,23 @@
 import { parseTja } from './parse';
 import { factorize } from './factorize';
-import { readdirSync, readFileSync, writeFileSync, existsSync } from 'fs';
-import { join } from 'path';
+import { readdirSync, readFileSync, writeFileSync, existsSync, mkdirSync } from 'fs';
+import { join, dirname } from 'path';
 import iconv from 'iconv-lite';
 
 const [,, workingDir, dataDir] = process.argv;
 
-if (existsSync(join(workingDir, 'features.json'))) {
-    console.log('features.json already exists. Skipping extraction.');
+const featurePath = join(workingDir, 'factors.json');
+
+if (existsSync(featurePath)) {
+    console.log('factors.json already exists. Skipping extraction.');
     process.exit(0);
 }
 
+// 작업 디렉토리 존재 확인 및 생성
+if (!existsSync(workingDir)) {
+    mkdirSync(workingDir, { recursive: true });
+}
+
 const tjaDir = join(dataDir, 'tja');
 const results: any[] = [];
 
@@ -38,5 +45,5 @@ for (const file of readdirSync(tjaDir)) {
         results.push({ songno, diff: diff === 'oni' ? 'oni' : 'ura', ...factors });
     }
 }
-writeFileSync(join(workingDir, 'features.json'), JSON.stringify(results, null, 2));
-console.log(`Features extracted to ${workingDir}/features.json`);
+writeFileSync(featurePath, JSON.stringify(results, null, 2));
+console.log(`Features extracted to ${featurePath}`);

.old/predict-all.ts

@@ -0,0 +1,71 @@
+import { readdirSync, writeFileSync, existsSync } from 'fs';
+import { join, extname, dirname } from 'path';
+import { execSync } from 'child_process';
+
+async function predictAll() {
+    const args = process.argv.slice(2);
+    if (args.length < 3) {
+        console.log("Usage: bun script/predict-all.ts <workingDir> <tjaDir> <outputPath>");
+        process.exit(1);
+    }
+
+    const workingDir = args[0];
+    const tjaDir = args[1];
+    const outputPath = args[2];
+    const targetDifficulties = ['oni', 'edit'];
+
+    if (!existsSync(tjaDir)) {
+        console.error(`❌ Directory not found: ${tjaDir}`);
+        process.exit(1);
+    }
+
+    const files = readdirSync(tjaDir).filter(file => extname(file).toLowerCase() === '.tja');
+    console.log(`📂 Found ${files.length} TJA files. Processing Oni & Edit...`);
+
+    const finalResults: any[] = [];
+
+    let i = 0;
+    for (const file of files) {
+        const fullPath = join(tjaDir, file);
+
+        for (const diff of targetDifficulties) {
+            try {
+                // 개별 예측 실행
+                const cmd = `bun script/predict.ts "${workingDir}" "${fullPath}" "${diff}"`;
+                const output = execSync(cmd, { stdio: ['pipe', 'pipe', 'ignore'] }).toString();
+
+                const match = output.match(/(?:Prediction|Predicted Difficulty):\s*(\d+\.\d+)/);
+
+                if (match) {
+                    const value = parseFloat(match[1]);
+                    console.log(`✅ [${file}] ${diff.toUpperCase()}: ${value}`);
+                    finalResults.push({
+                        filename: file,
+                        difficulty_type: diff,
+                        predicted_measure: value
+                    });
+                }
+            } catch (e) {
+                // 해당 난이도가 없는 경우 조용히 넘어감
+            }
+        }
+        i++;
+        console.log(`${i}/${files.length}`)
+    }
+
+    // 결과 저장 (확장자에 따라 분기)
+    if (outputPath.endsWith('.json')) {
+        writeFileSync(outputPath, JSON.stringify(finalResults, null, 2));
+    } else {
+        // CSV 저장 (기본값)
+        const csvContent = [
+            "filename,difficulty,predicted_measure",
+            ...finalResults.map(r => `"${r.filename}",${r.difficulty_type},${r.predicted_measure}`)
+        ].join('\n');
+        writeFileSync(outputPath, csvContent);
+    }
+
+    console.log(`\n✨ Prediction complete! Results saved to: ${outputPath}`);
+}
+
+predictAll();

.old/predict.sh

@@ -0,0 +1,15 @@
+#!/bin/bash
+
+# Arguments: working_dir tja_path difficulty
+WORKING_DIR=${1:-"."}
+TJA_PATH=${2}
+DIFFICULTY=${3:-"oni"}
+
+if [ -z "$TJA_PATH" ]; then
+    echo "Usage: ./predict.sh <working_dir> <tja_path> [difficulty: oni|edit]"
+    exit 1
+fi
+
+echo "Running prediction for $TJA_PATH..."
+# script_dir 인자가 사라졌으므로, 예측 스크립트 실행 시 인자 순서에 맞춰 실행
+bun script/predict.ts "$WORKING_DIR" "$TJA_PATH" "$DIFFICULTY"

.old/predict.ts

@@ -0,0 +1,97 @@
+import { parseTja } from './parse';
+import { factorize } from './factorize';
+import { join } from 'path';
+import { readFileSync, writeFileSync, unlinkSync, existsSync } from 'fs';
+import { execSync } from 'child_process';
+
+/**
+ * 예측 엔진: 모델 실행을 위해 Python 환경을 호출
+ */
+function runInference(workingDir: string, factors: Record<string, number>): number {
+    const tempFactorPath = join(workingDir, `temp_feat_${Date.now()}.json`);
+    writeFileSync(tempFactorPath, JSON.stringify(factors));
+
+    const pythonScript = `
+import json, torch, sys
+import numpy as np
+import torch.nn as nn
+
+# 모델 정의 (학습된 구조와 동일해야 함)
+class DifficultyNet(nn.Module):
+    def __init__(self, input_dim):
+        super().__init__()
+        self.net = nn.Sequential(
+            nn.Linear(input_dim, 128),
+            nn.BatchNorm1d(128),
+            nn.ReLU(),
+            nn.Dropout(0.4),
+            nn.Linear(128, 64),
+            nn.ReLU(),
+            nn.Dropout(0.3),
+            nn.Linear(64, 32),
+            nn.ReLU(),
+            nn.Linear(32, 1),
+            nn.Sigmoid()
+        )
+    def forward(self, x): return self.net(x)
+
+def predict():
+    try:
+        with open('${tempFactorPath}', 'r') as f: factors = json.load(f)
+        with open('${join(workingDir, 'scaler.json')}', 'r') as f: s = json.load(f)
+
+        X = np.array([[factors[col] for col in s['cols']]], dtype=np.float32)
+        
+        # 정규화 (Robust/Standard 지원)
+        if 'center' in s: X_scaled = (X - np.array(s['center'])) / np.array(s['scale'])
+        else: X_scaled = (X - np.array(s.get('mean', 0))) / np.array(s.get('std', s.get('scale', 1)))
+
+        model = DifficultyNet(len(s['cols']))
+        model.load_state_dict(torch.load('${join(workingDir, 'model.pth')}', map_location='cpu'))
+        model.eval()
+
+        with torch.no_grad():
+            input_tensor = torch.from_numpy(X_scaled).float()
+            pred = model(input_tensor) * 12 + 0.5
+        print(float(pred.item()))
+    except Exception as e:
+        print(f"ERROR: {e}", file=sys.stderr)
+        sys.exit(1)
+
+if __name__ == "__main__": predict()
+`;
+
+    const tempPyPath = join(workingDir, `inference_${Date.now()}.py`);
+    writeFileSync(tempPyPath, pythonScript);
+
+    try {
+        const result = execSync(`python3 ${tempPyPath}`).toString().trim();
+        return parseFloat(result);
+    } finally {
+        if (existsSync(tempFactorPath)) unlinkSync(tempFactorPath);
+        if (existsSync(tempPyPath)) unlinkSync(tempPyPath);
+    }
+}
+
+function main() {
+    const [workingDir, tjaPath, difficulty = 'oni'] = process.argv.slice(2);
+    
+    if (!workingDir || !tjaPath) {
+        console.log("Usage: bun run script/predict.ts <workingDir> <tjaPath> [difficulty]");
+        process.exit(1);
+    }
+
+    const tjaContent = readFileSync(tjaPath, 'utf-8');
+    const parsed = parseTja(tjaContent);
+    const chart = parsed[difficulty.toLowerCase() as 'oni' | 'edit'];
+
+    if (!chart) {
+        console.error(`Error: Difficulty '${difficulty}' not found.`);
+        process.exit(1);
+    }
+    
+    const score = runInference(workingDir, factorize(chart));
+    console.log(`\n🎯 Predicted Difficulty: ${score.toFixed(2)}`);
+}
+
+main();

.old/spec.md

@@ -0,0 +1,70 @@
+# 학습 
+## CLI
+- cli로 학습할 수 있는 쉘 스크립트를 작성한다.
+- argument
+    - script_dir
+    - working_dir
+    - data_dir
+    - train_data_count = 100
+    - validate_data_count = 20
+    - margin = 0.5
+    - validate_iterations = 5
+
+## 학습 모델
+- **추천:** PyTorch 기반의 **Deep MLP** (Dropout 및 Weight Decay 필수 적용)
+- **과적합 방지 전략:**
+    - **Dropout (0.3 ~ 0.5):** 학습 시 뉴런을 의도적으로 꺼서 특정 데이터 암기를 방지한다.
+    - **L2 Regularization (Weight Decay):** 가중치가 비정상적으로 커지는 것을 막는다.
+    - **Output Mapping:** 최종 출력에 `Sigmoid`를 걸고 (y * 11) + 1을 적용해 [1, 12] 범위를 강제한다.
+
+## 학습 과정
+1. `{data_dir}/tja`의 tja를 파싱한다. (`{script_dir}/parse.ts`)
+2. `oni`와 `edit` course를 factorize하여 `{working_dir}/factors.json`에 저장한다.
+    - 이미 존재하면 생략하되, Factor 로직 수정 시 삭제 후 재실행한다.
+3. 데이터 표준화(Standardization)를 위한 스케일링 파라미터를 계산하고 `{working_dir}/scaler.json`에 저장한다.
+4. **학습 단계:** 전체 데이터 중 `train_data_count`만큼 데이터를 랜덤하게 추출하여 학습을 진행한다.
+    - `edit` 코스는 `ura`와 매칭하여 오차를 계산한다.
+    - 모델이 충분히 수렴할 때까지 학습을 수행하며, 기존 모델 존재 시 가중치를 로드하여 업데이트한다.
+5. **검증 단계:** 학습이 완료된 모델을 대상으로 아래 과정을 수행한다.
+    - `validate_data_count`만큼의 데이터를 랜덤하게 추출하여 오차를 계산하고 `validate_result_{n}.csv`에 저장한다.
+    - 이 과정을 `validate_iterations` 횟수만큼 반복한다.
+    - **재시도 조건:** 검증 루프 도중 단 한 번이라도 Mean Absolute Error(MAE)가 `margin`보다 크다면, 즉시 검증을 중단하고 4번(학습 단계)으로 돌아가 데이터를 새로 샘플링하여 재학습한다.
+6. **성공 및 종료:** 모든 `validate_iterations` 회차에서 MAE가 `margin` 이하일 경우에만 학습 성공으로 간주하고 최종 모델 파일을 저장한다.
+
+### 유의 사항
+- 모든 데이터는 `working_dir`에 저장한다.
+- Apple Silicon 환경에서 `mps` 디바이스를 할당하여 GPU 가속을 사용한다.
+
+# 예측
+## CLI
+- argument: `working_dir`, `tja`, `difficulty` (oni | edit)
+
+## 예측 과정
+- 로드된 모델과 학습 시 사용했던 스케일링 파라미터(Mean, Std)를 동일하게 적용하여 예측을 진행한다.
+
+# Factor
+## 추천 Factor 세트 (난이도 결정 핵심 지표)
+1.  **density_avg**: 평균 노트 밀도 (총 노트 수 / 곡 길이)
+2.  **density_peak**: 1초당 최대 밀도 (가장 촘촘한 구간의 노트 수)
+3.  **bpm_avg**: 곡의 평균 BPM
+4.  **bpm_change**: BPM 변화율 (BPM 변속의 다양성)
+5.  **note_count**: 총 노트 수
+6.  **complexity**: 동캇 꼬임
+
+## Factor 계산 로직
+- **밀도 계열**: 
+    - 노트 간 시간 간격을 `delta`라고 할 때, `1 / delta`로 순시 밀도를 구한 뒤 최대값(`peak`)과 평균값(`avg`)을 도출합니다.
+- **패턴 계열**:
+    - 노트 타입을 배열로 매핑(예: 0=none, 1=don, 2=ka, 3=big_don, 4=big_ka)합니다.
+    - 슬라이딩 윈도우(예: 4~8개 노트)를 사용하여 인접한 노트의 타입 변화가 잦은 구간을 복합 패턴으로 판정(`complex_ratio`)합니다.
+- **연타 계열**:
+    - `#ROLL` 명령 혹은 특정 시간 내에 입력이 몰려있는 노트를 연타로 간주하여 카운트합니다.
+
+## 구현 가이드
+- `factorize.ts`에서 각 곡의 `Course` 데이터를 순회하며 위 지표들을 계산합니다.
+- 계산된 값은 `Record<string, number>` 형태로 반환하여 학습기에 입력합니다.
+- 모든 Factor는 이상치(Outlier) 처리를 위해 학습 단계에서 `StandardScaler` 또는 `RobustScaler`를 반드시 거쳐야 합니다.
+
+# Scripts
+- `parse.ts`: TJA 문자열을 `Course` 객체로 변환.
+- `factorize.ts`: `Course`를 입력받아 확장된 `Factor` 세트를 반환.

.old/t.sh

@@ -0,0 +1,6 @@
+for val in 0.5 0.45 0.4 0.35 0.3 0.25 0.2; do
+  echo "🚀 현재 학습 파라미터: $val (50회 반복 중)"
+  for i in {1..50}; do
+    ./train.sh script test datas 1000 200 $val 10
+  done
+done

.old/test/scaler.json

@@ -0,0 +1 @@
+{"mean": [630.3327571305099, 5.390975807838123, 11.292999135695766, 173.114216056683, 2.808124459809853, 0.011057130071469854, 4.286084701815039], "std": [211.64659003608378, 1.4301422132137462, 3.1270522447781857, 42.402697977620676, 7.900275348859312, 0.002796230062876389, 15.234164907711529], "cols": ["note_count", "density_avg", "density_peak", "bpm_avg", "bpm_change", "complexity", "scroll_change"]}

.old/test/validate_result_1.csv

@@ -0,0 +1,201 @@
+songno,diff,상수,predicted_measure,error
+828,oni,4.1,4.206385,0.10638513565063512
+63,oni,6.1,4.832204,1.2677961349487301
+1216,oni,10.0,9.5192,0.48079967498779297
+243,oni,11.0,9.915412,1.0845880508422852
+999,oni,10.8,9.808392,0.9916084289550788
+32,oni,6.8,6.970789,0.17078895568847674
+1012,oni,9.7,9.215172,0.48482818603515554
+959,oni,7.0,7.0853252,0.08532524108886719
+224,oni,6.8,7.196848,0.39684791564941424
+349,oni,6.8,6.479022,0.3209779739379881
+397,oni,7.9,7.942276,0.042276000976562145
+539,oni,6.0,5.876666,0.12333393096923828
+1100,oni,3.9,3.809228,0.09077205657958975
+1269,oni,3.2,2.7388601,0.4611398696899416
+1214,oni,6.0,6.4056187,0.40561866760253906
+161,oni,3.0,2.962341,0.0376589298248291
+193,oni,4.4,4.2237315,0.1762684822082523
+1002,oni,4.9,4.9559927,0.05599269866943324
+117,oni,8.0,8.2179575,0.2179574966430664
+625,oni,5.5,5.444084,0.05591583251953125
+567,oni,7.1,7.0908155,0.009184455871581676
+122,oni,1.9,2.9562376,1.056237554550171
+519,oni,7.3,7.4535465,0.15354652404785174
+559,oni,5.4,5.4344945,0.03449449539184535
+417,oni,3.3,3.3136492,0.013649177551269709
+60,oni,7.9,7.9494324,0.049432373046874645
+629,oni,7.3,6.906698,0.3933017730712889
+990,oni,10.9,9.670632,1.2293676376342777
+91,oni,3.8,3.716854,0.08314590454101545
+604,oni,1.0,1.1668034,0.16680335998535156
+166,oni,6.0,5.829277,0.17072296142578125
+388,oni,9.0,9.137562,0.13756179809570312
+1073,oni,2.3,2.394328,0.09432811737060565
+1175,oni,4.5,4.54766,0.047659873962402344
+1208,oni,4.3,4.2226963,0.07730369567871076
+367,oni,5.8,6.021642,0.2216422080993654
+223,oni,8.1,7.3584685,0.7415314674377438
+308,oni,5.2,4.990783,0.20921678543090838
+613,oni,3.0,2.9797726,0.020227432250976562
+1402,oni,5.8,5.8168125,0.01681251525878924
+1335,oni,4.0,3.8683329,0.1316671371459961
+882,oni,1.9,1.8341815,0.06581845283508292
+933,oni,7.1,7.331907,0.23190679550170934
+1422,oni,5.6,5.7453094,0.14530935287475621
+430,oni,1.5,1.7298045,0.22980451583862305
+454,oni,7.7,7.780956,0.0809557914733885
+887,oni,7.2,7.051265,0.1487352371215822
+894,oni,1.7,1.6517099,0.04829008579254146
+719,oni,9.9,9.321081,0.5789188385009769
+660,oni,6.6,6.25142,0.34857997894287074
+560,oni,6.0,5.845188,0.15481185913085938
+1421,oni,8.1,8.032013,0.06798706054687464
+434,oni,4.1,4.1712093,0.07120933532714879
+1356,oni,9.4,9.225898,0.1741022109985355
+262,oni,6.8,6.8975925,0.09759254455566424
+918,oni,6.4,6.2478523,0.15214767456054723
+1240,oni,4.2,4.3363075,0.13630752563476545
+1164,oni,5.4,5.845188,0.44518814086914027
+962,oni,2.6,3.1786516,0.5786515712738036
+304,oni,5.6,5.4814534,0.11854658126831019
+1255,oni,10.4,9.691372,0.708628082275391
+126,oni,4.6,4.485203,0.11479721069335902
+842,oni,2.1,1.9751469,0.12485311031341562
+930,oni,7.1,7.913493,0.8134931564331058
+775,oni,10.6,9.802061,0.7979389190673825
+1359,oni,2.2,2.4671395,0.26713948249816877
+1219,oni,4.4,4.62963,0.229630088806152
+42,oni,1.1,2.9574435,1.8574434757232665
+993,oni,10.0,9.799214,0.20078563690185547
+1059,oni,5.6,5.5380435,0.06195650100707972
+504,oni,6.1,5.845188,0.254811859130859
+1282,oni,3.8,4.334233,0.534232807159424
+1285,oni,2.4,2.4647148,0.06471476554870614
+720,oni,9.6,9.387812,0.21218833923339808
+66,oni,5.9,5.845188,0.05481185913085973
+1377,oni,4.9,4.9466968,0.04669675827026332
+74,oni,4.5,4.536579,0.036579132080078125
+85,oni,7.4,7.427013,0.027012920379638317
+51,oni,3.2,2.998054,0.20194597244262713
+487,oni,3.5,3.6010914,0.10109138488769531
+137,oni,6.1,6.0426464,0.05735359191894496
+649,oni,7.6,7.212104,0.3878961563110348
+1295,oni,1.3,1.133282,0.16671805381774907
+257,oni,6.3,6.122114,0.17788581848144513
+1015,oni,8.7,8.804349,0.10434894561767649
+127,oni,5.9,5.9468956,0.04689559936523402
+927,oni,8.9,8.5859165,0.3140834808349613
+654,oni,4.5,4.328125,0.171875
+96,oni,8.8,8.476414,0.3235862731933601
+366,oni,1.5,2.3543668,0.8543667793273926
+98,oni,3.1,3.4718246,0.37182464599609366
+1319,oni,8.8,8.910327,0.11032695770263601
+1330,oni,1.6,2.3295865,0.7295865058898925
+1369,oni,11.8,11.984631,0.18463058471679616
+273,oni,3.3,3.53368,0.2336799621582033
+465,oni,6.9,6.837492,0.06250801086425817
+1107,oni,6.5,6.6370926,0.13709259033203125
+431,oni,7.6,7.6814866,0.08148660659790075
+568,oni,7.5,7.328082,0.17191791534423828
+738,oni,7.9,7.6613083,0.2386917114257816
+307,oni,4.1,3.9636607,0.1363392829895016
+1280,oni,3.9,4.2937455,0.393745517730713
+1457,oni,5.4,5.485291,0.08529100418090785
+1314,oni,4.4,4.248893,0.151107215881348
+1109,oni,6.8,6.6508837,0.1491163253784178
+1233,oni,4.1,3.903399,0.19660100936889613
+1234,oni,6.6,6.50439,0.09561023712158168
+1366,oni,4.7,4.8216677,0.1216676712036131
+115,oni,2.3,2.2282612,0.07173876762390119
+947,oni,5.2,5.7635384,0.563538360595703
+261,oni,7.4,7.6588984,0.2588983535766598
+458,oni,6.3,6.2306204,0.06937961578369123
+319,oni,4.4,4.567095,0.16709480285644496
+1349,oni,6.3,5.845188,0.4548118591308592
+730,oni,5.1,4.697263,0.40273723602294886
+226,oni,6.7,6.548712,0.15128822326660174
+998,oni,5.0,5.098602,0.0986018180847168
+816,oni,8.2,8.676427,0.47642688751220774
+1034,oni,5.9,5.8499846,0.050015354156494496
+1049,oni,9.5,9.412386,0.08761405944824219
+12,oni,7.5,7.4275417,0.07245826721191406
+1407,oni,2.8,2.7094536,0.09054641723632795
+432,oni,6.6,6.7569847,0.15698471069335973
+406,oni,7.7,7.693886,0.006114196777343928
+68,oni,9.2,8.833591,0.36640853881835866
+1000,oni,8.0,8.003995,0.003994941711425781
+757,oni,10.5,10.120777,0.3792228698730469
+274,oni,4.3,4.3155355,0.015535545349121271
+1077,oni,3.9,3.902987,0.0029870033264161044
+447,oni,4.0,4.0462713,0.046271324157714844
+577,oni,9.8,9.342824,0.4571760177612312
+518,oni,11.4,11.650173,0.250173187255859
+1147,oni,9.1,8.942031,0.15796909332275355
+614,oni,6.4,5.970553,0.42944707870483434
+148,oni,9.7,9.349513,0.3504869461059563
+739,oni,7.1,7.3886776,0.2886775970458988
+335,oni,9.4,9.030155,0.36984481811523473
+220,oni,9.3,9.154587,0.1454132080078132
+566,oni,7.6,7.663207,0.06320705413818395
+110,oni,10.9,10.075016,0.8249839782714847
+694,oni,3.5,4.001005,0.5010051727294922
+1351,oni,9.3,9.762559,0.4625589370727532
+256,oni,4.9,4.7010565,0.1989435195922855
+822,oni,7.2,7.1134377,0.08656234741210955
+1136,oni,2.7,3.2574682,0.5574682235717772
+898,oni,3.8,4.029142,0.22914190292358416
+167,oni,7.1,7.137214,0.0372141838073734
+352,oni,2.5,2.549781,0.049781084060668945
+675,oni,5.2,5.875123,0.6751230239868162
+483,oni,7.1,7.3588047,0.2588047027587894
+626,oni,6.3,6.4142413,0.11424131393432635
+1036,oni,6.5,6.2653823,0.2346177101135254
+672,oni,2.4,2.623148,0.22314796447753915
+1252,oni,4.9,5.0239687,0.12396869659423793
+639,oni,7.0,6.931817,0.06818294525146484
+610,oni,6.0,5.80618,0.19381999969482422
+350,oni,5.5,5.536312,0.036312103271484375
+906,oni,9.1,9.289001,0.18900146484375036
+1048,oni,9.4,9.316141,0.08385887145996129
+310,oni,6.2,6.251908,0.051907825469970525
+415,oni,1.8,2.84784,1.0478400707244873
+1354,oni,7.0,7.4809437,0.4809436798095703
+831,oni,1.2,1.2000558,5.5837631225630346e-05
+322,oni,7.0,7.011821,0.011820793151855469
+1196,oni,6.6,6.2691593,0.3308406829833981
+482,oni,1.8,1.6762199,0.12378005981445317
+687,oni,4.3,4.1856074,0.11439256668090803
+1080,oni,8.0,8.098514,0.09851360321044922
+391,oni,7.4,6.5624576,0.8375424385070804
+402,oni,6.5,6.452997,0.04700279235839844
+1218,oni,1.4,1.6566312,0.2566312313079835
+1110,oni,1.7,2.0302389,0.3302388668060303
+1304,oni,4.7,4.2533813,0.4466187477111818
+340,oni,5.2,5.115123,0.08487720489501971
+1124,oni,2.8,2.4958134,0.30418663024902326
+1451,oni,5.1,5.1919827,0.09198274612426793
+1322,oni,7.9,7.850716,0.0492838859558109
+1382,oni,3.2,3.5220692,0.32206921577453596
+819,oni,8.6,8.7340145,0.1340145111083988
+1174,oni,6.2,6.517852,0.3178518295288084
+1256,oni,6.2,6.084713,0.1152870178222658
+321,oni,6.3,6.2436576,0.05634241104125959
+1424,oni,6.4,6.2954216,0.10457839965820348
+26,oni,4.1,5.500279,1.4002789497375492
+43,oni,8.1,8.08124,0.018760299682616832
+480,oni,3.2,3.3859277,0.18592767715454084
+13,oni,7.2,7.3293734,0.1293733596801756
+968,oni,7.0,6.966305,0.033695220947265625
+558,oni,8.5,8.246681,0.25331878662109375
+795,oni,2.6,2.3113527,0.2886472702026368
+469,oni,4.5,4.6131096,0.11310958862304688
+593,oni,10.6,9.9727,0.627299880981445
+551,oni,4.5,4.616651,0.11665105819702148
+797,oni,8.8,7.9628973,0.8371026992797859
+1133,oni,5.7,5.578541,0.12145919799804705
+929,oni,8.3,8.203068,0.09693222045898509
+1232,oni,7.2,6.9321103,0.2678896903991701
+686,oni,8.7,8.237562,0.4624378204345696
+732,oni,2.4,2.3426375,0.0573624610900878
+651,oni,7.1,6.437669,0.6623311996459957

.old/test/validate_result_10.csv

@@ -0,0 +1,201 @@
+songno,diff,상수,predicted_measure,error
+897,oni,4.5,4.108138,0.3918619155883789
+1338,oni,10.0,9.66922,0.330780029296875
+116,oni,2.8,2.8755562,0.07555623054504412
+1357,oni,8.1,8.198347,0.09834709167480504
+927,oni,8.9,8.5859165,0.3140834808349613
+1073,oni,2.3,2.394328,0.09432811737060565
+1150,oni,4.1,3.2386503,0.8613496780395504
+1354,oni,7.0,7.4809437,0.4809436798095703
+477,oni,9.3,9.341164,0.04116363525390554
+1309,oni,6.5,6.507756,0.007756233215332031
+495,oni,10.1,9.74755,0.3524499893188473
+616,oni,5.8,5.77202,0.027980136871337713
+542,oni,4.2,4.0955725,0.10442752838134783
+360,oni,9.2,9.504893,0.3048933029174812
+1076,oni,2.1,2.0889874,0.011012649536132901
+922,oni,5.7,5.6340675,0.06593246459960955
+1331,oni,2.2,2.7129545,0.512954521179199
+89,oni,7.6,7.661222,0.06122198104858434
+1180,oni,6.5,6.422063,0.07793712615966797
+22,oni,5.5,5.295011,0.20498895645141602
+925,oni,5.2,4.9190726,0.28092737197875994
+1308,oni,8.4,8.65023,0.2502304077148434
+782,oni,2.7,2.7994647,0.099464702606201
+630,oni,9.3,8.637385,0.6626146316528327
+1205,oni,6.1,6.03403,0.06597003936767543
+797,oni,8.8,7.9628973,0.8371026992797859
+200,oni,2.3,2.223773,0.0762269973754881
+279,oni,4.7,4.948142,0.24814205169677717
+1125,oni,3.6,3.4895444,0.11045560836792001
+861,oni,4.2,4.4093747,0.2093747138977049
+469,oni,4.5,4.6131096,0.11310958862304688
+375,oni,8.7,8.550151,0.14984912872314382
+949,oni,5.1,4.900959,0.19904098510742152
+150,oni,6.3,6.2841473,0.015852737426757635
+971,oni,10.3,9.681194,0.6188056945800788
+938,oni,5.3,5.209937,0.09006290435790998
+140,oni,3.3,3.0745137,0.2254863262176512
+1097,oni,5.1,5.135783,0.035783195495605824
+483,oni,7.1,7.3588047,0.2588047027587894
+1426,oni,6.2,5.8172493,0.38275070190429705
+513,oni,8.4,8.350839,0.0491613388061527
+107,oni,4.6,4.2985578,0.3014422416687008
+1327,oni,3.7,4.1225753,0.42257528305053693
+1342,oni,2.2,2.297284,0.097283887863159
+1,oni,9.0,9.170969,0.17096900939941406
+1023,oni,3.4,3.542378,0.14237794876098642
+438,oni,7.7,7.6389027,0.061097335815429865
+343,oni,6.4,5.75634,0.6436599731445316
+1363,oni,4.4,4.349381,0.05061902999877965
+1425,oni,7.9,7.9018764,0.0018764495849605822
+1037,oni,9.2,8.873722,0.32627792358398366
+990,oni,10.9,9.670632,1.2293676376342777
+774,oni,9.0,9.228863,0.22886276245117188
+1279,oni,8.4,8.668176,0.2681756973266598
+1208,oni,4.3,4.2226963,0.07730369567871076
+924,oni,2.8,2.7967556,0.003244447708129705
+864,oni,8.1,8.247111,0.14711132049560582
+1412,oni,6.2,6.3555965,0.15559654235839826
+608,oni,3.5,3.7859712,0.28597116470336914
+587,oni,5.2,5.1716547,0.02834529876709002
+852,oni,6.6,6.5446224,0.05537757873535121
+1172,oni,11.9,11.048339,0.8516611099243168
+976,oni,8.7,8.713133,0.013132858276367898
+1165,oni,7.9,7.9121323,0.012132263183593395
+3,oni,3.6,2.7672272,0.8327728271484376
+1295,oni,1.3,1.133282,0.16671805381774907
+1422,oni,5.6,5.7453094,0.14530935287475621
+471,oni,7.2,7.574646,0.3746459960937498
+383,oni,6.1,6.0199423,0.08005771636962855
+48,oni,5.6,5.4978957,0.10210428237915004
+381,oni,2.3,2.3997626,0.09976263046264666
+707,oni,6.4,6.3381214,0.06187858581543004
+221,oni,6.7,6.725834,0.025833892822265447
+1181,oni,3.3,3.7384562,0.4384562492370607
+257,oni,6.3,6.122114,0.17788581848144513
+445,oni,7.1,7.3118544,0.21185436248779332
+432,oni,6.6,6.7569847,0.15698471069335973
+462,oni,7.3,6.9185348,0.3814652442932127
+595,oni,9.5,9.133109,0.36689090728759766
+524,oni,8.9,9.106796,0.20679626464843714
+1001,oni,3.6,3.725905,0.1259049415588378
+392,oni,9.0,8.880889,0.1191110610961914
+741,oni,3.5,4.1137867,0.6137866973876953
+1012,oni,9.7,9.215172,0.48482818603515554
+1435,oni,5.2,5.0266085,0.1733915328979494
+36,oni,1.6,1.8594168,0.25941684246063224
+1024,oni,3.3,3.3856637,0.08566374778747576
+152,oni,7.9,7.8858066,0.014193439483642933
+1085,oni,10.0,9.612744,0.3872556686401367
+397,oni,7.9,7.942276,0.042276000976562145
+204,oni,2.4,2.719502,0.3195019721984864
+736,oni,3.5,3.618028,0.11802792549133301
+644,oni,4.4,4.3509674,0.049032592773437855
+1349,oni,6.3,5.845188,0.4548118591308592
+1056,oni,4.7,4.598622,0.10137815475463885
+727,oni,6.7,6.5126877,0.18731231689453143
+738,oni,7.9,7.6613083,0.2386917114257816
+163,oni,5.8,5.5594893,0.24051074981689435
+11,oni,1.5,2.3513732,0.8513731956481934
+1374,oni,1.6,2.4462283,0.846228265762329
+1137,oni,7.1,7.197917,0.09791698455810582
+284,oni,3.8,3.552865,0.24713497161865217
+926,oni,3.7,3.3538485,0.3461515426635744
+931,oni,7.7,7.59026,0.10973997116088885
+851,oni,10.1,9.837338,0.26266155242919886
+1274,oni,4.8,4.9417405,0.14174051284790057
+157,oni,3.3,3.1216764,0.1783235549926756
+889,oni,4.2,4.020527,0.17947311401367205
+1029,oni,1.6,1.3594662,0.24053380489349374
+1061,oni,3.9,3.5892773,0.3107227325439452
+352,oni,2.5,2.549781,0.049781084060668945
+141,oni,4.6,4.4628506,0.1371494293212887
+601,oni,3.5,3.6112447,0.11124467849731445
+1433,oni,5.5,5.6608744,0.1608743667602539
+114,oni,7.0,6.7927685,0.2072315216064453
+198,oni,2.4,2.7392402,0.3392401695251466
+882,oni,1.9,1.8341815,0.06581845283508292
+1120,oni,8.2,8.383063,0.18306331634521555
+1090,oni,4.2,4.540201,0.3402011871337889
+1282,oni,3.8,4.334233,0.534232807159424
+1148,oni,3.4,3.4094017,0.009401655197143644
+1297,oni,9.4,9.196398,0.20360221862793004
+808,oni,6.2,6.37263,0.1726301193237303
+345,oni,4.5,3.4398248,1.0601751804351807
+184,oni,5.2,4.5262094,0.6737906455993654
+243,oni,11.0,9.915412,1.0845880508422852
+1020,oni,5.0,5.0843077,0.08430767059326172
+1188,oni,3.9,4.303166,0.4031659126281739
+333,oni,1.1,1.4720416,0.3720416069030761
+94,oni,9.9,9.473127,0.42687263488769567
+1280,oni,3.9,4.2937455,0.393745517730713
+1104,oni,9.7,9.212923,0.4870769500732415
+539,oni,6.0,5.876666,0.12333393096923828
+928,oni,10.3,9.718238,0.5817621231079109
+1098,oni,5.9,5.3342714,0.5657285690307621
+179,oni,10.2,9.583749,0.6162511825561516
+1134,oni,9.6,9.668004,0.06800403594970739
+1418,oni,8.0,7.8037148,0.19628524780273438
+72,oni,7.4,7.6135597,0.21355972290039027
+83,oni,6.0,6.248336,0.2483358383178711
+406,oni,7.7,7.693886,0.006114196777343928
+626,oni,6.3,6.4142413,0.11424131393432635
+455,oni,5.7,5.487079,0.21292085647583026
+583,oni,2.9,2.784802,0.11519804000854483
+500,oni,4.5,4.253699,0.24630117416381836
+672,oni,2.4,2.623148,0.22314796447753915
+1216,oni,10.0,9.5192,0.48079967498779297
+1057,oni,6.4,6.3955936,0.004406356811523793
+686,oni,8.7,8.237562,0.4624378204345696
+1369,oni,11.8,11.984631,0.18463058471679616
+1380,oni,6.9,6.8520164,0.04798355102539098
+1403,oni,5.0,4.905402,0.09459781646728516
+496,oni,6.1,6.430355,0.33035507202148473
+771,oni,3.3,3.3331902,0.03319020271301287
+571,oni,9.1,9.056726,0.04327354431152308
+815,oni,8.9,8.579678,0.320322418212891
+596,oni,6.8,6.777798,0.022201824188232244
+458,oni,6.3,6.2306204,0.06937961578369123
+755,oni,6.9,7.2296677,0.3296676635742184
+277,oni,3.2,2.558311,0.6416890144348146
+637,oni,6.6,6.266298,0.33370218276977504
+1146,oni,7.5,7.736816,0.2368159294128418
+224,oni,6.8,7.196848,0.39684791564941424
+869,oni,10.9,10.199179,0.7008213043212894
+428,oni,6.5,6.475257,0.024743080139160156
+514,oni,7.7,7.8003054,0.1003053665161131
+340,oni,5.2,5.115123,0.08487720489501971
+1256,oni,6.2,6.084713,0.1152870178222658
+8,oni,2.2,2.6536784,0.45367841720581037
+1405,oni,10.3,9.282177,1.0178230285644538
+1154,oni,8.9,9.2293825,0.3293825149536129
+1140,oni,5.6,5.3424892,0.2575107574462887
+846,oni,8.5,8.355557,0.14444255828857422
+555,oni,11.3,10.049599,1.2504013061523445
+1036,oni,6.5,6.2653823,0.2346177101135254
+1173,oni,9.2,8.834013,0.3659870147705071
+977,oni,8.8,8.352481,0.4475191116333015
+1419,oni,9.8,9.345617,0.4543827056884773
+429,oni,4.7,4.3966427,0.30335731506347674
+935,oni,9.5,8.687765,0.8122348785400391
+534,oni,4.0,3.9437454,0.05625462532043457
+552,oni,6.5,6.4038367,0.09616327285766602
+193,oni,4.4,4.2237315,0.1762684822082523
+560,oni,6.0,5.845188,0.15481185913085938
+386,oni,7.0,7.1123323,0.11233234405517578
+1039,oni,9.2,9.421019,0.2210186004638679
+1312,oni,2.9,3.1262274,0.22622737884521493
+1278,oni,11.7,11.715541,0.01554088592529368
+1396,oni,4.6,3.8423722,0.7576278209686276
+1367,oni,11.6,11.049116,0.550883865356445
+1242,oni,5.8,5.7851844,0.014815616607665838
+160,oni,5.3,5.2640634,0.03593664169311506
+283,oni,8.5,8.440965,0.059035301208496094
+199,oni,3.1,2.4848018,0.6151982307434083
+1347,oni,6.9,6.590419,0.3095811843872074
+350,oni,5.5,5.536312,0.036312103271484375
+64,oni,6.3,6.4002404,0.1002404212951662
+750,oni,8.8,9.02407,0.22406978607177663
+740,oni,8.8,8.772017,0.027983474731446023
+1391,oni,7.8,7.8338575,0.033857536315918146

.old/test/validate_result_2.csv

@@ -0,0 +1,201 @@
+songno,diff,상수,predicted_measure,error
+430,oni,1.5,1.7298045,0.22980451583862305
+401,oni,11.3,10.439279,0.860721397399903
+51,oni,3.2,2.998054,0.20194597244262713
+1268,oni,6.3,6.229351,0.07064895629882795
+676,oni,2.6,2.7677126,0.1677125930786132
+1260,oni,8.8,9.02407,0.22406978607177663
+83,oni,6.0,6.248336,0.2483358383178711
+621,oni,8.2,8.37565,0.17565040588378977
+304,oni,5.6,5.4814534,0.11854658126831019
+138,oni,7.7,7.8239584,0.12395839691162092
+207,oni,4.2,4.2350945,0.03509454727172834
+380,oni,4.6,4.6236706,0.023670578002930043
+680,oni,7.9,7.9497805,0.049780464172362926
+38,oni,5.7,5.6291122,0.07088775634765643
+938,oni,5.3,5.209937,0.09006290435790998
+689,oni,4.3,4.1817994,0.11820058822631818
+396,oni,5.8,6.0741944,0.2741944313049318
+1438,oni,7.4,7.79311,0.39310989379882777
+502,oni,8.8,8.9434595,0.14345951080322195
+610,oni,6.0,5.80618,0.19381999969482422
+1021,oni,7.0,6.821661,0.17833900451660156
+917,oni,6.7,5.3931637,1.3068363189697267
+907,oni,7.9,8.020189,0.12018928527831996
+209,oni,3.1,3.3612163,0.2612163066864013
+982,oni,9.9,9.456101,0.44389858245849645
+887,oni,7.2,7.051265,0.1487352371215822
+1031,oni,4.7,5.2932096,0.5932095527648924
+422,oni,3.0,3.2640421,0.2640421390533447
+1457,oni,5.4,5.485291,0.08529100418090785
+434,oni,4.1,4.1712093,0.07120933532714879
+1400,oni,1.6,2.1304216,0.5304216384887694
+974,oni,1.5,2.088167,0.5881669521331787
+611,oni,3.8,3.8965096,0.09650964736938494
+111,oni,4.7,4.5066733,0.19332666397094744
+266,oni,6.3,6.4265485,0.126548480987549
+302,oni,6.1,5.975836,0.12416419982910121
+1057,oni,6.4,6.3955936,0.004406356811523793
+843,oni,10.0,9.909245,0.09075546264648438
+1080,oni,8.0,8.098514,0.09851360321044922
+1380,oni,6.9,6.8520164,0.04798355102539098
+794,oni,7.6,6.201171,1.398829078674316
+356,oni,8.1,8.2789545,0.1789545059204105
+1408,oni,4.5,4.593335,0.09333515167236328
+34,oni,6.3,6.27956,0.020439910888671697
+734,oni,8.6,8.177258,0.42274246215820277
+1035,oni,5.4,5.1531134,0.2468866348266605
+310,oni,6.2,6.251908,0.051907825469970525
+1079,oni,5.6,5.3822875,0.21771249771118129
+344,oni,5.2,5.2222204,0.022220420837402166
+1241,oni,5.0,4.6463585,0.3536415100097656
+1251,oni,4.5,4.1223273,0.3776726722717285
+1302,oni,1.4,1.5472724,0.1472724437713624
+1234,oni,6.6,6.50439,0.09561023712158168
+492,oni,1.9,2.398783,0.49878296852111825
+915,oni,5.7,5.4805393,0.21946067810058612
+1218,oni,1.4,1.6566312,0.2566312313079835
+1054,oni,3.9,3.99446,0.09446010589599618
+450,oni,9.0,8.956529,0.043471336364746094
+578,oni,6.5,6.2173667,0.28263330459594727
+932,oni,7.5,7.25756,0.24244022369384766
+541,oni,5.6,5.419343,0.18065700531005824
+484,oni,5.3,5.3123913,0.012391281127929865
+951,oni,4.2,4.1377635,0.06223649978637713
+1346,oni,8.8,8.989751,0.18975086212158132
+1235,oni,7.8,7.8989544,0.09895439147949237
+1247,oni,10.0,9.550696,0.44930362701416016
+1092,oni,4.7,4.495435,0.2045647621154787
+770,oni,6.9,6.8700523,0.02994766235351598
+323,oni,4.3,4.327031,0.02703113555908221
+525,oni,6.6,6.3501854,0.24981460571289027
+1349,oni,6.3,5.845188,0.4548118591308592
+1406,oni,11.4,10.735728,0.6642717361450199
+452,oni,8.5,8.498182,0.0018177032470703125
+1236,oni,11.2,11.153378,0.046622467041014914
+1381,oni,7.3,7.888981,0.5889808654785158
+1125,oni,3.6,3.4895444,0.11045560836792001
+410,oni,6.6,6.6413183,0.0413183212280277
+775,oni,10.6,9.802061,0.7979389190673825
+716,oni,6.8,6.8853464,0.08534641265869158
+1136,oni,2.7,3.2574682,0.5574682235717772
+956,oni,2.1,2.52589,0.4258901119232177
+1040,oni,11.6,10.210405,1.3895946502685543
+1170,oni,6.9,7.0019712,0.10197124481201136
+1321,oni,5.8,5.845188,0.0451881408691408
+1292,oni,4.4,4.468697,0.0686970710754391
+1029,oni,1.6,1.3594662,0.24053380489349374
+1026,oni,4.1,3.8297968,0.2702032089233395
+556,oni,4.4,4.3982005,0.0017994880676273084
+1177,oni,3.4,3.871611,0.4716111183166505
+759,oni,9.0,8.656189,0.34381103515625
+537,oni,6.9,6.9082613,0.008261299133300426
+738,oni,7.9,7.6613083,0.2386917114257816
+905,oni,10.1,9.582528,0.517471885681152
+681,oni,7.0,7.2444925,0.2444925308227539
+36,oni,1.6,1.8594168,0.25941684246063224
+473,oni,5.4,5.588812,0.18881187438964808
+777,oni,10.8,9.979303,0.8206966400146491
+758,oni,10.8,9.948399,0.8516014099121101
+294,oni,4.3,4.851549,0.5515491485595705
+1355,oni,9.0,9.127055,0.12705516815185547
+751,oni,7.7,7.925638,0.22563819885253888
+713,oni,8.3,8.379393,0.07939262390136648
+976,oni,8.7,8.713133,0.013132858276367898
+723,oni,9.9,9.376441,0.5235589981079105
+100,oni,5.3,5.1635313,0.1364686965942381
+429,oni,4.7,4.3966427,0.30335731506347674
+341,oni,3.3,4.5935755,1.2935754776000978
+755,oni,6.9,7.2296677,0.3296676635742184
+823,oni,5.8,5.9487357,0.1487357139587404
+654,oni,4.5,4.328125,0.171875
+510,oni,5.3,5.2113357,0.08866434097290021
+1297,oni,9.4,9.196398,0.20360221862793004
+93,oni,5.9,6.05892,0.15891990661621058
+608,oni,3.5,3.7859712,0.28597116470336914
+183,oni,7.0,7.051115,0.05111503601074219
+1328,oni,9.9,9.375269,0.5247310638427738
+395,oni,7.7,7.528426,0.17157382965087908
+208,oni,4.5,4.484102,0.01589822769165039
+81,oni,4.4,5.829715,1.4297147750854489
+26,oni,4.1,5.500279,1.4002789497375492
+941,oni,4.6,4.5694156,0.030584430694579723
+970,oni,7.2,7.0208073,0.17919273376464862
+601,oni,3.5,3.6112447,0.11124467849731445
+1293,oni,3.8,4.05702,0.25702018737792987
+316,oni,7.1,6.120003,0.9799967765808102
+459,oni,5.6,5.5550127,0.044987297058105113
+968,oni,7.0,6.966305,0.033695220947265625
+1153,oni,1.8,2.1704838,0.37048382759094234
+375,oni,8.7,8.550151,0.14984912872314382
+39,oni,5.6,5.4114456,0.1885543823242184
+365,oni,10.0,8.945875,1.0541248321533203
+149,oni,4.8,5.013694,0.21369380950927752
+952,oni,10.0,9.769096,0.23090362548828125
+204,oni,2.4,2.719502,0.3195019721984864
+1197,oni,4.5,4.3139343,0.186065673828125
+945,oni,5.0,4.583455,0.41654491424560547
+1257,oni,9.0,8.956529,0.043471336364746094
+156,oni,5.8,5.7743926,0.025607395172118963
+470,oni,5.0,4.7149334,0.2850666046142578
+127,oni,5.9,5.9468956,0.04689559936523402
+1274,oni,4.8,4.9417405,0.14174051284790057
+552,oni,6.5,6.4038367,0.09616327285766602
+1422,oni,5.6,5.7453094,0.14530935287475621
+824,oni,1.3,1.4652954,0.16529543399810787
+154,oni,7.7,7.9209757,0.22097568511962873
+735,oni,6.4,6.3566375,0.043362522125244496
+164,oni,8.0,7.877853,0.1221470832824707
+139,oni,4.8,4.6042676,0.1957324028015135
+90,oni,5.6,5.5476413,0.05235872268676722
+368,oni,1.8,2.6962068,0.8962068080902099
+1048,oni,9.4,9.316141,0.08385887145996129
+615,oni,9.0,8.879463,0.12053680419921875
+994,oni,8.3,8.048112,0.2518880844116218
+576,oni,8.0,8.001074,0.0010738372802734375
+479,oni,4.9,4.708866,0.19113388061523473
+762,oni,10.3,9.964611,0.33538894653320384
+1111,oni,4.0,4.145422,0.14542198181152344
+719,oni,9.9,9.321081,0.5789188385009769
+1384,oni,9.3,9.402317,0.10231704711913991
+808,oni,6.2,6.37263,0.1726301193237303
+70,oni,6.1,5.9111953,0.18880472183227504
+571,oni,9.1,9.056726,0.04327354431152308
+488,oni,5.7,5.5653105,0.13468952178955096
+855,oni,5.7,4.9179006,0.7820994377136232
+491,oni,8.2,8.159609,0.04039115905761648
+498,oni,6.7,5.881829,0.8181712150573732
+79,oni,3.2,3.9292216,0.7292216300964354
+487,oni,3.5,3.6010914,0.10109138488769531
+922,oni,5.7,5.6340675,0.06593246459960955
+496,oni,6.1,6.430355,0.33035507202148473
+1249,oni,8.3,8.288431,0.011568832397461648
+846,oni,8.5,8.355557,0.14444255828857422
+700,oni,9.6,9.639939,0.03993930816650426
+103,oni,8.2,8.3685665,0.16856651306152415
+74,oni,4.5,4.536579,0.036579132080078125
+345,oni,4.5,3.4398248,1.0601751804351807
+404,oni,8.5,8.037392,0.46260833740234375
+1137,oni,7.1,7.197917,0.09791698455810582
+94,oni,9.9,9.473127,0.42687263488769567
+1412,oni,6.2,6.3555965,0.15559654235839826
+1300,oni,2.0,2.610785,0.6107850074768066
+315,oni,3.2,3.581308,0.3813078880310057
+364,oni,5.2,5.120959,0.07904119491577166
+1254,oni,5.9,5.639204,0.26079597473144567
+926,oni,3.7,3.3538485,0.3461515426635744
+732,oni,2.4,2.3426375,0.0573624610900878
+523,oni,5.1,4.688936,0.41106376647949183
+1425,oni,7.9,7.9018764,0.0018764495849605822
+1058,oni,6.4,6.4812675,0.08126745223998988
+543,oni,5.5,5.6081347,0.10813474655151367
+150,oni,6.3,6.2841473,0.015852737426757635
+471,oni,7.2,7.574646,0.3746459960937498
+666,oni,6.7,6.532254,0.1677457809448244
+833,oni,9.6,9.165459,0.4345413208007809
+765,oni,11.9,11.395069,0.5049308776855472
+166,oni,6.0,5.829277,0.17072296142578125
+1411,oni,9.1,8.809523,0.29047737121581996
+53,oni,6.4,6.430334,0.030334091186523082
+1176,oni,4.0,3.6295004,0.3704996109008789
+1098,oni,5.9,5.3342714,0.5657285690307621

.old/test/validate_result_3.csv

@@ -0,0 +1,201 @@
+songno,diff,상수,predicted_measure,error
+1349,oni,6.3,5.845188,0.4548118591308592
+1407,oni,2.8,2.7094536,0.09054641723632795
+1133,oni,5.7,5.578541,0.12145919799804705
+204,oni,2.4,2.719502,0.3195019721984864
+694,oni,3.5,4.001005,0.5010051727294922
+148,oni,9.7,9.349513,0.3504869461059563
+1332,oni,1.5,2.2710903,0.7710902690887451
+305,oni,3.7,3.7643456,0.06434564590454084
+1138,oni,4.1,4.1449833,0.04498329162597692
+437,oni,2.6,2.7578714,0.157871389389038
+1383,oni,9.3,9.224361,0.07563858032226634
+239,oni,7.6,9.329233,1.7292331695556644
+758,oni,10.8,9.948399,0.8516014099121101
+202,oni,7.9,8.050922,0.15092239379882777
+737,oni,9.9,9.715982,0.1840175628662113
+565,oni,4.7,4.9257283,0.22572832107543928
+820,oni,10.7,10.07424,0.6257602691650384
+851,oni,10.1,9.837338,0.26266155242919886
+1353,oni,2.8,3.3464475,0.5464474678039553
+15,oni,5.4,5.427866,0.027865982055663707
+198,oni,2.4,2.7392402,0.3392401695251466
+1418,oni,8.0,7.8037148,0.19628524780273438
+739,oni,7.1,7.3886776,0.2886775970458988
+396,oni,5.8,6.0741944,0.2741944313049318
+880,oni,5.0,5.18964,0.18964004516601562
+1327,oni,3.7,4.1225753,0.42257528305053693
+417,oni,3.3,3.3136492,0.013649177551269709
+661,oni,2.9,2.848915,0.05108489990234366
+933,oni,7.1,7.331907,0.23190679550170934
+230,oni,7.4,7.2959986,0.1040014266967777
+898,oni,3.8,4.029142,0.22914190292358416
+935,oni,9.5,8.687765,0.8122348785400391
+954,oni,9.7,9.612374,0.08762569427490163
+1237,oni,4.5,4.42432,0.07567977905273438
+1124,oni,2.8,2.4958134,0.30418663024902326
+926,oni,3.7,3.3538485,0.3461515426635744
+1227,oni,10.8,10.015021,0.7849786758422859
+287,oni,2.8,3.2971406,0.4971405982971193
+214,oni,7.8,7.899379,0.09937877655029315
+408,oni,3.6,3.389768,0.2102321147918702
+776,oni,7.7,7.219965,0.4800350189208986
+261,oni,7.4,7.6588984,0.2588983535766598
+60,oni,7.9,7.9494324,0.049432373046874645
+1052,oni,8.1,7.9752064,0.12479362487792933
+824,oni,1.3,1.4652954,0.16529543399810787
+982,oni,9.9,9.456101,0.44389858245849645
+161,oni,3.0,2.962341,0.0376589298248291
+1084,oni,2.6,3.0994637,0.49946370124816886
+1233,oni,4.1,3.903399,0.19660100936889613
+194,oni,9.4,9.450678,0.05067787170410121
+666,oni,6.7,6.532254,0.1677457809448244
+445,oni,7.1,7.3118544,0.21185436248779332
+315,oni,3.2,3.581308,0.3813078880310057
+1412,oni,6.2,6.3555965,0.15559654235839826
+1236,oni,11.2,11.153378,0.046622467041014914
+1055,oni,1.8,2.166351,0.36635107994079585
+1096,oni,5.1,3.3114526,1.7885473728179928
+1181,oni,3.3,3.7384562,0.4384562492370607
+608,oni,3.5,3.7859712,0.28597116470336914
+1312,oni,2.9,3.1262274,0.22622737884521493
+90,oni,5.6,5.5476413,0.05235872268676722
+861,oni,4.2,4.4093747,0.2093747138977049
+626,oni,6.3,6.4142413,0.11424131393432635
+469,oni,4.5,4.6131096,0.11310958862304688
+181,oni,8.5,8.829673,0.32967281341552734
+466,oni,3.0,4.420792,1.4207921028137207
+16,oni,8.3,8.095219,0.20478134155273509
+1067,oni,9.2,9.369811,0.1698110580444343
+111,oni,4.7,4.5066733,0.19332666397094744
+817,oni,8.0,8.014091,0.014090538024902344
+472,oni,5.9,6.009906,0.10990581512451136
+255,oni,5.5,5.6039057,0.10390567779541016
+1339,oni,4.5,4.5587,0.0587000846862793
+180,oni,7.3,7.391402,0.09140176773071307
+542,oni,4.2,4.0955725,0.10442752838134783
+483,oni,7.1,7.3588047,0.2588047027587894
+1320,oni,8.0,7.8748026,0.1251974105834961
+1370,oni,8.6,8.680969,0.08096923828125036
+1308,oni,8.4,8.65023,0.2502304077148434
+67,oni,7.6,7.61247,0.01247014999389684
+770,oni,6.9,6.8700523,0.02994766235351598
+1243,oni,1.1,1.061276,0.03872404098510751
+568,oni,7.5,7.328082,0.17191791534423828
+435,oni,5.7,5.4369884,0.26301164627075213
+947,oni,5.2,5.7635384,0.563538360595703
+358,oni,7.5,7.3555374,0.14446258544921875
+348,oni,3.9,3.733858,0.1661418914794921
+930,oni,7.1,7.913493,0.8134931564331058
+997,oni,4.0,4.2168503,0.21685028076171875
+1279,oni,8.4,8.668176,0.2681756973266598
+1013,oni,8.3,8.191456,0.10854415893554759
+1311,oni,6.7,6.815501,0.11550121307373029
+753,oni,5.3,5.040696,0.2593038558959959
+951,oni,4.2,4.1377635,0.06223649978637713
+157,oni,3.3,3.1216764,0.1783235549926756
+751,oni,7.7,7.925638,0.22563819885253888
+775,oni,10.6,9.802061,0.7979389190673825
+74,oni,4.5,4.536579,0.036579132080078125
+317,oni,2.9,3.59739,0.6973899364471436
+728,oni,4.9,4.727743,0.17225685119628942
+1382,oni,3.2,3.5220692,0.32206921577453596
+572,oni,8.3,8.527885,0.22788543701171804
+1452,oni,8.1,8.237134,0.13713397979736364
+1301,oni,2.0,2.1946142,0.19461417198181152
+39,oni,5.6,5.4114456,0.1885543823242184
+1450,oni,2.9,2.7411995,0.1588005065917968
+1034,oni,5.9,5.8499846,0.050015354156494496
+97,oni,7.5,7.3623753,0.13762474060058594
+1285,oni,2.4,2.4647148,0.06471476554870614
+390,oni,4.8,4.1861615,0.613838481903076
+952,oni,10.0,9.769096,0.23090362548828125
+208,oni,4.5,4.484102,0.01589822769165039
+193,oni,4.4,4.2237315,0.1762684822082523
+1116,oni,4.2,4.2934246,0.09342460632324201
+1315,oni,5.9,6.147772,0.24777183532714808
+233,oni,3.1,3.4442387,0.3442386627197265
+403,oni,5.3,5.125764,0.17423610687255842
+1277,oni,5.0,5.122813,0.12281322479248047
+639,oni,7.0,6.931817,0.06818294525146484
+498,oni,6.7,5.881829,0.8181712150573732
+179,oni,10.2,9.583749,0.6162511825561516
+793,oni,3.1,3.1706154,0.07061543464660636
+1160,oni,6.3,6.065338,0.23466186523437482
+1276,oni,9.2,9.156151,0.04384918212890554
+561,oni,5.5,5.413027,0.08697319030761719
+278,oni,3.8,4.709072,0.9090721130371096
+1083,oni,7.5,5.1249447,2.3750553131103516
+1019,oni,5.5,5.5518045,0.051804542541503906
+108,oni,6.8,6.9671035,0.1671034812927248
+929,oni,8.3,8.203068,0.09693222045898509
+988,oni,5.3,5.2196455,0.08035449981689435
+1154,oni,8.9,9.2293825,0.3293825149536129
+1401,oni,7.4,7.3951006,0.004899406433105824
+906,oni,9.1,9.289001,0.18900146484375036
+1368,oni,7.5,7.64944,0.14943981170654297
+1027,oni,3.9,4.020381,0.12038097381591806
+865,oni,5.6,5.53293,0.06707010269165004
+397,oni,7.9,7.942276,0.042276000976562145
+310,oni,6.2,6.251908,0.051907825469970525
+1110,oni,1.7,2.0302389,0.3302388668060303
+42,oni,1.1,2.9574435,1.8574434757232665
+431,oni,7.6,7.6814866,0.08148660659790075
+1381,oni,7.3,7.888981,0.5889808654785158
+395,oni,7.7,7.528426,0.17157382965087908
+514,oni,7.7,7.8003054,0.1003053665161131
+1362,oni,6.1,5.8406997,0.25930032730102504
+1305,oni,7.8,7.7893615,0.010638523101806463
+741,oni,3.5,4.1137867,0.6137866973876953
+958,oni,8.4,8.552175,0.15217456817626918
+1228,oni,4.1,3.8578901,0.24210987091064418
+203,oni,4.4,4.3201413,0.07985868453979528
+308,oni,5.2,4.990783,0.20921678543090838
+1266,oni,6.9,6.620808,0.279191875457764
+984,oni,4.2,4.5411806,0.3411806106567381
+764,oni,11.2,10.063731,1.1362688064575188
+702,oni,10.6,10.179897,0.42010269165039027
+418,oni,8.2,8.12527,0.07473011016845632
+1115,oni,3.8,3.9233134,0.1233133792877199
+534,oni,4.0,3.9437454,0.05625462532043457
+293,oni,3.9,4.5801744,0.6801744461059571
+1213,oni,5.3,5.074323,0.22567682266235334
+224,oni,6.8,7.196848,0.39684791564941424
+387,oni,4.9,4.804673,0.09532680511474645
+644,oni,4.4,4.3509674,0.049032592773437855
+1029,oni,1.6,1.3594662,0.24053380489349374
+1325,oni,8.2,8.245262,0.04526214599609446
+49,oni,4.4,4.6545525,0.2545524597167965
+1172,oni,11.9,11.048339,0.8516611099243168
+783,oni,6.0,5.882972,0.11702823638916016
+341,oni,3.3,4.5935755,1.2935754776000978
+527,oni,7.3,7.4277773,0.12777729034423846
+118,oni,3.0,2.7463155,0.25368452072143555
+448,oni,5.5,5.4255238,0.07447624206542969
+288,oni,8.2,8.141322,0.05867786407470632
+672,oni,2.4,2.623148,0.22314796447753915
+973,oni,4.0,3.9920924,0.007907629013061523
+1108,oni,10.4,9.689839,0.7101606369018558
+442,oni,10.3,9.434371,0.8656290054321296
+981,oni,3.3,3.3986528,0.09865279197692889
+319,oni,4.4,4.567095,0.16709480285644496
+1211,oni,6.0,5.586297,0.41370296478271484
+8,oni,2.2,2.6536784,0.45367841720581037
+750,oni,8.8,9.02407,0.22406978607177663
+1259,oni,6.1,6.0426464,0.05735359191894496
+992,oni,10.1,9.259041,0.8409591674804684
+507,oni,5.8,5.805967,0.005966854095459162
+1432,oni,10.0,9.590779,0.40922069549560547
+574,oni,8.6,8.241462,0.3585382461547848
+416,oni,6.0,5.845188,0.15481185913085938
+1114,oni,2.0,2.5233107,0.523310661315918
+1076,oni,2.1,2.0889874,0.011012649536132901
+63,oni,6.1,4.832204,1.2677961349487301
+31,oni,8.1,8.09505,0.004950141906737926
+115,oni,2.3,2.2282612,0.07173876762390119
+112,oni,6.5,6.330585,0.16941499710083008
+1044,oni,5.3,5.208531,0.091469097137451
+9,oni,2.0,2.2515485,0.25154852867126465
+1421,oni,8.1,8.032013,0.06798706054687464
+1218,oni,1.4,1.6566312,0.2566312313079835
+151,oni,4.8,4.623654,0.17634611129760724

.old/test/validate_result_4.csv

@@ -0,0 +1,201 @@
+songno,diff,상수,predicted_measure,error
+1200,oni,2.4,2.7606864,0.3606863975524903
+1420,oni,3.7,3.7231805,0.02318053245544416
+852,oni,6.6,6.5446224,0.05537757873535121
+406,oni,7.7,7.693886,0.006114196777343928
+112,oni,6.5,6.330585,0.16941499710083008
+37,oni,5.3,5.221942,0.0780580520629881
+187,oni,7.4,7.523672,0.12367210388183558
+1342,oni,2.2,2.297284,0.097283887863159
+1051,oni,8.7,8.919339,0.2193391799926765
+593,oni,10.6,9.9727,0.627299880981445
+242,oni,2.9,2.8971946,0.0028053760528563565
+1269,oni,3.2,2.7388601,0.4611398696899416
+663,oni,4.1,3.8161051,0.2838948726654049
+914,oni,8.7,8.76552,0.06552009582519602
+113,oni,5.7,5.845188,0.14518814086914045
+484,oni,5.3,5.3123913,0.012391281127929865
+469,oni,4.5,4.6131096,0.11310958862304688
+783,oni,6.0,5.882972,0.11702823638916016
+1188,oni,3.9,4.303166,0.4031659126281739
+1411,oni,9.1,8.809523,0.29047737121581996
+1043,oni,9.8,9.788532,0.011467742919922586
+32,oni,6.8,6.970789,0.17078895568847674
+47,oni,5.1,5.0852995,0.014700508117675426
+465,oni,6.9,6.837492,0.06250801086425817
+207,oni,4.2,4.2350945,0.03509454727172834
+824,oni,1.3,1.4652954,0.16529543399810787
+499,oni,8.8,8.752506,0.047493743896485086
+661,oni,2.9,2.848915,0.05108489990234366
+1379,oni,9.5,9.345877,0.15412330627441406
+662,oni,7.4,7.6551266,0.2551265716552731
+211,oni,7.8,7.7671843,0.032815742492675604
+794,oni,7.6,6.201171,1.398829078674316
+1173,oni,9.2,8.834013,0.3659870147705071
+1034,oni,5.9,5.8499846,0.050015354156494496
+1199,oni,4.3,4.3580647,0.05806465148925799
+450,oni,9.0,8.956529,0.043471336364746094
+1026,oni,4.1,3.8297968,0.2702032089233395
+539,oni,6.0,5.876666,0.12333393096923828
+1410,oni,3.0,2.8184204,0.18157958984375
+1412,oni,6.2,6.3555965,0.15559654235839826
+877,oni,1.3,1.7480694,0.44806940555572505
+808,oni,6.2,6.37263,0.1726301193237303
+623,oni,5.0,4.8877587,0.11224126815795898
+617,oni,5.6,6.084587,0.4845870971679691
+675,oni,5.2,5.875123,0.6751230239868162
+1209,oni,4.1,4.285281,0.18528118133544957
+462,oni,7.3,6.9185348,0.3814652442932127
+1191,oni,5.4,5.603399,0.20339879989623988
+418,oni,8.2,8.12527,0.07473011016845632
+1280,oni,3.9,4.2937455,0.393745517730713
+177,oni,2.0,2.734377,0.7343769073486328
+467,oni,2.9,2.862545,0.037454986572265536
+1071,oni,9.6,9.450291,0.14970932006835902
+471,oni,7.2,7.574646,0.3746459960937498
+220,oni,9.3,9.154587,0.1454132080078132
+1334,oni,4.0,4.209813,0.20981311798095703
+506,oni,5.6,3.9512367,1.648763275146484
+511,oni,2.3,2.8341331,0.5341331481933596
+1292,oni,4.4,4.468697,0.0686970710754391
+887,oni,7.2,7.051265,0.1487352371215822
+1145,oni,7.2,7.3018465,0.1018465042114256
+1332,oni,1.5,2.2710903,0.7710902690887451
+951,oni,4.2,4.1377635,0.06223649978637713
+1406,oni,11.4,10.735728,0.6642717361450199
+51,oni,3.2,2.998054,0.20194597244262713
+758,oni,10.8,9.948399,0.8516014099121101
+1422,oni,5.6,5.7453094,0.14530935287475621
+88,oni,8.1,7.868636,0.23136386871337855
+1073,oni,2.3,2.394328,0.09432811737060565
+549,oni,8.3,7.9438715,0.35612850189209055
+248,oni,6.2,6.27658,0.07657985687255842
+171,oni,9.6,9.666423,0.06642284393310582
+1258,oni,4.6,4.4628506,0.1371494293212887
+536,oni,6.7,6.679579,0.02042121887207049
+302,oni,6.1,5.975836,0.12416419982910121
+702,oni,10.6,10.179897,0.42010269165039027
+1415,oni,3.6,3.5990567,0.0009432792663575107
+981,oni,3.3,3.3986528,0.09865279197692889
+1169,oni,3.6,4.2148876,0.6148876190185546
+445,oni,7.1,7.3118544,0.21185436248779332
+858,oni,11.1,10.486108,0.6138921737670895
+1098,oni,5.9,5.3342714,0.5657285690307621
+1324,oni,9.3,9.130554,0.1694458007812507
+1226,oni,3.5,3.291145,0.20885491371154785
+1086,oni,10.6,10.089752,0.5102478027343746
+546,oni,6.8,6.8765535,0.07655353546142596
+1164,oni,5.4,5.845188,0.44518814086914027
+596,oni,6.8,6.777798,0.022201824188232244
+945,oni,5.0,4.583455,0.41654491424560547
+1184,oni,6.4,6.7978306,0.3978305816650387
+534,oni,4.0,3.9437454,0.05625462532043457
+1450,oni,2.9,2.7411995,0.1588005065917968
+1157,oni,4.6,4.355861,0.24413881301879847
+157,oni,3.3,3.1216764,0.1783235549926756
+1295,oni,1.3,1.133282,0.16671805381774907
+1107,oni,6.5,6.6370926,0.13709259033203125
+1196,oni,6.6,6.2691593,0.3308406829833981
+1070,oni,3.5,3.736236,0.2362360954284668
+1045,oni,8.0,7.8687744,0.1312255859375
+579,oni,7.2,7.204522,0.004522132873534979
+785,oni,8.5,8.505435,0.005434989929199219
+73,oni,8.5,8.4226885,0.07731151580810547
+879,oni,8.5,8.33581,0.16419029235839844
+869,oni,10.9,10.199179,0.7008213043212894
+340,oni,5.2,5.115123,0.08487720489501971
+401,oni,11.3,10.439279,0.860721397399903
+1306,oni,4.9,4.731516,0.1684841156005863
+297,oni,6.0,5.7628894,0.23711061477661133
+765,oni,11.9,11.395069,0.5049308776855472
+1248,oni,6.6,6.713462,0.1134618759155277
+384,oni,2.1,2.4554389,0.3554388523101806
+753,oni,5.3,5.040696,0.2593038558959959
+687,oni,4.3,4.1856074,0.11439256668090803
+505,oni,3.0,3.9480019,0.9480018615722656
+241,oni,2.6,2.7528176,0.15281763076782218
+110,oni,10.9,10.075016,0.8249839782714847
+881,oni,7.7,7.814657,0.11465721130371076
+544,oni,5.9,6.0155296,0.11552963256835902
+999,oni,10.8,9.808392,0.9916084289550788
+994,oni,8.3,8.048112,0.2518880844116218
+1099,oni,11.0,9.8232155,1.1767845153808594
+920,oni,4.9,4.69801,0.20199003219604528
+275,oni,3.5,3.5875602,0.08756017684936523
+1313,oni,5.3,5.4312534,0.13125343322753924
+1144,oni,2.7,2.792343,0.09234290122985822
+1172,oni,11.9,11.048339,0.8516611099243168
+266,oni,6.3,6.4265485,0.126548480987549
+14,oni,3.7,3.3840866,0.3159133911132814
+101,oni,1.4,1.6238686,0.22386858463287362
+115,oni,2.3,2.2282612,0.07173876762390119
+1354,oni,7.0,7.4809437,0.4809436798095703
+52,oni,5.2,5.161411,0.038589191436767756
+1418,oni,8.0,7.8037148,0.19628524780273438
+843,oni,10.0,9.909245,0.09075546264648438
+612,oni,6.7,6.813999,0.11399917602539045
+1323,oni,6.8,6.8110046,0.011004638671875178
+844,oni,8.1,8.043919,0.05608139038085902
+87,oni,5.6,5.726159,0.1261590957641605
+1090,oni,4.2,4.540201,0.3402011871337889
+111,oni,4.7,4.5066733,0.19332666397094744
+339,oni,6.8,6.780081,0.019919204711913885
+1198,oni,7.0,7.4989223,0.49892234802246094
+154,oni,7.7,7.9209757,0.22097568511962873
+291,oni,5.9,5.845188,0.05481185913085973
+739,oni,7.1,7.3886776,0.2886775970458988
+1137,oni,7.1,7.197917,0.09791698455810582
+725,oni,4.4,4.192197,0.20780315399169957
+1382,oni,3.2,3.5220692,0.32206921577453596
+253,oni,5.0,5.038617,0.03861713409423828
+1163,oni,8.3,8.122807,0.17719345092773509
+948,oni,8.3,8.308329,0.008328628540038352
+991,oni,6.3,6.262135,0.037864971160888494
+415,oni,1.8,2.84784,1.0478400707244873
+452,oni,8.5,8.498182,0.0018177032470703125
+503,oni,11.3,10.671308,0.6286924362182624
+1394,oni,10.3,10.036249,0.26375083923339915
+169,oni,3.7,3.7415326,0.04153256416320783
+1240,oni,4.2,4.3363075,0.13630752563476545
+515,oni,3.5,3.4623504,0.03764963150024414
+1106,oni,9.1,9.420722,0.3207220077514652
+685,oni,10.4,9.864383,0.5356172561645511
+412,oni,7.0,7.036152,0.036151885986328125
+453,oni,6.5,6.6112123,0.11121225357055664
+977,oni,8.8,8.352481,0.4475191116333015
+828,oni,4.1,4.206385,0.10638513565063512
+492,oni,1.9,2.398783,0.49878296852111825
+388,oni,9.0,9.137562,0.13756179809570312
+1131,oni,3.6,3.4794557,0.12054429054260263
+640,oni,8.5,8.225748,0.27425193786621094
+1122,oni,7.8,7.899935,0.09993476867675799
+888,oni,6.1,5.907568,0.1924320220947262
+1317,oni,11.2,10.480424,0.7195760726928704
+748,oni,10.8,9.854418,0.9455821990966804
+1147,oni,9.1,8.942031,0.15796909332275355
+666,oni,6.7,6.532254,0.1677457809448244
+489,oni,4.7,4.4432297,0.2567703247070314
+434,oni,4.1,4.1712093,0.07120933532714879
+706,oni,3.3,3.252023,0.047977018356323065
+1214,oni,6.0,6.4056187,0.40561866760253906
+80,oni,8.4,8.461607,0.06160697937011683
+897,oni,4.5,4.108138,0.3918619155883789
+565,oni,4.7,4.9257283,0.22572832107543928
+1377,oni,4.9,4.9466968,0.04669675827026332
+389,oni,6.2,5.975665,0.22433490753173846
+867,oni,9.0,9.131069,0.13106918334960938
+428,oni,6.5,6.475257,0.024743080139160156
+935,oni,9.5,8.687765,0.8122348785400391
+358,oni,7.5,7.3555374,0.14446258544921875
+279,oni,4.7,4.948142,0.24814205169677717
+954,oni,9.7,9.612374,0.08762569427490163
+703,oni,5.1,5.1526203,0.05262031555175817
+1059,oni,5.6,5.5380435,0.06195650100707972
+574,oni,8.6,8.241462,0.3585382461547848
+1309,oni,6.5,6.507756,0.007756233215332031
+998,oni,5.0,5.098602,0.0986018180847168
+124,oni,5.5,5.4560986,0.04390144348144531
+572,oni,8.3,8.527885,0.22788543701171804
+1134,oni,9.6,9.668004,0.06800403594970739
+738,oni,7.9,7.6613083,0.2386917114257816
+775,oni,10.6,9.802061,0.7979389190673825

.old/test/validate_result_5.csv

@@ -0,0 +1,201 @@
+songno,diff,상수,predicted_measure,error
+173,oni,6.0,5.917573,0.0824270248413086
+1061,oni,3.9,3.5892773,0.3107227325439452
+1104,oni,9.7,9.212923,0.4870769500732415
+424,oni,6.3,6.343747,0.04374713897705096
+339,oni,6.8,6.780081,0.019919204711913885
+730,oni,5.1,4.697263,0.40273723602294886
+1412,oni,6.2,6.3555965,0.15559654235839826
+1282,oni,3.8,4.334233,0.534232807159424
+109,oni,4.6,4.4378033,0.16219673156738246
+441,oni,10.7,10.057486,0.6425144195556634
+771,oni,3.3,3.3331902,0.03319020271301287
+180,oni,7.3,7.391402,0.09140176773071307
+140,oni,3.3,3.0745137,0.2254863262176512
+1233,oni,4.1,3.903399,0.19660100936889613
+1295,oni,1.3,1.133282,0.16671805381774907
+71,oni,6.0,5.845188,0.15481185913085938
+1012,oni,9.7,9.215172,0.48482818603515554
+623,oni,5.0,4.8877587,0.11224126815795898
+361,oni,8.0,8.176176,0.1761760711669922
+1304,oni,4.7,4.2533813,0.4466187477111818
+644,oni,4.4,4.3509674,0.049032592773437855
+1323,oni,6.8,6.8110046,0.011004638671875178
+491,oni,8.2,8.159609,0.04039115905761648
+1391,oni,7.8,7.8338575,0.033857536315918146
+267,oni,3.6,3.689395,0.08939495086669913
+1324,oni,9.3,9.130554,0.1694458007812507
+1379,oni,9.5,9.345877,0.15412330627441406
+846,oni,8.5,8.355557,0.14444255828857422
+1267,oni,4.7,4.7183113,0.018311309814452947
+1105,oni,5.5,4.235367,1.2646331787109375
+880,oni,5.0,5.18964,0.18964004516601562
+103,oni,8.2,8.3685665,0.16856651306152415
+1226,oni,3.5,3.291145,0.20885491371154785
+911,oni,4.7,4.432701,0.2672988891601564
+1403,oni,5.0,4.905402,0.09459781646728516
+106,oni,5.1,4.919034,0.18096599578857386
+250,oni,3.0,3.369525,0.3695249557495117
+1018,oni,4.9,4.7685223,0.13147773742675817
+545,oni,4.8,4.993164,0.19316406250000018
+386,oni,7.0,7.1123323,0.11233234405517578
+184,oni,5.2,4.5262094,0.6737906455993654
+790,oni,2.5,2.8459394,0.34593939781188965
+1056,oni,4.7,4.598622,0.10137815475463885
+459,oni,5.6,5.5550127,0.044987297058105113
+1200,oni,2.4,2.7606864,0.3606863975524903
+804,oni,5.7,5.78445,0.084450054168701
+345,oni,4.5,3.4398248,1.0601751804351807
+1276,oni,9.2,9.156151,0.04384918212890554
+1455,oni,4.6,4.585906,0.014093971252441051
+961,oni,10.2,9.469546,0.7304536819458001
+26,oni,4.1,5.500279,1.4002789497375492
+640,oni,8.5,8.225748,0.27425193786621094
+1420,oni,3.7,3.7231805,0.02318053245544416
+126,oni,4.6,4.485203,0.11479721069335902
+1117,oni,2.9,3.0611537,0.16115365028381357
+570,oni,9.0,8.652149,0.3478507995605469
+1083,oni,7.5,5.1249447,2.3750553131103516
+1428,oni,5.0,4.738045,0.26195478439331055
+190,oni,7.7,7.912362,0.21236209869384748
+1465,oni,6.8,7.052265,0.2522651672363283
+967,oni,8.8,7.3483477,1.4516523361206062
+1397,oni,3.3,3.8423722,0.5423721790313722
+1062,oni,5.0,4.9341946,0.06580543518066406
+713,oni,8.3,8.379393,0.07939262390136648
+524,oni,8.9,9.106796,0.20679626464843714
+627,oni,11.5,10.365782,1.1342182159423828
+1092,oni,4.7,4.495435,0.2045647621154787
+1037,oni,9.2,8.873722,0.32627792358398366
+1287,oni,3.2,3.1268172,0.07318277359008807
+454,oni,7.7,7.780956,0.0809557914733885
+63,oni,6.1,4.832204,1.2677961349487301
+474,oni,5.3,5.03934,0.2606599807739256
+724,oni,8.5,8.412328,0.08767223358154297
+689,oni,4.3,4.1817994,0.11820058822631818
+1386,oni,5.4,5.1236534,0.276346588134766
+115,oni,2.3,2.2282612,0.07173876762390119
+1438,oni,7.4,7.79311,0.39310989379882777
+753,oni,5.3,5.040696,0.2593038558959959
+266,oni,6.3,6.4265485,0.126548480987549
+1313,oni,5.3,5.4312534,0.13125343322753924
+1370,oni,8.6,8.680969,0.08096923828125036
+1122,oni,7.8,7.899935,0.09993476867675799
+1345,oni,4.5,4.4040623,0.09593772888183594
+167,oni,7.1,7.137214,0.0372141838073734
+828,oni,4.1,4.206385,0.10638513565063512
+358,oni,7.5,7.3555374,0.14446258544921875
+107,oni,4.6,4.2985578,0.3014422416687008
+1230,oni,3.4,3.2455196,0.15448036193847647
+1072,oni,9.8,9.565875,0.234124946594239
+917,oni,6.7,5.3931637,1.3068363189697267
+293,oni,3.9,4.5801744,0.6801744461059571
+11,oni,1.5,2.3513732,0.8513731956481934
+1351,oni,9.3,9.762559,0.4625589370727532
+1393,oni,5.2,4.9068255,0.293174457550049
+1302,oni,1.4,1.5472724,0.1472724437713624
+122,oni,1.9,2.9562376,1.056237554550171
+1109,oni,6.8,6.6508837,0.1491163253784178
+604,oni,1.0,1.1668034,0.16680335998535156
+960,oni,10.1,9.670273,0.42972717285156214
+349,oni,6.8,6.479022,0.3209779739379881
+778,oni,11.0,10.150578,0.8494224548339844
+1188,oni,3.9,4.303166,0.4031659126281739
+1010,oni,4.8,5.020252,0.2202522277832033
+763,oni,10.8,9.695452,1.1045482635498054
+1332,oni,1.5,2.2710903,0.7710902690887451
+541,oni,5.6,5.419343,0.18065700531005824
+314,oni,1.3,1.4790187,0.17901868820190425
+218,oni,7.7,7.211827,0.48817319869995135
+1280,oni,3.9,4.2937455,0.393745517730713
+751,oni,7.7,7.925638,0.22563819885253888
+952,oni,10.0,9.769096,0.23090362548828125
+131,oni,8.2,8.129029,0.07097072601318288
+277,oni,3.2,2.558311,0.6416890144348146
+1014,oni,5.3,5.1404095,0.15959053039550763
+700,oni,9.6,9.639939,0.03993930816650426
+348,oni,3.9,3.733858,0.1661418914794921
+408,oni,3.6,3.389768,0.2102321147918702
+819,oni,8.6,8.7340145,0.1340145111083988
+121,oni,9.3,8.394699,0.9053009033203132
+401,oni,11.3,10.439279,0.860721397399903
+1377,oni,4.9,4.9466968,0.04669675827026332
+360,oni,9.2,9.504893,0.3048933029174812
+365,oni,10.0,8.945875,1.0541248321533203
+487,oni,3.5,3.6010914,0.10109138488769531
+461,oni,7.1,7.155115,0.05511512756347692
+792,oni,3.7,4.1929874,0.4929874420166014
+1069,oni,8.0,7.9372826,0.06271743774414062
+621,oni,8.2,8.37565,0.17565040588378977
+968,oni,7.0,6.966305,0.033695220947265625
+297,oni,6.0,5.7628894,0.23711061477661133
+1212,oni,1.4,1.6360978,0.23609778881073007
+1232,oni,7.2,6.9321103,0.2678896903991701
+845,oni,7.5,7.5258875,0.025887489318847656
+259,oni,6.6,6.505166,0.09483394622802699
+432,oni,6.6,6.7569847,0.15698471069335973
+1059,oni,5.6,5.5380435,0.06195650100707972
+1157,oni,4.6,4.355861,0.24413881301879847
+43,oni,8.1,8.08124,0.018760299682616832
+653,oni,6.7,6.263284,0.4367157936096193
+865,oni,5.6,5.53293,0.06707010269165004
+209,oni,3.1,3.3612163,0.2612163066864013
+910,oni,7.3,7.5151834,0.21518344879150408
+1120,oni,8.2,8.383063,0.18306331634521555
+896,oni,11.1,10.215234,0.8847661972045895
+480,oni,3.2,3.3859277,0.18592767715454084
+1053,oni,4.6,4.43534,0.1646600723266598
+457,oni,6.5,6.5087705,0.008770465850830078
+443,oni,4.4,4.634786,0.23478612899780238
+566,oni,7.6,7.663207,0.06320705413818395
+136,oni,9.1,9.059532,0.040467834472655895
+212,oni,5.7,5.6477895,0.052210521697998225
+928,oni,10.3,9.718238,0.5817621231079109
+1168,oni,4.0,4.149542,0.14954185485839844
+554,oni,5.9,5.8169847,0.08301534652709996
+867,oni,9.0,9.131069,0.13106918334960938
+1255,oni,10.4,9.691372,0.708628082275391
+1154,oni,8.9,9.2293825,0.3293825149536129
+1110,oni,1.7,2.0302389,0.3302388668060303
+852,oni,6.6,6.5446224,0.05537757873535121
+1164,oni,5.4,5.845188,0.44518814086914027
+1133,oni,5.7,5.578541,0.12145919799804705
+261,oni,7.4,7.6588984,0.2588983535766598
+319,oni,4.4,4.567095,0.16709480285644496
+907,oni,7.9,8.020189,0.12018928527831996
+721,oni,9.3,8.969906,0.33009414672851634
+722,oni,9.4,9.258499,0.14150085449218786
+194,oni,9.4,9.450678,0.05067787170410121
+1090,oni,4.2,4.540201,0.3402011871337889
+988,oni,5.3,5.2196455,0.08035449981689435
+13,oni,7.2,7.3293734,0.1293733596801756
+416,oni,6.0,5.845188,0.15481185913085938
+304,oni,5.6,5.4814534,0.11854658126831019
+1408,oni,4.5,4.593335,0.09333515167236328
+1160,oni,6.3,6.065338,0.23466186523437482
+1262,oni,4.3,4.327031,0.02703113555908221
+1180,oni,6.5,6.422063,0.07793712615966797
+820,oni,10.7,10.07424,0.6257602691650384
+762,oni,10.3,9.964611,0.33538894653320384
+1165,oni,7.9,7.9121323,0.012132263183593395
+118,oni,3.0,2.7463155,0.25368452072143555
+72,oni,7.4,7.6135597,0.21355972290039027
+1421,oni,8.1,8.032013,0.06798706054687464
+242,oni,2.9,2.8971946,0.0028053760528563565
+886,oni,1.9,1.9517515,0.05175147056579599
+864,oni,8.1,8.247111,0.14711132049560582
+733,oni,3.8,3.5901544,0.2098455905914305
+281,oni,4.2,4.1172986,0.08270139694213885
+214,oni,7.8,7.899379,0.09937877655029315
+423,oni,5.9,6.2007046,0.3007045745849606
+1150,oni,4.1,3.2386503,0.8613496780395504
+1451,oni,5.1,5.1919827,0.09198274612426793
+951,oni,4.2,4.1377635,0.06223649978637713
+1039,oni,9.2,9.421019,0.2210186004638679
+53,oni,6.4,6.430334,0.030334091186523082
+1103,oni,4.8,4.630667,0.16933279037475568
+300,oni,4.0,4.3146954,0.3146953582763672
+377,oni,4.3,3.7255836,0.5744164466857908
+1220,oni,3.6,3.5168161,0.08318386077880868
+933,oni,7.1,7.331907,0.23190679550170934
+860,oni,10.6,10.310149,0.28985080718994105

.old/test/validate_result_6.csv

@@ -0,0 +1,201 @@
+songno,diff,상수,predicted_measure,error
+644,oni,4.4,4.3509674,0.049032592773437855
+1159,oni,7.1,7.237465,0.1374649047851566
+1322,oni,7.9,7.850716,0.0492838859558109
+387,oni,4.9,4.804673,0.09532680511474645
+434,oni,4.1,4.1712093,0.07120933532714879
+539,oni,6.0,5.876666,0.12333393096923828
+894,oni,1.7,1.6517099,0.04829008579254146
+541,oni,5.6,5.419343,0.18065700531005824
+306,oni,6.3,6.4742217,0.17422170639038104
+1306,oni,4.9,4.731516,0.1684841156005863
+267,oni,3.6,3.689395,0.08939495086669913
+920,oni,4.9,4.69801,0.20199003219604528
+1086,oni,10.6,10.089752,0.5102478027343746
+1214,oni,6.0,6.4056187,0.40561866760253906
+990,oni,10.9,9.670632,1.2293676376342777
+354,oni,4.8,4.4523187,0.3476813316345213
+90,oni,5.6,5.5476413,0.05235872268676722
+307,oni,4.1,3.9636607,0.1363392829895016
+761,oni,10.4,9.505325,0.8946746826171879
+896,oni,11.1,10.215234,0.8847661972045895
+874,oni,7.0,7.0029125,0.0029125213623046875
+465,oni,6.9,6.837492,0.06250801086425817
+833,oni,9.6,9.165459,0.4345413208007809
+983,oni,8.9,8.690888,0.20911159515380895
+1165,oni,7.9,7.9121323,0.012132263183593395
+866,oni,7.9,7.9159746,0.015974617004394176
+834,oni,9.7,9.58336,0.11664028167724538
+398,oni,6.3,6.204332,0.09566812515258771
+835,oni,10.2,9.703024,0.49697608947753835
+1304,oni,4.7,4.2533813,0.4466187477111818
+111,oni,4.7,4.5066733,0.19332666397094744
+797,oni,8.8,7.9628973,0.8371026992797859
+875,oni,5.1,5.04557,0.0544301033020016
+534,oni,4.0,3.9437454,0.05625462532043457
+182,oni,6.4,6.657069,0.2570692062377926
+467,oni,2.9,2.862545,0.037454986572265536
+925,oni,5.2,4.9190726,0.28092737197875994
+260,oni,8.8,7.5779324,1.2220676422119148
+221,oni,6.7,6.725834,0.025833892822265447
+184,oni,5.2,4.5262094,0.6737906455993654
+565,oni,4.7,4.9257283,0.22572832107543928
+915,oni,5.7,5.4805393,0.21946067810058612
+1194,oni,4.2,3.9531407,0.24685926437377947
+288,oni,8.2,8.141322,0.05867786407470632
+385,oni,5.1,5.0345116,0.06548843383789027
+668,oni,3.4,3.5601308,0.16013083457946786
+595,oni,9.5,9.133109,0.36689090728759766
+505,oni,3.0,3.9480019,0.9480018615722656
+1307,oni,6.9,6.8420954,0.0579046249389652
+1243,oni,1.1,1.061276,0.03872404098510751
+1241,oni,5.0,4.6463585,0.3536415100097656
+1298,oni,5.7,5.728607,0.028607177734374822
+1218,oni,1.4,1.6566312,0.2566312313079835
+1153,oni,1.8,2.1704838,0.37048382759094234
+675,oni,5.2,5.875123,0.6751230239868162
+108,oni,6.8,6.9671035,0.1671034812927248
+732,oni,2.4,2.3426375,0.0573624610900878
+34,oni,6.3,6.27956,0.020439910888671697
+120,oni,11.9,11.8358965,0.06410350799560582
+933,oni,7.1,7.331907,0.23190679550170934
+348,oni,3.9,3.733858,0.1661418914794921
+372,oni,6.4,6.345234,0.05476608276367223
+1421,oni,8.1,8.032013,0.06798706054687464
+1181,oni,3.3,3.7384562,0.4384562492370607
+136,oni,9.1,9.059532,0.040467834472655895
+504,oni,6.1,5.845188,0.254811859130859
+630,oni,9.3,8.637385,0.6626146316528327
+279,oni,4.7,4.948142,0.24814205169677717
+1015,oni,8.7,8.804349,0.10434894561767649
+343,oni,6.4,5.75634,0.6436599731445316
+1171,oni,6.1,6.1189337,0.0189336776733402
+237,oni,5.3,6.2021265,0.9021265029907228
+1278,oni,11.7,11.715541,0.01554088592529368
+1072,oni,9.8,9.565875,0.234124946594239
+621,oni,8.2,8.37565,0.17565040588378977
+62,oni,6.1,6.0770845,0.022915458679198863
+291,oni,5.9,5.845188,0.05481185913085973
+333,oni,1.1,1.4720416,0.3720416069030761
+32,oni,6.8,6.970789,0.17078895568847674
+1053,oni,4.6,4.43534,0.1646600723266598
+970,oni,7.2,7.0208073,0.17919273376464862
+1233,oni,4.1,3.903399,0.19660100936889613
+1255,oni,10.4,9.691372,0.708628082275391
+133,oni,4.8,5.009082,0.2090818405151369
+536,oni,6.7,6.679579,0.02042121887207049
+700,oni,9.6,9.639939,0.03993930816650426
+1084,oni,2.6,3.0994637,0.49946370124816886
+897,oni,4.5,4.108138,0.3918619155883789
+1352,oni,8.0,8.376083,0.3760833740234375
+321,oni,6.3,6.2436576,0.05634241104125959
+428,oni,6.5,6.475257,0.024743080139160156
+626,oni,6.3,6.4142413,0.11424131393432635
+287,oni,2.8,3.2971406,0.4971405982971193
+449,oni,8.1,7.9776797,0.12232027053832972
+617,oni,5.6,6.084587,0.4845870971679691
+946,oni,3.4,3.4010017,0.0010016918182373935
+1390,oni,5.7,5.536209,0.16379089355468768
+430,oni,1.5,1.7298045,0.22980451583862305
+60,oni,7.9,7.9494324,0.049432373046874645
+988,oni,5.3,5.2196455,0.08035449981689435
+530,oni,4.7,4.619874,0.08012599945068377
+730,oni,5.1,4.697263,0.40273723602294886
+594,oni,9.7,9.348086,0.35191364288330007
+359,oni,8.3,8.350011,0.05001087188720632
+209,oni,3.1,3.3612163,0.2612163066864013
+1227,oni,10.8,10.015021,0.7849786758422859
+1317,oni,11.2,10.480424,0.7195760726928704
+824,oni,1.3,1.4652954,0.16529543399810787
+1228,oni,4.1,3.8578901,0.24210987091064418
+911,oni,4.7,4.432701,0.2672988891601564
+615,oni,9.0,8.879463,0.12053680419921875
+699,oni,9.7,9.337942,0.36205787658691335
+422,oni,3.0,3.2640421,0.2640421390533447
+684,oni,7.5,7.035693,0.4643068313598633
+404,oni,8.5,8.037392,0.46260833740234375
+549,oni,8.3,7.9438715,0.35612850189209055
+452,oni,8.5,8.498182,0.0018177032470703125
+1400,oni,1.6,2.1304216,0.5304216384887694
+1061,oni,3.9,3.5892773,0.3107227325439452
+1402,oni,5.8,5.8168125,0.01681251525878924
+183,oni,7.0,7.051115,0.05111503601074219
+317,oni,2.9,3.59739,0.6973899364471436
+776,oni,7.7,7.219965,0.4800350189208986
+1350,oni,11.9,11.538368,0.3616317749023441
+714,oni,9.4,8.93766,0.4623397827148441
+1145,oni,7.2,7.3018465,0.1018465042114256
+868,oni,9.5,8.869226,0.6307735443115234
+82,oni,5.7,5.5701885,0.129811477661133
+1183,oni,6.5,6.3902035,0.10979652404785156
+612,oni,6.7,6.813999,0.11399917602539045
+1361,oni,1.5,1.7912393,0.29123926162719727
+1368,oni,7.5,7.64944,0.14943981170654297
+1342,oni,2.2,2.297284,0.097283887863159
+995,oni,6.4,6.4859424,0.08594236373901332
+930,oni,7.1,7.913493,0.8134931564331058
+1230,oni,3.4,3.2455196,0.15448036193847647
+1058,oni,6.4,6.4812675,0.08126745223998988
+1391,oni,7.8,7.8338575,0.033857536315918146
+323,oni,4.3,4.327031,0.02703113555908221
+800,oni,7.3,7.135235,0.16476516723632795
+1386,oni,5.4,5.1236534,0.276346588134766
+688,oni,6.3,6.2337136,0.06628637313842756
+451,oni,7.2,8.086525,0.8865249633789061
+276,oni,3.7,3.782309,0.08230905532836896
+1116,oni,4.2,4.2934246,0.09342460632324201
+914,oni,8.7,8.76552,0.06552009582519602
+454,oni,7.7,7.780956,0.0809557914733885
+880,oni,5.0,5.18964,0.18964004516601562
+1336,oni,2.9,3.0399966,0.13999662399292
+346,oni,6.4,6.5222197,0.12221965789794886
+1424,oni,6.4,6.2954216,0.10457839965820348
+1209,oni,4.1,4.285281,0.18528118133544957
+1226,oni,3.5,3.291145,0.20885491371154785
+1080,oni,8.0,8.098514,0.09851360321044922
+746,oni,8.0,8.106806,0.10680580139160156
+340,oni,5.2,5.115123,0.08487720489501971
+1203,oni,4.9,4.7891917,0.11080827713012731
+1118,oni,1.2,2.1188107,0.9188106536865235
+1295,oni,1.3,1.133282,0.16671805381774907
+1328,oni,9.9,9.375269,0.5247310638427738
+429,oni,4.7,4.3966427,0.30335731506347674
+1030,oni,4.9,5.055099,0.15509901046752894
+1457,oni,5.4,5.485291,0.08529100418090785
+471,oni,7.2,7.574646,0.3746459960937498
+1240,oni,4.2,4.3363075,0.13630752563476545
+56,oni,6.6,6.737589,0.13758888244628942
+360,oni,9.2,9.504893,0.3048933029174812
+1388,oni,6.1,5.9238286,0.17617139816284144
+304,oni,5.6,5.4814534,0.11854658126831019
+1323,oni,6.8,6.8110046,0.011004638671875178
+1191,oni,5.4,5.603399,0.20339879989623988
+881,oni,7.7,7.814657,0.11465721130371076
+999,oni,10.8,9.808392,0.9916084289550788
+1305,oni,7.8,7.7893615,0.010638523101806463
+1375,oni,1.1,1.3882558,0.2882558345794677
+278,oni,3.8,4.709072,0.9090721130371096
+523,oni,5.1,4.688936,0.41106376647949183
+856,oni,8.8,8.237783,0.5622165679931648
+1076,oni,2.1,2.0889874,0.011012649536132901
+859,oni,10.2,9.688175,0.5118247985839837
+510,oni,5.3,5.2113357,0.08866434097290021
+860,oni,10.6,10.310149,0.28985080718994105
+1098,oni,5.9,5.3342714,0.5657285690307621
+96,oni,8.8,8.476414,0.3235862731933601
+1211,oni,6.0,5.586297,0.41370296478271484
+103,oni,8.2,8.3685665,0.16856651306152415
+302,oni,6.1,5.975836,0.12416419982910121
+635,oni,8.5,8.438788,0.061211585998535156
+1346,oni,8.8,8.989751,0.18975086212158132
+851,oni,10.1,9.837338,0.26266155242919886
+264,oni,4.8,4.4578743,0.3421257019042967
+1401,oni,7.4,7.3951006,0.004899406433105824
+242,oni,2.9,2.8971946,0.0028053760528563565
+118,oni,3.0,2.7463155,0.25368452072143555
+65,oni,6.2,6.0759215,0.12407846450805682
+1329,oni,7.4,7.584384,0.18438396453857386
+1099,oni,11.0,9.8232155,1.1767845153808594
+1215,oni,8.3,8.44902,0.1490203857421868
+972,oni,7.5,7.1393,0.3607001304626465
+1172,oni,11.9,11.048339,0.8516611099243168

.old/test/validate_result_7.csv

@@ -0,0 +1,201 @@
+songno,diff,상수,predicted_measure,error
+723,oni,9.9,9.376441,0.5235589981079105
+879,oni,8.5,8.33581,0.16419029235839844
+293,oni,3.9,4.5801744,0.6801744461059571
+297,oni,6.0,5.7628894,0.23711061477661133
+988,oni,5.3,5.2196455,0.08035449981689435
+726,oni,4.6,4.15228,0.44772014617919886
+236,oni,2.2,2.4775774,0.2775774478912352
+1011,oni,5.8,5.5946207,0.20537929534912092
+422,oni,3.0,3.2640421,0.2640421390533447
+1339,oni,4.5,4.5587,0.0587000846862793
+845,oni,7.5,7.5258875,0.025887489318847656
+13,oni,7.2,7.3293734,0.1293733596801756
+459,oni,5.6,5.5550127,0.044987297058105113
+548,oni,6.5,6.6141577,0.11415767669677734
+399,oni,7.6,7.566705,0.033294773101806285
+1110,oni,1.7,2.0302389,0.3302388668060303
+1085,oni,10.0,9.612744,0.3872556686401367
+96,oni,8.8,8.476414,0.3235862731933601
+1035,oni,5.4,5.1531134,0.2468866348266605
+937,oni,5.1,5.195075,0.0950750350952152
+316,oni,7.1,6.120003,0.9799967765808102
+1103,oni,4.8,4.630667,0.16933279037475568
+984,oni,4.2,4.5411806,0.3411806106567381
+983,oni,8.9,8.690888,0.20911159515380895
+501,oni,7.6,7.7123227,0.11232271194458043
+1376,oni,1.9,2.5668554,0.6668554306030274
+144,oni,2.5,2.725388,0.2253880500793457
+390,oni,4.8,4.1861615,0.613838481903076
+270,oni,6.9,6.9537883,0.05378828048706019
+1400,oni,1.6,2.1304216,0.5304216384887694
+204,oni,2.4,2.719502,0.3195019721984864
+552,oni,6.5,6.4038367,0.09616327285766602
+467,oni,2.9,2.862545,0.037454986572265536
+1287,oni,3.2,3.1268172,0.07318277359008807
+527,oni,7.3,7.4277773,0.12777729034423846
+1105,oni,5.5,4.235367,1.2646331787109375
+640,oni,8.5,8.225748,0.27425193786621094
+577,oni,9.8,9.342824,0.4571760177612312
+1073,oni,2.3,2.394328,0.09432811737060565
+349,oni,6.8,6.479022,0.3209779739379881
+997,oni,4.0,4.2168503,0.21685028076171875
+134,oni,6.4,6.312104,0.08789577484130895
+400,oni,7.6,7.566705,0.033294773101806285
+1371,oni,6.5,6.3326335,0.16736650466918945
+402,oni,6.5,6.452997,0.04700279235839844
+1079,oni,5.6,5.3822875,0.21771249771118129
+1204,oni,3.1,3.1182175,0.01821746826171866
+51,oni,3.2,2.998054,0.20194597244262713
+1332,oni,1.5,2.2710903,0.7710902690887451
+1312,oni,2.9,3.1262274,0.22622737884521493
+925,oni,5.2,4.9190726,0.28092737197875994
+1331,oni,2.2,2.7129545,0.512954521179199
+860,oni,10.6,10.310149,0.28985080718994105
+1177,oni,3.4,3.871611,0.4716111183166505
+368,oni,1.8,2.6962068,0.8962068080902099
+262,oni,6.8,6.8975925,0.09759254455566424
+21,oni,8.1,7.9611,0.13889989852905238
+483,oni,7.1,7.3588047,0.2588047027587894
+1021,oni,7.0,6.821661,0.17833900451660156
+169,oni,3.7,3.7415326,0.04153256416320783
+574,oni,8.6,8.241462,0.3585382461547848
+266,oni,6.3,6.4265485,0.126548480987549
+1369,oni,11.8,11.984631,0.18463058471679616
+415,oni,1.8,2.84784,1.0478400707244873
+265,oni,10.2,9.425652,0.7743484497070305
+1451,oni,5.1,5.1919827,0.09198274612426793
+1324,oni,9.3,9.130554,0.1694458007812507
+90,oni,5.6,5.5476413,0.05235872268676722
+131,oni,8.2,8.129029,0.07097072601318288
+1111,oni,4.0,4.145422,0.14542198181152344
+212,oni,5.7,5.6477895,0.052210521697998225
+1259,oni,6.1,6.0426464,0.05735359191894496
+820,oni,10.7,10.07424,0.6257602691650384
+1271,oni,4.8,4.724819,0.07518081665039045
+530,oni,4.7,4.619874,0.08012599945068377
+680,oni,7.9,7.9497805,0.049780464172362926
+694,oni,3.5,4.001005,0.5010051727294922
+1176,oni,4.0,3.6295004,0.3704996109008789
+709,oni,2.5,3.2942243,0.7942242622375488
+536,oni,6.7,6.679579,0.02042121887207049
+211,oni,7.8,7.7671843,0.032815742492675604
+1053,oni,4.6,4.43534,0.1646600723266598
+157,oni,3.3,3.1216764,0.1783235549926756
+1137,oni,7.1,7.197917,0.09791698455810582
+1264,oni,7.5,7.736816,0.2368159294128418
+1261,oni,7.3,7.4277773,0.12777729034423846
+1404,oni,9.5,9.176101,0.3238992691040039
+1048,oni,9.4,9.316141,0.08385887145996129
+462,oni,7.3,6.9185348,0.3814652442932127
+580,oni,3.6,3.1813672,0.4186328411102296
+1434,oni,9.7,9.5976305,0.10236949920654226
+1263,oni,7.4,7.5086164,0.10861644744873011
+246,oni,1.1,1.8426442,0.7426442146301269
+1245,oni,8.4,8.364225,0.03577461242675817
+954,oni,9.7,9.612374,0.08762569427490163
+1391,oni,7.8,7.8338575,0.033857536315918146
+970,oni,7.2,7.0208073,0.17919273376464862
+861,oni,4.2,4.4093747,0.2093747138977049
+1405,oni,10.3,9.282177,1.0178230285644538
+652,oni,3.7,4.190357,0.49035720825195295
+348,oni,3.9,3.733858,0.1661418914794921
+1240,oni,4.2,4.3363075,0.13630752563476545
+824,oni,1.3,1.4652954,0.16529543399810787
+45,oni,7.9,8.064715,0.16471538543701136
+1236,oni,11.2,11.153378,0.046622467041014914
+332,oni,2.5,2.9927447,0.49274468421936035
+962,oni,2.6,3.1786516,0.5786515712738036
+1030,oni,4.9,5.055099,0.15509901046752894
+36,oni,1.6,1.8594168,0.25941684246063224
+1426,oni,6.2,5.8172493,0.38275070190429705
+896,oni,11.1,10.215234,0.8847661972045895
+611,oni,3.8,3.8965096,0.09650964736938494
+609,oni,5.8,5.289323,0.5106771469116209
+1307,oni,6.9,6.8420954,0.0579046249389652
+771,oni,3.3,3.3331902,0.03319020271301287
+154,oni,7.7,7.9209757,0.22097568511962873
+76,oni,3.5,3.6149366,0.11493659019470215
+200,oni,2.3,2.223773,0.0762269973754881
+94,oni,9.9,9.473127,0.42687263488769567
+615,oni,9.0,8.879463,0.12053680419921875
+880,oni,5.0,5.18964,0.18964004516601562
+822,oni,7.2,7.1134377,0.08656234741210955
+765,oni,11.9,11.395069,0.5049308776855472
+56,oni,6.6,6.737589,0.13758888244628942
+856,oni,8.8,8.237783,0.5622165679931648
+1419,oni,9.8,9.345617,0.4543827056884773
+1420,oni,3.7,3.7231805,0.02318053245544416
+456,oni,8.2,8.731038,0.5310380935668952
+917,oni,6.7,5.3931637,1.3068363189697267
+834,oni,9.7,9.58336,0.11664028167724538
+1145,oni,7.2,7.3018465,0.1018465042114256
+108,oni,6.8,6.9671035,0.1671034812927248
+736,oni,3.5,3.618028,0.11802792549133301
+345,oni,4.5,3.4398248,1.0601751804351807
+1149,oni,3.8,3.710602,0.08939795494079572
+472,oni,5.9,6.009906,0.10990581512451136
+31,oni,8.1,8.09505,0.004950141906737926
+714,oni,9.4,8.93766,0.4623397827148441
+466,oni,3.0,4.420792,1.4207921028137207
+779,oni,8.5,8.482459,0.017540931701660156
+734,oni,8.6,8.177258,0.42274246215820277
+1308,oni,8.4,8.65023,0.2502304077148434
+173,oni,6.0,5.917573,0.0824270248413086
+253,oni,5.0,5.038617,0.03861713409423828
+1042,oni,10.1,9.15089,0.9491096496582028
+612,oni,6.7,6.813999,0.11399917602539045
+1186,oni,1.9,2.0640535,0.16405353546142587
+1436,oni,6.0,5.912111,0.08788919448852539
+150,oni,6.3,6.2841473,0.015852737426757635
+1083,oni,7.5,5.1249447,2.3750553131103516
+777,oni,10.8,9.979303,0.8206966400146491
+98,oni,3.1,3.4718246,0.37182464599609366
+1118,oni,1.2,2.1188107,0.9188106536865235
+890,oni,11.0,11.081855,0.08185482025146484
+277,oni,3.2,2.558311,0.6416890144348146
+301,oni,2.8,3.031713,0.2317130088806154
+359,oni,8.3,8.350011,0.05001087188720632
+907,oni,7.9,8.020189,0.12018928527831996
+1373,oni,2.1,2.7734246,0.6734246253967284
+149,oni,4.8,5.013694,0.21369380950927752
+1181,oni,3.3,3.7384562,0.4384562492370607
+1126,oni,3.9,4.236051,0.3360510826110841
+1350,oni,11.9,11.538368,0.3616317749023441
+434,oni,4.1,4.1712093,0.07120933532714879
+420,oni,4.3,4.4294934,0.1294934272766115
+1337,oni,3.9,4.010494,0.11049423217773446
+1012,oni,9.7,9.215172,0.48482818603515554
+511,oni,2.3,2.8341331,0.5341331481933596
+1375,oni,1.1,1.3882558,0.2882558345794677
+1447,oni,10.4,10.219149,0.18085136413574254
+1187,oni,1.8,2.606092,0.8060919761657714
+711,oni,2.0,2.2765596,0.27655959129333496
+1146,oni,7.5,7.736816,0.2368159294128418
+350,oni,5.5,5.536312,0.036312103271484375
+976,oni,8.7,8.713133,0.013132858276367898
+559,oni,5.4,5.4344945,0.03449449539184535
+538,oni,8.6,8.172002,0.4279981613159176
+363,oni,8.1,7.6119237,0.48807630538940394
+356,oni,8.1,8.2789545,0.1789545059204105
+1438,oni,7.4,7.79311,0.39310989379882777
+111,oni,4.7,4.5066733,0.19332666397094744
+1389,oni,1.2,2.3112829,1.1112828731536866
+1243,oni,1.1,1.061276,0.03872404098510751
+312,oni,1.1,1.9396806,0.8396805763244628
+1015,oni,8.7,8.804349,0.10434894561767649
+785,oni,8.5,8.505435,0.005434989929199219
+1316,oni,2.8,2.868266,0.06826610565185565
+956,oni,2.1,2.52589,0.4258901119232177
+198,oni,2.4,2.7392402,0.3392401695251466
+1296,oni,7.9,7.9847984,0.08479843139648402
+351,oni,8.7,8.912447,0.21244697570800852
+505,oni,3.0,3.9480019,0.9480018615722656
+492,oni,1.9,2.398783,0.49878296852111825
+732,oni,2.4,2.3426375,0.0573624610900878
+247,oni,5.5,5.405316,0.09468412399291992
+1099,oni,11.0,9.8232155,1.1767845153808594
+695,oni,8.3,8.208186,0.09181385040283274
+1034,oni,5.9,5.8499846,0.050015354156494496
+1190,oni,5.5,5.4641275,0.035872459411621094
+746,oni,8.0,8.106806,0.10680580139160156

.old/test/validate_result_8.csv

@@ -0,0 +1,201 @@
+songno,diff,상수,predicted_measure,error
+1451,oni,5.1,5.1919827,0.09198274612426793
+317,oni,2.9,3.59739,0.6973899364471436
+285,oni,4.3,4.3387833,0.03878326416015643
+261,oni,7.4,7.6588984,0.2588983535766598
+359,oni,8.3,8.350011,0.05001087188720632
+1171,oni,6.1,6.1189337,0.0189336776733402
+861,oni,4.2,4.4093747,0.2093747138977049
+22,oni,5.5,5.295011,0.20498895645141602
+755,oni,6.9,7.2296677,0.3296676635742184
+406,oni,7.7,7.693886,0.006114196777343928
+127,oni,5.9,5.9468956,0.04689559936523402
+183,oni,7.0,7.051115,0.05111503601074219
+48,oni,5.6,5.4978957,0.10210428237915004
+233,oni,3.1,3.4442387,0.3442386627197265
+408,oni,3.6,3.389768,0.2102321147918702
+568,oni,7.5,7.328082,0.17191791534423828
+293,oni,3.9,4.5801744,0.6801744461059571
+1051,oni,8.7,8.919339,0.2193391799926765
+167,oni,7.1,7.137214,0.0372141838073734
+101,oni,1.4,1.6238686,0.22386858463287362
+454,oni,7.7,7.780956,0.0809557914733885
+852,oni,6.6,6.5446224,0.05537757873535121
+178,oni,6.8,6.8800697,0.0800697326660158
+1286,oni,2.2,2.5343087,0.33430867195129377
+1411,oni,9.1,8.809523,0.29047737121581996
+1166,oni,7.8,7.9244876,0.1244875907897951
+716,oni,6.8,6.8853464,0.08534641265869158
+1050,oni,9.2,9.424095,0.22409515380859446
+758,oni,10.8,9.948399,0.8516014099121101
+828,oni,4.1,4.206385,0.10638513565063512
+1303,oni,4.3,2.948968,1.3510320663452147
+53,oni,6.4,6.430334,0.030334091186523082
+930,oni,7.1,7.913493,0.8134931564331058
+525,oni,6.6,6.3501854,0.24981460571289027
+577,oni,9.8,9.342824,0.4571760177612312
+457,oni,6.5,6.5087705,0.008770465850830078
+538,oni,8.6,8.172002,0.4279981613159176
+1423,oni,4.7,4.6261406,0.0738594055175783
+1297,oni,9.4,9.196398,0.20360221862793004
+763,oni,10.8,9.695452,1.1045482635498054
+614,oni,6.4,5.970553,0.42944707870483434
+117,oni,8.0,8.2179575,0.2179574966430664
+87,oni,5.6,5.726159,0.1261590957641605
+346,oni,6.4,6.5222197,0.12221965789794886
+1422,oni,5.6,5.7453094,0.14530935287475621
+1196,oni,6.6,6.2691593,0.3308406829833981
+267,oni,3.6,3.689395,0.08939495086669913
+992,oni,10.1,9.259041,0.8409591674804684
+927,oni,8.9,8.5859165,0.3140834808349613
+1170,oni,6.9,7.0019712,0.10197124481201136
+429,oni,4.7,4.3966427,0.30335731506347674
+71,oni,6.0,5.845188,0.15481185913085938
+1410,oni,3.0,2.8184204,0.18157958984375
+438,oni,7.7,7.6389027,0.061097335815429865
+1114,oni,2.0,2.5233107,0.523310661315918
+1392,oni,5.4,7.2053556,1.8053556442260739
+304,oni,5.6,5.4814534,0.11854658126831019
+1159,oni,7.1,7.237465,0.1374649047851566
+1136,oni,2.7,3.2574682,0.5574682235717772
+732,oni,2.4,2.3426375,0.0573624610900878
+313,oni,2.6,2.9117153,0.31171526908874503
+754,oni,5.0,5.8170137,0.8170137405395508
+505,oni,3.0,3.9480019,0.9480018615722656
+191,oni,6.9,6.620808,0.279191875457764
+114,oni,7.0,6.7927685,0.2072315216064453
+701,oni,6.6,6.5500507,0.04994926452636683
+1271,oni,4.8,4.724819,0.07518081665039045
+111,oni,4.7,4.5066733,0.19332666397094744
+1406,oni,11.4,10.735728,0.6642717361450199
+1187,oni,1.8,2.606092,0.8060919761657714
+997,oni,4.0,4.2168503,0.21685028076171875
+1318,oni,8.9,8.928999,0.028998947143554332
+1216,oni,10.0,9.5192,0.48079967498779297
+1385,oni,3.2,2.8330605,0.36693949699401873
+420,oni,4.3,4.4294934,0.1294934272766115
+1108,oni,10.4,9.689839,0.7101606369018558
+31,oni,8.1,8.09505,0.004950141906737926
+1431,oni,7.8,7.896584,0.09658403396606463
+1281,oni,2.2,2.5469916,0.3469915866851805
+1317,oni,11.2,10.480424,0.7195760726928704
+1082,oni,5.0,4.8203955,0.17960453033447266
+555,oni,11.3,10.049599,1.2504013061523445
+171,oni,9.6,9.666423,0.06642284393310582
+637,oni,6.6,6.266298,0.33370218276977504
+323,oni,4.3,4.327031,0.02703113555908221
+105,oni,2.9,3.073173,0.17317304611206064
+1013,oni,8.3,8.191456,0.10854415893554759
+1311,oni,6.7,6.815501,0.11550121307373029
+126,oni,4.6,4.485203,0.11479721069335902
+36,oni,1.6,1.8594168,0.25941684246063224
+702,oni,10.6,10.179897,0.42010269165039027
+350,oni,5.5,5.536312,0.036312103271484375
+1219,oni,4.4,4.62963,0.229630088806152
+1020,oni,5.0,5.0843077,0.08430767059326172
+874,oni,7.0,7.0029125,0.0029125213623046875
+649,oni,7.6,7.212104,0.3878961563110348
+349,oni,6.8,6.479022,0.3209779739379881
+1169,oni,3.6,4.2148876,0.6148876190185546
+759,oni,9.0,8.656189,0.34381103515625
+302,oni,6.1,5.975836,0.12416419982910121
+1000,oni,8.0,8.003995,0.003994941711425781
+1175,oni,4.5,4.54766,0.047659873962402344
+513,oni,8.4,8.350839,0.0491613388061527
+1190,oni,5.5,5.4641275,0.035872459411621094
+1052,oni,8.1,7.9752064,0.12479362487792933
+1414,oni,5.2,5.329712,0.12971191406249982
+1070,oni,3.5,3.736236,0.2362360954284668
+898,oni,3.8,4.029142,0.22914190292358416
+247,oni,5.5,5.405316,0.09468412399291992
+1396,oni,4.6,3.8423722,0.7576278209686276
+343,oni,6.4,5.75634,0.6436599731445316
+405,oni,3.1,3.4590964,0.35909643173217765
+455,oni,5.7,5.487079,0.21292085647583026
+826,oni,10.5,9.781674,0.7183256149291992
+1115,oni,3.8,3.9233134,0.1233133792877199
+993,oni,10.0,9.799214,0.20078563690185547
+630,oni,9.3,8.637385,0.6626146316528327
+594,oni,9.7,9.348086,0.35191364288330007
+1160,oni,6.3,6.065338,0.23466186523437482
+397,oni,7.9,7.942276,0.042276000976562145
+1129,oni,9.8,9.580477,0.2195232391357429
+1366,oni,4.7,4.8216677,0.1216676712036131
+1054,oni,3.9,3.99446,0.09446010589599618
+855,oni,5.7,4.9179006,0.7820994377136232
+518,oni,11.4,11.650173,0.250173187255859
+1279,oni,8.4,8.668176,0.2681756973266598
+415,oni,1.8,2.84784,1.0478400707244873
+644,oni,4.4,4.3509674,0.049032592773437855
+616,oni,5.8,5.77202,0.027980136871337713
+1097,oni,5.1,5.135783,0.035783195495605824
+717,oni,5.6,5.3476925,0.2523075103759762
+947,oni,5.2,5.7635384,0.563538360595703
+73,oni,8.5,8.4226885,0.07731151580810547
+365,oni,10.0,8.945875,1.0541248321533203
+1257,oni,9.0,8.956529,0.043471336364746094
+549,oni,8.3,7.9438715,0.35612850189209055
+42,oni,1.1,2.9574435,1.8574434757232665
+44,oni,3.4,3.387909,0.012091064453124911
+275,oni,3.5,3.5875602,0.08756017684936523
+860,oni,10.6,10.310149,0.28985080718994105
+144,oni,2.5,2.725388,0.2253880500793457
+460,oni,2.7,2.8139534,0.11395339965820295
+137,oni,6.1,6.0426464,0.05735359191894496
+798,oni,9.1,8.862417,0.2375827789306637
+713,oni,8.3,8.379393,0.07939262390136648
+627,oni,11.5,10.365782,1.1342182159423828
+960,oni,10.1,9.670273,0.42972717285156214
+1259,oni,6.1,6.0426464,0.05735359191894496
+1350,oni,11.9,11.538368,0.3616317749023441
+1234,oni,6.6,6.50439,0.09561023712158168
+574,oni,8.6,8.241462,0.3585382461547848
+890,oni,11.0,11.081855,0.08185482025146484
+1403,oni,5.0,4.905402,0.09459781646728516
+1198,oni,7.0,7.4989223,0.49892234802246094
+989,oni,6.4,6.363348,0.03665199279785192
+612,oni,6.7,6.813999,0.11399917602539045
+197,oni,1.7,2.5009499,0.8009498596191407
+390,oni,4.8,4.1861615,0.613838481903076
+283,oni,8.5,8.440965,0.059035301208496094
+857,oni,7.7,7.948821,0.24882106781005842
+1255,oni,10.4,9.691372,0.708628082275391
+400,oni,7.6,7.566705,0.033294773101806285
+416,oni,6.0,5.845188,0.15481185913085938
+1455,oni,4.6,4.585906,0.014093971252441051
+834,oni,9.7,9.58336,0.11664028167724538
+1334,oni,4.0,4.209813,0.20981311798095703
+1415,oni,3.6,3.5990567,0.0009432792663575107
+1153,oni,1.8,2.1704838,0.37048382759094234
+226,oni,6.7,6.548712,0.15128822326660174
+1282,oni,3.8,4.334233,0.534232807159424
+150,oni,6.3,6.2841473,0.015852737426757635
+1210,oni,3.2,3.3614755,0.1614754676818846
+507,oni,5.8,5.805967,0.005966854095459162
+1071,oni,9.6,9.450291,0.14970932006835902
+569,oni,7.2,7.382146,0.18214588165283185
+81,oni,4.4,5.829715,1.4297147750854489
+91,oni,3.8,3.716854,0.08314590454101545
+2,oni,2.7,2.8962004,0.19620041847228986
+651,oni,7.1,6.437669,0.6623311996459957
+1237,oni,4.5,4.42432,0.07567977905273438
+335,oni,9.4,9.030155,0.36984481811523473
+593,oni,10.6,9.9727,0.627299880981445
+1347,oni,6.9,6.590419,0.3095811843872074
+503,oni,11.3,10.671308,0.6286924362182624
+40,oni,4.6,4.5613666,0.03863344192504847
+1165,oni,7.9,7.9121323,0.012132263183593395
+1432,oni,10.0,9.590779,0.40922069549560547
+585,oni,6.7,6.5668163,0.1331836700439455
+440,oni,6.8,6.4244413,0.3755586624145506
+1149,oni,3.8,3.710602,0.08939795494079572
+715,oni,8.4,8.413227,0.01322708129882777
+392,oni,9.0,8.880889,0.1191110610961914
+200,oni,2.3,2.223773,0.0762269973754881
+552,oni,6.5,6.4038367,0.09616327285766602
+1146,oni,7.5,7.736816,0.2368159294128418
+621,oni,8.2,8.37565,0.17565040588378977
+270,oni,6.9,6.9537883,0.05378828048706019
+1351,oni,9.3,9.762559,0.4625589370727532
+750,oni,8.8,9.02407,0.22406978607177663
+124,oni,5.5,5.4560986,0.04390144348144531

.old/test/validate_result_9.csv

@@ -0,0 +1,201 @@
+songno,diff,상수,predicted_measure,error
+697,oni,9.9,9.322553,0.5774473190307621
+593,oni,10.6,9.9727,0.627299880981445
+354,oni,4.8,4.4523187,0.3476813316345213
+1053,oni,4.6,4.43534,0.1646600723266598
+1302,oni,1.4,1.5472724,0.1472724437713624
+1258,oni,4.6,4.4628506,0.1371494293212887
+260,oni,8.8,7.5779324,1.2220676422119148
+1437,oni,7.5,7.7957687,0.29576873779296875
+484,oni,5.3,5.3123913,0.012391281127929865
+592,oni,7.1,7.0801096,0.01989040374755824
+186,oni,10.9,9.880772,1.0192283630371097
+103,oni,8.2,8.3685665,0.16856651306152415
+549,oni,8.3,7.9438715,0.35612850189209055
+572,oni,8.3,8.527885,0.22788543701171804
+388,oni,9.0,9.137562,0.13756179809570312
+757,oni,10.5,10.120777,0.3792228698730469
+36,oni,1.6,1.8594168,0.25941684246063224
+1324,oni,9.3,9.130554,0.1694458007812507
+1030,oni,4.9,5.055099,0.15509901046752894
+518,oni,11.4,11.650173,0.250173187255859
+742,oni,6.9,6.7455025,0.15449752807617223
+1435,oni,5.2,5.0266085,0.1733915328979494
+680,oni,7.9,7.9497805,0.049780464172362926
+43,oni,8.1,8.08124,0.018760299682616832
+198,oni,2.4,2.7392402,0.3392401695251466
+1333,oni,1.7,1.838375,0.13837497234344487
+309,oni,6.6,6.26643,0.33357009887695277
+120,oni,11.9,11.8358965,0.06410350799560582
+1292,oni,4.4,4.468697,0.0686970710754391
+675,oni,5.2,5.875123,0.6751230239868162
+1383,oni,9.3,9.224361,0.07563858032226634
+467,oni,2.9,2.862545,0.037454986572265536
+472,oni,5.9,6.009906,0.10990581512451136
+139,oni,4.8,4.6042676,0.1957324028015135
+63,oni,6.1,4.832204,1.2677961349487301
+525,oni,6.6,6.3501854,0.24981460571289027
+931,oni,7.7,7.59026,0.10973997116088885
+1117,oni,2.9,3.0611537,0.16115365028381357
+672,oni,2.4,2.623148,0.22314796447753915
+628,oni,10.7,9.843618,0.8563816070556634
+1006,oni,5.3,5.5312557,0.23125572204589862
+1204,oni,3.1,3.1182175,0.01821746826171866
+95,oni,7.3,7.362153,0.062153053283691584
+946,oni,3.4,3.4010017,0.0010016918182373935
+774,oni,9.0,9.228863,0.22886276245117188
+83,oni,6.0,6.248336,0.2483358383178711
+237,oni,5.3,6.2021265,0.9021265029907228
+1129,oni,9.8,9.580477,0.2195232391357429
+301,oni,2.8,3.031713,0.2317130088806154
+1158,oni,8.6,8.740654,0.1406539916992191
+471,oni,7.2,7.574646,0.3746459960937498
+1350,oni,11.9,11.538368,0.3616317749023441
+1364,oni,7.7,7.450409,0.24959106445312518
+898,oni,3.8,4.029142,0.22914190292358416
+1354,oni,7.0,7.4809437,0.4809436798095703
+403,oni,5.3,5.125764,0.17423610687255842
+698,oni,9.9,9.641409,0.2585910797119144
+151,oni,4.8,4.623654,0.17634611129760724
+248,oni,6.2,6.27658,0.07657985687255842
+993,oni,10.0,9.799214,0.20078563690185547
+1308,oni,8.4,8.65023,0.2502304077148434
+1331,oni,2.2,2.7129545,0.512954521179199
+1033,oni,4.3,4.3674026,0.06740255355834979
+614,oni,6.4,5.970553,0.42944707870483434
+1209,oni,4.1,4.285281,0.18528118133544957
+86,oni,6.8,7.345725,0.5457250595092775
+1411,oni,9.1,8.809523,0.29047737121581996
+1197,oni,4.5,4.3139343,0.186065673828125
+786,oni,9.3,8.820966,0.47903423309326243
+347,oni,5.7,5.845188,0.14518814086914045
+1276,oni,9.2,9.156151,0.04384918212890554
+895,oni,1.0,1.2011725,0.20117247104644775
+1400,oni,1.6,2.1304216,0.5304216384887694
+187,oni,7.4,7.523672,0.12367210388183558
+1359,oni,2.2,2.4671395,0.26713948249816877
+149,oni,4.8,5.013694,0.21369380950927752
+1245,oni,8.4,8.364225,0.03577461242675817
+1026,oni,4.1,3.8297968,0.2702032089233395
+312,oni,1.1,1.9396806,0.8396805763244628
+92,oni,2.4,3.0877912,0.6877912044525147
+1361,oni,1.5,1.7912393,0.29123926162719727
+1320,oni,8.0,7.8748026,0.1251974105834961
+977,oni,8.8,8.352481,0.4475191116333015
+247,oni,5.5,5.405316,0.09468412399291992
+179,oni,10.2,9.583749,0.6162511825561516
+1424,oni,6.4,6.2954216,0.10457839965820348
+167,oni,7.1,7.137214,0.0372141838073734
+138,oni,7.7,7.8239584,0.12395839691162092
+804,oni,5.7,5.78445,0.084450054168701
+1335,oni,4.0,3.8683329,0.1316671371459961
+220,oni,9.3,9.154587,0.1454132080078132
+1277,oni,5.0,5.122813,0.12281322479248047
+888,oni,6.1,5.907568,0.1924320220947262
+1125,oni,3.6,3.4895444,0.11045560836792001
+242,oni,2.9,2.8971946,0.0028053760528563565
+1348,oni,6.7,5.7091274,0.9908725738525392
+267,oni,3.6,3.689395,0.08939495086669913
+750,oni,8.8,9.02407,0.22406978607177663
+1388,oni,6.1,5.9238286,0.17617139816284144
+435,oni,5.7,5.4369884,0.26301164627075213
+1170,oni,6.9,7.0019712,0.10197124481201136
+1306,oni,4.9,4.731516,0.1684841156005863
+253,oni,5.0,5.038617,0.03861713409423828
+108,oni,6.8,6.9671035,0.1671034812927248
+817,oni,8.0,8.014091,0.014090538024902344
+241,oni,2.6,2.7528176,0.15281763076782218
+1180,oni,6.5,6.422063,0.07793712615966797
+574,oni,8.6,8.241462,0.3585382461547848
+1275,oni,4.1,4.2565866,0.15658655166626012
+1084,oni,2.6,3.0994637,0.49946370124816886
+1362,oni,6.1,5.8406997,0.25930032730102504
+1106,oni,9.1,9.420722,0.3207220077514652
+1027,oni,3.9,4.020381,0.12038097381591806
+747,oni,7.3,7.7551966,0.45519657135009783
+891,oni,6.8,6.728759,0.07124118804931623
+651,oni,7.1,6.437669,0.6623311996459957
+930,oni,7.1,7.913493,0.8134931564331058
+1149,oni,3.8,3.710602,0.08939795494079572
+498,oni,6.7,5.881829,0.8181712150573732
+868,oni,9.5,8.869226,0.6307735443115234
+1022,oni,2.7,2.4909775,0.2090224742889406
+537,oni,6.9,6.9082613,0.008261299133300426
+1037,oni,9.2,8.873722,0.32627792358398366
+703,oni,5.1,5.1526203,0.05262031555175817
+270,oni,6.9,6.9537883,0.05378828048706019
+992,oni,10.1,9.259041,0.8409591674804684
+1337,oni,3.9,4.010494,0.11049423217773446
+239,oni,7.6,9.329233,1.7292331695556644
+427,oni,9.1,8.9220085,0.17799148559570277
+654,oni,4.5,4.328125,0.171875
+1349,oni,6.3,5.845188,0.4548118591308592
+383,oni,6.1,6.0199423,0.08005771636962855
+1316,oni,2.8,2.868266,0.06826610565185565
+514,oni,7.7,7.8003054,0.1003053665161131
+860,oni,10.6,10.310149,0.28985080718994105
+541,oni,5.6,5.419343,0.18065700531005824
+530,oni,4.7,4.619874,0.08012599945068377
+1322,oni,7.9,7.850716,0.0492838859558109
+53,oni,6.4,6.430334,0.030334091186523082
+80,oni,8.4,8.461607,0.06160697937011683
+974,oni,1.5,2.088167,0.5881669521331787
+293,oni,3.9,4.5801744,0.6801744461059571
+1323,oni,6.8,6.8110046,0.011004638671875178
+335,oni,9.4,9.030155,0.36984481811523473
+816,oni,8.2,8.676427,0.47642688751220774
+1261,oni,7.3,7.4277773,0.12777729034423846
+924,oni,2.8,2.7967556,0.003244447708129705
+580,oni,3.6,3.1813672,0.4186328411102296
+1387,oni,4.0,3.7799487,0.22005128860473633
+358,oni,7.5,7.3555374,0.14446258544921875
+1284,oni,2.3,2.797853,0.4978529930114748
+934,oni,9.5,8.944782,0.5552177429199219
+1160,oni,6.3,6.065338,0.23466186523437482
+881,oni,7.7,7.814657,0.11465721130371076
+1293,oni,3.8,4.05702,0.25702018737792987
+507,oni,5.8,5.805967,0.005966854095459162
+1373,oni,2.1,2.7734246,0.6734246253967284
+430,oni,1.5,1.7298045,0.22980451583862305
+208,oni,4.5,4.484102,0.01589822769165039
+523,oni,5.1,4.688936,0.41106376647949183
+411,oni,4.8,4.3224406,0.4775593757629393
+1192,oni,5.7,5.762822,0.06282215118408185
+1455,oni,4.6,4.585906,0.014093971252441051
+511,oni,2.3,2.8341331,0.5341331481933596
+363,oni,8.1,7.6119237,0.48807630538940394
+1415,oni,3.6,3.5990567,0.0009432792663575107
+361,oni,8.0,8.176176,0.1761760711669922
+1079,oni,5.6,5.3822875,0.21771249771118129
+194,oni,9.4,9.450678,0.05067787170410121
+366,oni,1.5,2.3543668,0.8543667793273926
+534,oni,4.0,3.9437454,0.05625462532043457
+1111,oni,4.0,4.145422,0.14542198181152344
+131,oni,8.2,8.129029,0.07097072601318288
+1049,oni,9.5,9.412386,0.08761405944824219
+976,oni,8.7,8.713133,0.013132858276367898
+539,oni,6.0,5.876666,0.12333393096923828
+1231,oni,5.7,5.6410522,0.05894775390625018
+758,oni,10.8,9.948399,0.8516014099121101
+1118,oni,1.2,2.1188107,0.9188106536865235
+1220,oni,3.6,3.5168161,0.08318386077880868
+1368,oni,7.5,7.64944,0.14943981170654297
+1273,oni,6.9,5.924207,0.9757927894592289
+40,oni,4.6,4.5613666,0.03863344192504847
+1072,oni,9.8,9.565875,0.234124946594239
+1050,oni,9.2,9.424095,0.22409515380859446
+432,oni,6.6,6.7569847,0.15698471069335973
+1301,oni,2.0,2.1946142,0.19461417198181152
+315,oni,3.2,3.581308,0.3813078880310057
+926,oni,3.7,3.3538485,0.3461515426635744
+773,oni,4.0,3.9387102,0.06128978729248047
+782,oni,2.7,2.7994647,0.099464702606201
+1144,oni,2.7,2.792343,0.09234290122985822
+880,oni,5.0,5.18964,0.18964004516601562
+1043,oni,9.8,9.788532,0.011467742919922586
+1096,oni,5.1,3.3114526,1.7885473728179928
+1069,oni,8.0,7.9372826,0.06271743774414062
+790,oni,2.5,2.8459394,0.34593939781188965
+646,oni,2.3,2.4966152,0.1966151714324953
+456,oni,8.2,8.731038,0.5310380935668952
+699,oni,9.7,9.337942,0.36205787658691335

.old/train.py

@@ -0,0 +1,152 @@
+import pandas as pd
+import json, os, sys
+import numpy as np
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from sklearn.preprocessing import StandardScaler
+
+MAX_MARGIN_LIMIT = 10
+
+class DifficultyNet(nn.Module):
+    def __init__(self, input_dim):
+        super(DifficultyNet, self).__init__()
+        self.net = nn.Sequential(
+            nn.Linear(input_dim, 128),
+            nn.BatchNorm1d(128),
+            nn.ReLU(),
+            nn.Dropout(0.4),
+            nn.Linear(128, 64),
+            nn.ReLU(),
+            nn.Dropout(0.3),
+            nn.Linear(64, 32),
+            nn.ReLU(),
+            nn.Linear(32, 1),
+            nn.Sigmoid()
+        )
+
+    def forward(self, x):
+        return self.net(x)
+
+def train(script_dir, working_dir, data_dir, train_count, val_count, margin, val_iterations):
+    device = torch.device("mps" if torch.backends.mps.is_available() else "cpu")
+    print(f"🚀 Using Device: {device}")
+
+    # 데이터 로드 및 병합
+    factors_path = os.path.join(working_dir, 'factors.json')
+    if not os.path.exists(factors_path):
+        os.system(f"bun run {os.path.join(script_dir, 'factorize.ts')} {data_dir} {working_dir}")
+
+    df_feat = pd.read_json(factors_path)
+    df_meas = pd.read_csv(os.path.join(data_dir, 'measure.csv'))
+    df_meas['diff'] = df_meas['diff'].replace('ura', 'edit')
+    df_feat['songno'], df_meas['songno'] = df_feat['songno'].astype(int), df_meas['songno'].astype(int)
+    df = pd.merge(df_feat, df_meas, on=['songno', 'diff'])
+
+    exclude_cols = ['songno', 'diff', 'title', 'course', '상수', 'predicted_measure', 'error']
+    X_cols = [c for c in df_feat.columns if c not in exclude_cols]
+    
+    scaler = StandardScaler()
+    df[X_cols] = scaler.fit_transform(df[X_cols])
+    with open(os.path.join(working_dir, 'scaler.json'), 'w') as f:
+        json.dump({'mean': scaler.mean_.tolist(), 'std': scaler.scale_.tolist(), 'cols': X_cols}, f)
+
+    model_path = os.path.join(working_dir, 'model.pth')
+    margin = float(margin)
+    val_iterations = int(val_iterations)
+    
+    model = DifficultyNet(len(X_cols)).to(device)
+    attempt = 1
+
+    while True:
+        print(f"\n[Attempt {attempt}] " + "-"*40)
+        if os.path.exists(model_path):
+            try: model.load_state_dict(torch.load(model_path, map_location=device))
+            except: pass
+
+        train_df = df.sample(n=min(int(train_count), len(df)))
+        X_train = torch.FloatTensor(train_df[X_cols].values).to(device)
+        y_train = torch.FloatTensor(((train_df['상수'].values - 1) / 11)).unsqueeze(1).to(device)
+
+        base_lr = 0.001
+        optimizer = optim.Adam(model.parameters(), lr=base_lr, weight_decay=1e-5)
+        criterion = nn.L1Loss()
+
+        model.train()
+        epoch = 0
+        while True:
+            optimizer.zero_grad()
+            preds_train = model(X_train)
+            loss = criterion(preds_train, y_train) # MSELoss 권장
+            loss.backward()
+            optimizer.step()
+            
+            # 현재 학습 데이터에 대한 오차 분석
+            with torch.no_grad():
+                diff = torch.abs((preds_train * 12 + 0.5) - (y_train * 12 + 0.5))
+                train_mae = torch.mean(diff).item()
+                train_max = torch.max(diff).item()
+            
+            if epoch % 1000 == 0:
+                print(f"   - Ep {epoch:5d} | MAE: {train_mae:.4f} | MAX: {train_max:.4f}")
+            
+            # ✅ 조건 강화: MAE뿐만 아니라 MAX도 어느 정도 잡혔을 때만 검증으로 이동
+            if train_mae < margin and train_max < (margin * MAX_MARGIN_LIMIT): 
+                print(f"   ✅ Train Goal Reached (MAE: {train_mae} < {margin:.4f}, MAX: {train_max:.4f} < {(margin * MAX_MARGIN_LIMIT):.4f}). Moving to Val.")
+                break
+            
+            epoch += 1
+            if epoch > 50000:
+                print("   ⚠️ Timed out. Resampling...")
+                break
+        
+        if train_mae >= margin:
+            torch.save(model.state_dict(), model_path)
+            attempt += 1
+            continue
+
+        # 3. 검증 단계
+        print(f"3. Validating {val_iterations} iterations...")
+        model.eval()
+        all_passed = True
+        with torch.no_grad():
+            for i in range(1, val_iterations + 1):
+                val_df = df.sample(n=min(int(val_count), len(df)))
+                X_val = torch.FloatTensor(val_df[X_cols].values).to(device)
+                y_val_raw = val_df['상수'].values
+                preds = model(X_val) * 12 + 0.5
+                y_val_tensor = torch.FloatTensor(y_val_raw).unsqueeze(1).to(device)
+                diff_tensor = torch.abs(preds - y_val_tensor)
+                mae = torch.mean(diff_tensor).item()
+                max_error = torch.max(diff_tensor).item()
+                
+
+                # csv 저장
+                # preds를 CPU 넘파이로 변환하고 1차원으로 펴주는 과정이 포함되어야 합니다.
+                preds_np = preds.detach().cpu().numpy().flatten() if torch.is_tensor(preds) else preds
+                
+                res_df = val_df[['songno', 'diff', '상수']].copy()
+                res_df['predicted_measure'] = preds_np
+                res_df['error'] = np.abs(preds_np - y_val_raw)
+                
+                output_file = os.path.join(working_dir, f'validate_result_{i}.csv')
+                res_df.to_csv(output_file, index=False)
+
+                if mae <= margin and max_error <= (margin * MAX_MARGIN_LIMIT):
+                    print(f"   [Iter {i}] ✅ PASS (MAE: {mae:.4f}, MAX: {max_error:.4f})")
+                else:
+                    print(f"   [Iter {i}] ❌ FAIL (MAE: {mae:.4f} > {margin} || MAX: {max_error:.4f} > {margin * MAX_MARGIN_LIMIT})")
+                    all_passed = False
+                    break 
+
+        if all_passed:
+            print(f"\n✨ Final Success!")
+            torch.save(model.state_dict(), model_path)
+            return
+        else:
+            torch.save(model.state_dict(), model_path)
+            attempt += 1
+
+if __name__ == "__main__":
+    if len(sys.argv) < 8: sys.exit(1)
+    train(*sys.argv[1:8])

.old/train.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+
+# Arguments: script_dir working_dir data_dir train_count validate_count margin validate_iterations
+SCRIPT_DIR=${1:-"script"}
+WORKING_DIR=${2:-"."}
+DATA_DIR=${3:-"datas"}
+TRAIN_COUNT=${4:-100}
+VAL_COUNT=${5:-20}
+MARGIN=${6:-0.5}
+VAL_ITERATIONS=${7:-5}
+
+# 작업 디렉토리가 없으면 생성
+if [ ! -d "$WORKING_DIR" ]; then
+    echo "Creating working directory: $WORKING_DIR"
+    mkdir -p "$WORKING_DIR"
+fi
+
+# factors.json이 없으면 생성
+if [ ! -f "$WORKING_DIR/factors.json" ]; then
+    echo "Factors file not found. Running factorization..."
+    bun run "$SCRIPT_DIR/extract.ts" "$WORKING_DIR" "$DATA_DIR"
+fi
+
+echo "Starting training..."
+python3 "$SCRIPT_DIR/train.py" "$SCRIPT_DIR" "$WORKING_DIR" "$DATA_DIR" "$TRAIN_COUNT" "$VAL_COUNT" "$MARGIN" "$VAL_ITERATIONS"

.old/train.ts

@@ -0,0 +1,118 @@
+import { parseTja } from './parse';
+import { factorize, Factors } from './factorize';
+import { join } from 'path';
+import { readFileSync, writeFileSync, readdirSync } from 'fs';
+import { parse } from 'csv-parse/sync';
+
+async function train() {
+    const args = process.argv.slice(2);
+    const workingDir = args[0];
+    const scriptDir = args[1];
+    const dataDir = args[2];
+    const trainCount = parseInt(args[3]) || 100;
+    const validateCount = parseInt(args[4]) || 20;
+    const margin = parseFloat(args[5]) || 0.1;
+
+    const factorJsonPath = join(scriptDir, 'factor.json');
+ // 항상 script/ 위치 참조
+    let weights: Factors = JSON.parse(readFileSync(factorJsonPath, 'utf-8'));
+    
+    // 결과물(로그 등)을 저장할 경로 (필요 시 활용)
+    const logPath = join(workingDir, 'training_log.txt'); 
+
+    const measureCsv = readFileSync(join(dataDir, 'measure.csv'), 'utf-8');
+    const records = parse(measureCsv, { columns: true, skip_empty_lines: true });
+
+    const tjaFiles = readdirSync(join(dataDir, 'tja')).filter(f => f.endsWith('.tja'));
+    const shuffled = tjaFiles.sort(() => 0.5 - Math.random());
+    const trainFiles = shuffled.slice(0, trainCount);
+    const validateFiles = shuffled.slice(trainCount, trainCount + validateCount);
+
+    console.log(`Training with ${trainFiles.length} files...`);
+
+    // 학습 로직 (Sigmoid + [1, 12] scaling)
+    const sigmoid = (x: number) => 1 / (1 + Math.exp(-x));
+    
+    let error = Infinity;
+    let iterations = 0;
+    while (error > margin && iterations < 10) { // Spec에 따라 10번 반복
+        let totalError = 0;
+        let count = 0;
+
+        for (const file of trainFiles) {
+            const songno = file.replace(/\D/g, '');
+            const tjaContent = readFileSync(join(dataDir, 'tja', file), 'utf-8');
+            const parsed = parseTja(tjaContent);
+            if (!parsed) continue;
+
+            for (const diff of ['oni', 'edit'] as const) {
+                const course = parsed[diff];
+                if (!course) continue;
+
+                const target = records.find((r: any) => r.songno === songno && (r.diff === (diff === 'oni' ? 'oni' : 'ura')));
+                if (!target) continue;
+
+                const factors = factorize(course);
+                // 가중치 적용 전 정규화 (임의 값으로 가정)
+                const normalizedFactors = {
+                    physical_density: Math.min(factors.physical_density / 20, 1),
+                    stamina_requirement: Math.min(factors.stamina_requirement, 1),
+                    pattern_complexity: Math.min(factors.pattern_complexity, 1),
+                    rhythmic_complexity: Math.min(factors.rhythmic_complexity, 1),
+                    reading_gimmick: Math.min(factors.reading_gimmick, 1)
+                };
+
+                const rawPrediction = Object.keys(normalizedFactors).reduce((sum, key) => 
+                    sum + (normalizedFactors[key as keyof Factors] * weights[key as keyof Factors]), 0);
+                
+                const prediction = (sigmoid(rawPrediction) * 11) + 1; // [1, 12]
+                
+                const targetValue = parseFloat(target.상수);
+                const diff_val = targetValue - prediction;
+                totalError += Math.abs(diff_val);
+                count++;
+
+                // 가중치 업데이트 (간단한 경사 하강법)
+                for (const key in weights) {
+                    const k = key as keyof Factors;
+                    weights[k] += diff_val * normalizedFactors[k] * 0.01;
+                }
+            }
+        }
+        error = totalError / (count || 1);
+        console.log(`Iteration ${iterations}: Mean Error = ${error.toFixed(4)}`);
+        iterations++;
+    }
+
+    writeFileSync(factorJsonPath, JSON.stringify(weights, null, 2));
+    writeFileSync(join(workingDir, 'training_result.json'), JSON.stringify({ finalError: error, weights }, null, 2));
+    console.log(`Training complete. Weights saved to ${factorJsonPath}, result saved to ${workingDir}`);
+
+    // 검증 로직
+    console.log('\nValidation Results:');
+    for (const file of validateFiles) {
+        const songno = file.replace(/\D/g, '');
+        const tjaContent = readFileSync(join(dataDir, 'tja', file), 'utf-8');
+        const parsed = parseTja(tjaContent);
+        if (!parsed) continue;
+
+        for (const diff of ['oni', 'edit'] as const) {
+            const course = parsed[diff];
+            if (!course) continue;
+
+            const target = records.find((r: any) => r.songno === songno && (r.diff === (diff === 'oni' ? 'oni' : 'ura')));
+            if (!target) {
+                console.log(`[!] No match for ${songno} diff=${diff === 'oni' ? 'oni' : 'ura'}`);
+                continue;
+            }
+
+            const factors = factorize(course);
+            const prediction = Object.keys(factors).reduce((sum, key) => 
+                sum + (factors[key as keyof Factors] * weights[key as keyof Factors]), 0);
+            
+            console.log(`[${songno}] Target: ${target.상수}, Predicted: ${prediction.toFixed(2)}, Diff: ${Math.abs(parseFloat(target.상수) - prediction).toFixed(2)}`);
+        }
+    }
+}
+
+train();

bun.lock

@@ -8,6 +8,7 @@
         "tja-parser": "^0.2.9",
       },
       "devDependencies": {
+        "@types/bun": "^1.3.13",
         "@types/iconv-lite": "^0.0.1",
         "@types/node": "^25.6.0",
         "typescript": "^6.0.3",
@@ -17,10 +18,14 @@
   "packages": {
     "@babel/runtime": ["@babel/runtime@7.29.2", "", {}, "sha512-JiDShH45zKHWyGe4ZNVRrCjBz8Nh9TMmZG1kh4QTK8hCBTWBi8Da+i7s1fJw7/lYpM4ccepSNfqzZ/QvABBi5g=="],
 
+    "@types/bun": ["@types/bun@1.3.13", "", { "dependencies": { "bun-types": "1.3.13" } }, "sha512-9fqXWk5YIHGGnUau9TEi+qdlTYDAnOj+xLCmSTwXfAIqXr2x4tytJb43E9uCvt09zJURKXwAtkoH4nLQfzeTXw=="],
+
     "@types/iconv-lite": ["@types/iconv-lite@0.0.1", "", { "dependencies": { "@types/node": "*" } }, "sha512-SsRBQxGw7/2/NxYJfBdiUx5a7Ms/voaUhOO9u2y9FTeTNBO1PXohzE4i3JfD8q2Te42HLTn5pyZtDf8j1bPKgQ=="],
 
     "@types/node": ["@types/node@25.6.0", "", { "dependencies": { "undici-types": "~7.19.0" } }, "sha512-+qIYRKdNYJwY3vRCZMdJbPLJAtGjQBudzZzdzwQYkEPQd+PJGixUL5QfvCLDaULoLv+RhT3LDkwEfKaAkgSmNQ=="],
 
+    "bun-types": ["bun-types@1.3.13", "", { "dependencies": { "@types/node": "*" } }, "sha512-QXKeHLlOLqQX9LgYaHJfzdBaV21T63HhFJnvuRCcjZiaUDpbs5ED1MgxbMra71CsryN/1dAoXuJJJwIv/2drVA=="],
+
     "complex.js": ["complex.js@2.4.3", "", {}, "sha512-UrQVSUur14tNX6tiP4y8T4w4FeJAX3bi2cIv0pu/DTLFNxoq7z2Yh83Vfzztj6Px3X/lubqQ9IrPp7Bpn6p4MQ=="],
 
     "csv-parse": ["csv-parse@6.2.1", "", {}, "sha512-LRLMV+UCyfMokp8Wb411duBf1gaBKJfOfBWU9eHMJ+b+cJYZsNu3AFmjJf3+yPGd59Exz1TsMjaSFyxnYB9+IQ=="],

docs/1. tja factorize.md

@@ -0,0 +1,12 @@
+# TJA factorize
+TJA를 DNN에 사용하기 위해 factorize한다.
+
+각 노트를 다음과 같이 변환한다.
+- [Note type, BPM, Scroll, Timing]
+- Note type은 0과 1만 사용한다.
+    - Don -> 0
+    - Ka -> 1
+- BPM은 100을 나누어 사용한다.
+- Scroll은 5를 나누어 사용한다.
+- Timing은 이전 노트와의 시간 차이를 사용한다. (단위 s)
+

docs/2. features.md

@@ -0,0 +1,40 @@
+# Feature
+DNN 분석을 위해 채보의 특성을 추출한다.
+
+## Average density
+평균 밀도를 나타낸다.
+
+$\frac{노트 수}{마지막 노트 타이밍 - 첫 노트 타이밍}$
+
+## Peak density
+최고 밀도를 나타낸다.
+
+어떤 노트를 기준으로 앞으로 1초내에 있는 노트들의 개수 중 최대값으로 구한다.
+
+## Average BPM
+평균 BPM을 나타낸다.
+
+$\frac{\Sigma BPM}{노트 수}$
+
+## Average BPM 2
+평균 BPM을 나타내나, 다른 방식으로 구한다.
+극소수의 노트만 BPM이 다를 경우 평균 BPM에 미치는 영향이 클 수 있기 때문에, BPM의 제곱을 총합하여 평균을 구한다.
+
+$\sqrt{\frac{\Sigma BPM^2}{노트수}}$
+
+## BPM Change
+BPM 변화 횟수를 나타낸다. BPM 흔들림을 제외하기 위해, 이전 노트와의 BPM차이가 1.5 이상일 떄 1 증가한다.
+
+## Scroll Change
+스크롤 변화 횟수를 나타낸다. 노트의 $BPM \times Scroll$의 값이 이전 노트와 1.5 이상 차이날 때 1 증가한다.
+
+## Rhythm Complexity
+i번째 노트와 i-1번쨰 노트의 간격이 i-1번쨰 노트와 i-2번째 노트의 간격의 비율이 2의 거듭제곱이 아닐 때 1 증가한다.
+
+## Color Complexity
+i번째 노트와 i-2번쨰 노트가 다른 종류일 때 증가하며, 간격의 제곱의 역수에 비례한다.
+
+$\frac{\mathrm{color\ changed\ ?\ 1\ :\ 0}}{\Delta t^2}$
+
+## Note Count
+노트의 개수

docs/3-1.xgboost.md

@@ -0,0 +1,56 @@
+# XGBoost 기반 난이도 상수 예측 모델
+
+이 문서는 프로젝트의 핵심 학습 엔진인 XGBoost 모델의 구조, 하이퍼파라미터 및 파이프라인 과정을 설명합니다.
+
+## 1. 모델 개요
+본 프로젝트는 태고의 달인 채보의 미세한 난이도 차이(상수)를 예측하기 위해 **XGBoost (Extreme Gradient Boosting)** 회귀 모델을 사용합니다. 8가지 주요 피처를 입력으로 받아 0.1 단위의 정밀한 난이도 상수를 출력하는 것을 목표로 합니다.
+
+## 2. 하이퍼파라미터 설정
+`train/train_xgboost.py`에 정의된 주요 설정값은 다음과 같습니다.
+
+| 파라미터 | 설정값 | 설명 |
+| :--- | :--- | :--- |
+| `N_ESTIMATORS` | 500 | 결정 트리의 개수 |
+| `MAX_DEPTH` | 6 | 각 트리의 최대 깊이 |
+| `LEARNING_RATE` | 0.05 | 학습률 (경사 하강 단계의 크기) |
+| `SUBSAMPLE` | 0.8 | 각 트리 학습 시 사용할 데이터 샘플 비율 (과적합 방지) |
+| `COLSAMPLE_BYTREE` | 0.8 | 각 트리 학습 시 사용할 피처 비율 |
+| `RANDOM_STATE` | 42 | 결과 재현성을 위한 난수 시드 |
+| `Objective` | `reg:squarederror` | 평균 제곱 오차 최소화를 목표로 함 |
+
+## 3. 학습 프로세스 (Training Pipeline)
+
+### A. 데이터 로드 및 매칭
+- `features.json`: 전처리 단계에서 추출된 채보별 피처 데이터.
+- `measure.csv`: 각 곡의 `songno`, `diff`에 대응하는 실제 난이도 상수(Label).
+- 두 데이터를 `(songno, difficulty)` 키를 기준으로 매칭하여 학습 데이터셋을 구성합니다.
+
+### B. 전처리 및 정규화
+- **Feature Selection**: `songno`, `difficulty`를 제외한 모든 수치형 데이터를 피처로 사용합니다.
+- **Scaling**: `sklearn.preprocessing.StandardScaler`를 사용하여 피처의 평균을 0, 분산을 1로 정규화합니다. 이는 `color_complexity`와 같이 수치 범위가 큰 피처가 모델에 과도한 영향을 주는 것을 방지합니다.
+
+### C. 모델 학습 및 지속 학습 (Warm Start)
+- `CONTINUE_TRAINING` 옵션이 활성화된 경우, 기존에 저장된 `model.pkl`과 `scaler.pkl`을 로드하여 이전 상태를 유지하며 추가 학습을 진행합니다.
+
+## 4. 평가 지표
+- **MAE (Mean Absolute Error)**: 실제 상수와 예측값 간의 절대 오차 평균.
+- **Accuracy (±0.1)**: 예측값이 실제 값과 0.1 이내로 일치하는 비율을 측정하여 실질적인 예측 성능을 평가합니다.
+- **Feature Importance**: 학습 완료 후 각 피처가 예측에 기여한 중요도를 출력하여 모델의 판단 근거를 확인합니다.
+
+## 5. 예측 프로세스 (Inference)
+`predict/predict_xgboost.py`를 통해 다음 과정을 거쳐 결과를 도출합니다.
+
+1. 학습 시 저장된 `model.pkl`, `scaler.pkl`, `features.txt` 로드.
+2. 입력받은 `songno`에 해당하는 피처를 `features.json`에서 탐색.
+3. 학습 시와 동일한 순서로 피처 벡터를 구성하고 Scaler를 적용.
+4. 모델을 통해 예측을 수행하고 결과를 JSON 배열 형태로 출력.
+
+```json
+[
+  {
+    "songno": "1000",
+    "diff": "oni",
+    "predicted": 11.2345
+  }
+]
+```

output/lightgbm/features_lgbm.txt

@@ -0,0 +1,8 @@
+bpm_avg
+bpm_change
+color_complexity
+density_avg
+density_peak
+note_count
+rhythm_complexity
+scroll_change

output/lightgbm2/features_lgbm.txt

@@ -0,0 +1,8 @@
+bpm_avg
+bpm_change
+color_complexity
+density_avg
+density_peak
+note_count
+rhythm_complexity
+scroll_change

output/lightgbm3/features_lgbm.txt

@@ -0,0 +1,8 @@
+bpm_avg
+bpm_change
+color_complexity
+density_avg
+density_peak
+note_count
+rhythm_complexity
+scroll_change

output/xgboost/features.txt

@@ -0,0 +1,8 @@
+bpm_avg
+bpm_change
+color_complexity
+density_avg
+density_peak
+note_count
+rhythm_complexity
+scroll_change

output/xgboost_copy/features.json

@@ -0,0 +1,74 @@
+[
+  {
+    "songno": "XODUS",
+    "difficulty": "oni",
+    "note_count": 1007,
+    "density_avg": 8.14959935897436,
+    "density_peak": 17,
+    "bpm_avg": 202,
+    "bpm_change": 0,
+    "scroll_change": 14,
+    "rhythm_complexity": 868,
+    "color_complexity": 0.012565274045963932
+  },
+  {
+    "songno": "Destructive Little Sister",
+    "difficulty": "oni",
+    "note_count": 1119,
+    "density_avg": 7.917452830188679,
+    "density_peak": 21,
+    "bpm_avg": 247.5,
+    "bpm_change": 0,
+    "scroll_change": 17,
+    "rhythm_complexity": 1079,
+    "color_complexity": 0.016291673920775917
+  },
+  {
+    "songno": "7 Wonders",
+    "difficulty": "oni",
+    "note_count": 794,
+    "density_avg": 5.571929824561403,
+    "density_peak": 12,
+    "bpm_avg": 168,
+    "bpm_change": 0,
+    "scroll_change": 0,
+    "rhythm_complexity": 726,
+    "color_complexity": 0.006656416024691356
+  },
+  {
+    "songno": "7 Wonders",
+    "difficulty": "ura",
+    "note_count": 1207,
+    "density_avg": 8.47017543859649,
+    "density_peak": 15,
+    "bpm_avg": 168,
+    "bpm_change": 0,
+    "scroll_change": 0,
+    "rhythm_complexity": 965,
+    "color_complexity": 0.01463965596985236
+  },
+  {
+    "songno": "Destruction 3 2 1",
+    "difficulty": "oni",
+    "note_count": 1160,
+    "density_avg": 7.46659375,
+    "density_peak": 15,
+    "bpm_avg": 321.3209999999973,
+    "bpm_change": 0,
+    "scroll_change": 5,
+    "rhythm_complexity": 1101,
+    "color_complexity": 0.012682773678672012
+  },
+  {
+    "songno": "Destruction 3 2 1",
+    "difficulty": "ura",
+    "note_count": 1491,
+    "density_avg": 9.59714765625,
+    "density_peak": 20,
+    "bpm_avg": 321.32099999999707,
+    "bpm_change": 0,
+    "scroll_change": 8,
+    "rhythm_complexity": 1417,
+    "color_complexity": 0.028926671611070473
+  }
+]

package.json

@@ -5,6 +5,7 @@
     "tja-parser": "^0.2.9"
   },
   "devDependencies": {
+    "@types/bun": "^1.3.13",
     "@types/iconv-lite": "^0.0.1",
     "@types/node": "^25.6.0",
     "typescript": "^6.0.3"

predict.sh

@@ -1,15 +0,0 @@
-#!/bin/bash
-
-# Arguments: script_dir working_dir tja_path difficulty
-SCRIPT_DIR=${1:-"script"}
-WORKING_DIR=${2:-"."}
-TJA_PATH=${3}
-DIFFICULTY=${4:-"oni"}
-
-if [ -z "$TJA_PATH" ]; then
-    echo "Usage: ./redict.sh <script_dir> <working_dir> <tja_path> [difficulty: oni|edit]"
-    exit 1
-fi
-
-echo "Running prediction for $TJA_PATH..."
-bun "$SCRIPT_DIR/predict.ts" "$WORKING_DIR" "$SCRIPT_DIR" "$TJA_PATH" "$DIFFICULTY"

predict/predict_lightgbm.py

@@ -0,0 +1,70 @@
+import argparse
+import json
+import math
+import os
+import joblib
+import numpy as np
+import warnings
+
+# 경고 무시 (Feature name 관련 경고 제거)
+warnings.filterwarnings("ignore", category=UserWarning)
+
+# =========================================================
+# 파일명
+# =========================================================
+
+FEATURES_FILENAME = "features.json"
+MODEL_FILENAME = "model_lgbm.pkl"
+SCALER_FILENAME = "scaler_lgbm.pkl"
+FEATURE_NAMES_FILENAME = "features_lgbm.txt"
+
+def safe_float(value):
+    if value is None: return 0.0
+    x = float(value)
+    return x if math.isfinite(x) else 0.0
+
+def predict(working_dir: str, songno: str, feature: str = None):
+    features_path = os.path.join(working_dir, FEATURES_FILENAME) if feature is None else feature
+    model_path = os.path.join(working_dir, MODEL_FILENAME)
+    scaler_path = os.path.join(working_dir, SCALER_FILENAME)
+    feature_names_path = os.path.join(working_dir, FEATURE_NAMES_FILENAME)
+
+    if not os.path.exists(model_path):
+        raise FileNotFoundError(f"Model not found at {model_path}")
+
+    model = joblib.load(model_path)
+    scaler = joblib.load(scaler_path)
+
+    with open(feature_names_path, "r", encoding="utf-8") as f:
+        feature_names = [line.strip() for line in f.readlines() if line.strip()]
+
+    with open(features_path, "r", encoding="utf-8") as f:
+        data = json.load(f)
+
+    targets = [item for item in data if str(item["songno"]) == str(songno)]
+    if len(targets) == 0:
+        raise ValueError(f"Chart not found: songno={songno}")
+
+    results = []
+    for target in targets:
+        row = [safe_float(target.get(k, 0)) for k in feature_names]
+        X = np.array([row], dtype=np.float32)
+        X = scaler.transform(X)
+        pred = model.predict(X)[0]
+
+        results.append({
+            "songno": str(songno),
+            "diff": target.get("difficulty", "unknown"),
+            "predicted": round(float(pred), 4)
+        })
+
+    print(json.dumps(results, indent=2, ensure_ascii=False))
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--workingDir", required=True)
+    parser.add_argument("--feature", required=False)
+    parser.add_argument("--songno", required=True)
+    args = parser.parse_args()
+
+    predict(args.workingDir, args.songno, args.feature)

predict/predict_xgboost.py

@@ -0,0 +1,164 @@
+import argparse
+import json
+import math
+import os
+import joblib
+import numpy as np
+
+
+# =========================================================
+# 파일명
+# =========================================================
+
+FEATURES_FILENAME = "features.json"
+
+MODEL_FILENAME = "model.pkl"
+SCALER_FILENAME = "scaler.pkl"
+FEATURE_NAMES_FILENAME = "features.txt"
+
+
+# =========================================================
+# safe float
+# =========================================================
+
+def safe_float(value):
+    if value is None:
+        return 0.0
+
+    x = float(value)
+
+    if not math.isfinite(x):
+        return 0.0
+
+    return x
+
+
+# =========================================================
+# 예측 함수
+# =========================================================
+
+def predict(
+    working_dir: str,
+    songno: str,
+    feature: str = None
+):
+    # =====================================================
+    # 경로
+    # =====================================================
+
+    features_path = (
+        os.path.join(working_dir, FEATURES_FILENAME)
+        if feature is None
+        else feature
+    )
+
+    model_path = os.path.join(
+        working_dir,
+        MODEL_FILENAME
+    )
+
+    scaler_path = os.path.join(
+        working_dir,
+        SCALER_FILENAME
+    )
+
+    feature_names_path = os.path.join(
+        working_dir,
+        FEATURE_NAMES_FILENAME
+    )
+
+    # =====================================================
+    # 모델 로드
+    # =====================================================
+
+    model = joblib.load(model_path)
+    scaler = joblib.load(scaler_path)
+
+    # =====================================================
+    # feature 이름 로드
+    # =====================================================
+
+    with open(feature_names_path, "r", encoding="utf-8") as f:
+        feature_names = [
+            line.strip()
+            for line in f.readlines()
+            if line.strip()
+        ]
+
+    # =====================================================
+    # features.json 로드
+    # =====================================================
+
+    with open(features_path, "r", encoding="utf-8") as f:
+        data = json.load(f)
+
+    # =====================================================
+    # target 찾기
+    # =====================================================
+
+    targets = []
+
+    for item in data:
+        if str(item["songno"]) == str(songno):
+            targets.append(item)
+
+    if len(targets) == 0:
+        raise ValueError(f"Chart not found: songno={songno}")
+
+    # =====================================================
+    # feature vector 생성
+    # =====================================================
+
+    results = []
+
+    for target in targets:
+        row = []
+
+        for k in feature_names:
+            value = target.get(k, 0)
+            row.append(safe_float(value))
+
+        X = np.array([row], dtype=np.float32)
+
+        X = scaler.transform(X)
+
+        pred = model.predict(X)[0]
+
+        results.append({
+            "songno": str(songno),
+            "diff": target.get("difficulty", "unknown"),
+            "predicted": round(float(pred), 4)
+        })
+
+    print(json.dumps(results, indent=2, ensure_ascii=False))
+
+
+# =========================================================
+# main
+# =========================================================
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+
+    parser.add_argument(
+        "--workingDir",
+        required=True
+    )
+    
+    parser.add_argument(
+        "--feature",
+        required=False
+    )
+
+    parser.add_argument(
+        "--songno",
+        required=True
+    )
+
+    args = parser.parse_args()
+
+    predict(
+        args.workingDir,
+        args.songno,
+        args.feature
+    )

preprocess/factorize.ts

@@ -0,0 +1,18 @@
+import { Course } from "tja-parser";
+import { filterHitNotes } from "./util";
+
+export type NoteFactor = [type: 0 | 1, bpm: number, scroll: number, delta: number]
+export function factorize(course: Course) {
+    const hitNotes = filterHitNotes(course);
+    const factors: NoteFactor[] = [];
+    for (let i = 0; i < hitNotes.length; i++) {
+        const note = hitNotes[i];
+        factors.push([
+            (note.type === 1 || note.type === 3) ? 0 : 1,
+            note.getBPM() / 100,
+            note.getScroll() / 5,
+            i === 0 ? 0 : (note.getTimingMS() - hitNotes[i - 1].getTimingMS()) / 1000
+        ])
+    }
+    return factors;
+}

preprocess/featurize.ts

@@ -0,0 +1,121 @@
+import { Course, Bar, Note, HitNote } from 'tja-parser';
+
+export namespace Feature {
+    export function filterHitNotes(course: Course) {
+        const notes: HitNote[] = [];
+        course.noteGroups.forEach((g) => {
+            if (g instanceof Bar) {
+                for (const note of g.getNotes()) {
+                    if (note instanceof HitNote) {
+                        notes.push(note);
+                    }
+                }
+            }
+        });
+        return notes;
+    }
+
+    export function getAllNotes(course: Course) {
+        const notes: { type: number, time: number }[] = [];
+        for (const group of course.noteGroups) {
+            if (group instanceof Bar) {
+                notes.push(...group.getNotes().map((note) => ({
+                    //@ts-expect-error
+                    type: note.type,
+                    time: note.getTimingMS()
+                })));
+            }
+        }
+        return notes.sort((a, b) => a.time - b.time).filter(n => [1, 2, 3, 4].includes(n.type));
+    }
+
+    // 밀도 관련
+    export function getAverageDensity(notes: HitNote[]) {
+        if (notes.length === 0) {
+            return 0;
+        }
+        return notes.length / (notes[notes.length - 1].getTimingMS() - notes[0].getTimingMS()) * 1000
+    }
+
+    export function getPeakDensity(notes: HitNote[]) {
+        let peakDensity = 0;
+        for (let i = 0; i < notes.length; i++) {
+            let noteCount = 1;
+            for (let j = i + 1; j < notes.length && (notes[j].getTimingMS() - notes[i].getTimingMS() <= 1000); j++) {
+                noteCount++;
+            }
+            if (peakDensity < noteCount) {
+                peakDensity = noteCount;
+            }
+        }
+        return peakDensity;
+    }
+
+    // BPM 관련
+    export function getAverageBPM(notes: HitNote[]) {
+        if(notes.length === 0) return 0;
+        const averageBPM = notes.reduce((p, note) => p + note.getBPM().valueOf(), 0) / notes.length;
+        return averageBPM;
+    }
+
+    export function getBpmChange(notes: HitNote[]) {
+        let bpmChange = 0;
+        for (let i = 1; i < notes.length; i++) {
+            if (Math.abs(notes[i].getBPM() - notes[i - 1].getBPM()) > 1.5) {
+                bpmChange++;
+            }
+        };
+        return bpmChange;
+    }
+
+    // 스크롤 변화
+    export function getScrollChange(notes: HitNote[]) {
+        let bpmChange = 0;
+        for (let i = 1; i < notes.length; i++) {
+            if (Math.abs(notes[i].getBPM() * notes[i].getScroll() - notes[i - 1].getBPM() * notes[i - 1].getScroll()) > 1.5) {
+                bpmChange++;
+            }
+        };
+        return bpmChange;
+    }
+
+    export function getRhythmComplexity(notes: HitNote[]) {
+        let complexity = 0;
+        for (let i = 2; i < notes.length; i++) {
+            const d1 = notes[i].getTimingMS() - notes[i - 1].getTimingMS()
+            const d2 = notes[i - 1].getTimingMS() - notes[i - 2].getTimingMS()
+
+            const ratio = d1 / d2;
+            const log = Math.log2(ratio);
+
+            if (Math.abs(log - Math.round(log)) < 1e-3) {
+                complexity++;
+            }
+        }
+        return complexity;
+    }
+
+    export function getColorComplexity(notes: HitNote[]) {
+        let complexity = 0;
+        for (let i = 2; i < notes.length; i++) {
+            if (notes[i].type % 2 != notes[i - 2].type % 2) {
+                complexity += 1 / ((notes[i].getTimingMS() - notes[i - 2].getTimingMS()) ** 2)
+            }
+        }
+        return complexity;
+    }
+}
+
+export function featurize(course: Course) {
+    const notes = Feature.filterHitNotes(course)
+    return {
+        note_count: notes.length,
+        density_avg: Feature.getAverageDensity(notes),
+        density_peak: Feature.getPeakDensity(notes),
+        bpm_avg: Feature.getAverageBPM(notes),
+        bpm_change: Feature.getBpmChange(notes),
+        scroll_change: Feature.getScrollChange(notes),
+        rhythm_complexity: Feature.getRhythmComplexity(notes),
+        color_complexity: Feature.getColorComplexity(notes)
+    };
+}

preprocess/parse.ts

@@ -0,0 +1,45 @@
+import tjaParser, { Bar, Branch, Course } from 'tja-parser'
+
+export function parseTja(tja: string): Partial<Record<'oni' | 'edit', Course>> {
+    const song = tjaParser.Song.parse(tja);
+
+    let oni: Course | undefined = undefined;
+    let edit: Course | undefined = undefined;
+
+    if (song.course?.oni) {
+        const noteGroups = song.course.oni.noteGroups;
+        oni = song.course.oni;
+        oni.noteGroups = []
+
+        for (const noteGroup of noteGroups) {
+            if (noteGroup instanceof Bar) {
+                oni.pushNoteGroups(noteGroup)
+            }
+            else if (noteGroup instanceof Branch) {
+                const bar = noteGroup.master || noteGroup.advanced || noteGroup.normal;
+                if (bar) {
+                    oni.pushNoteGroups(...bar)
+                }
+            }
+        }
+    }
+    if (song.course?.edit) {
+        const noteGroups = song.course.edit.noteGroups;
+        edit = song.course.edit;
+        edit.noteGroups = []
+
+        for (const noteGroup of noteGroups) {
+            if (noteGroup instanceof Bar) {
+                edit.pushNoteGroups(noteGroup)
+            }
+            else if (noteGroup instanceof Branch) {
+                const bar = noteGroup.master || noteGroup.advanced || noteGroup.normal;
+                if (bar) {
+                    edit.pushNoteGroups(...bar)
+                }
+            }
+        }
+    }
+
+    return { oni, edit }
+}

preprocess/util.ts

@@ -0,0 +1,15 @@
+import { Bar, Course, HitNote } from "tja-parser";
+
+export function filterHitNotes(course: Course): HitNote[] {
+    const notes: HitNote[] = [];
+    course.noteGroups.forEach((g) => {
+        if (g instanceof Bar) {
+            for (const note of g.getNotes()) {
+                if (note instanceof HitNote) {
+                    notes.push(note);
+                }
+            }
+        }
+    });
+    return notes;
+}

script/compare.ts

@@ -0,0 +1,148 @@
+import Bun from 'bun';
+import path from 'node:path';
+import { parseArgs } from 'node:util';
+import fs from 'node:fs';
+
+const { values } = parseArgs({
+    args: Bun.argv,
+    options: {
+        workingDir: { type: "string" },
+        dataDir: { type: "string" },
+        script: { type: "string" },
+    },
+    strict: true,
+    allowPositionals: true,
+});
+
+if (!values.workingDir || !values.dataDir || !values.script) {
+    console.error("Usage: bun run script/compare.ts --workingDir <dir> --dataDir <dir> --script <python_script>");
+    process.exit(1);
+}
+
+const workingDir = values.workingDir;
+const dataDir = values.dataDir;
+const predictScript = values.script;
+const tempFileName = "temp.json";
+const tempFilePath = path.join(workingDir, tempFileName);
+
+// 1. 전처리 실행 (temp.json 생성하여 기존 features.json 보존)
+console.log("Step 1: Running preprocessing to temp.json...");
+const preprocessResult = Bun.spawnSync([
+    "bun", "run", "script/preprocess.ts",
+    "--workingDir", workingDir,
+    "--dataDir", dataDir,
+    "--fileName", tempFileName
+]);
+
+if (!preprocessResult.success) {
+    console.error("Preprocessing failed");
+    console.error(preprocessResult.stderr.toString());
+    process.exit(1);
+}
+
+// 2. measure.csv 로드
+console.log("Step 2: Loading measure.csv...");
+const measurePath = path.join(dataDir, "measure.csv");
+const measureContent = fs.readFileSync(measurePath, "utf-8");
+const measureMap = new Map<string, number>();
+
+measureContent.split("\n").forEach((line, index) => {
+    if (index === 0 || !line.trim()) return;
+    const parts = line.split(",");
+    if (parts.length >= 3) {
+        const constant = parts[0];
+        const songno = parts[1];
+        const diff = parts[2];
+        measureMap.set(`${songno.trim()}_${diff.trim()}`, parseFloat(constant));
+    }
+});
+
+// 3. temp.json 로드하여 대상 곡 목록 추출
+const features = JSON.parse(fs.readFileSync(tempFilePath, "utf-8"));
+const uniqueSongnos = Array.from(new Set(features.map((f: any) => f.songno)));
+
+// 4. 예측 및 비교
+console.log(`Step 3: Predicting and comparing ${uniqueSongnos.length} songs...`);
+const comparisonResults: any[] = [];
+let processedCount = 0;
+
+for (const songno of uniqueSongnos) {
+    try {
+        const predictProcess = Bun.spawnSync([
+            "python3", predictScript,
+            "--workingDir", workingDir,
+            "--songno", songno as string,
+            "--feature", tempFilePath
+        ]);
+
+        if (!predictProcess.success) {
+            console.error(`\n[ERROR] Failed to predict songno ${songno}`);
+            console.error(predictProcess.stderr.toString());
+            processedCount++;
+            continue;
+        }
+
+        const output = predictProcess.stdout.toString().trim();
+        // JSON 부분만 추출 (경고문 등이 섞여있을 경우 대비)
+        const jsonStart = output.indexOf('[');
+        const jsonEnd = output.lastIndexOf(']') + 1;
+        
+        if (jsonStart === -1 || jsonEnd === 0) {
+            console.error(`\n[ERROR] Invalid output format for songno ${songno}`);
+            processedCount++;
+            continue;
+        }
+
+        const predictions = JSON.parse(output.substring(jsonStart, jsonEnd));
+
+        predictions.forEach((pred: any) => {
+            const key = `${pred.songno}_${pred.diff}`;
+            const actual = measureMap.get(key);
+
+            if (actual !== undefined) {
+                comparisonResults.push({
+                    songno: pred.songno,
+                    diff: pred.diff,
+                    actual: actual,
+                    predicted: pred.predicted,
+                    error: Math.abs(actual - pred.predicted)
+                });
+            }
+        });
+
+        processedCount++;
+        if (processedCount % 10 === 0 || processedCount === uniqueSongnos.length) {
+            const percent = ((processedCount / uniqueSongnos.length) * 100).toFixed(1);
+            process.stdout.write(`\rProgress: ${processedCount}/${uniqueSongnos.length} (${percent}%)   `);
+        }
+    } catch (err) {
+        console.error(`\nError processing songno ${songno}:`, err);
+        processedCount++;
+    }
+}
+console.log("\nPrediction finished.");
+
+// 5. 결과 분석 및 저장
+if (comparisonResults.length === 0) {
+    console.error("No comparison results were generated.");
+    process.exit(1);
+}
+
+const avgError = comparisonResults.reduce((acc, curr) => acc + curr.error, 0) / comparisonResults.length;
+const resultData = {
+    summary: {
+        total_compared: comparisonResults.length,
+        average_absolute_error: avgError,
+        timestamp: new Date().toISOString(),
+        script_used: predictScript
+    },
+    details: comparisonResults.sort((a, b) => b.error - a.error)
+};
+
+const comparePath = path.join(workingDir, "compare.json");
+fs.writeFileSync(comparePath, JSON.stringify(resultData, null, 2), "utf-8");
+
+console.log(`\nComparison complete!`);
+console.log(`Total compared: ${comparisonResults.length}`);
+console.log(`Average Error: ${avgError.toFixed(4)}`);
+console.log(`Results saved to: ${comparePath}`);

script/factor.json

@@ -1,7 +0,0 @@
-{
-  "physical_density": 1,
-  "stamina_requirement": 1,
-  "pattern_complexity": 1,
-  "rhythmic_complexity": 1,
-  "reading_gimmick": 1
-}

script/factorize.ts

@@ -1,86 +0,0 @@
-import { Bar, Course } from 'tja-parser';
-
-export interface Factors {
-    physical_density: number;
-    stamina_requirement: number;
-    pattern_complexity: number;
-    rhythmic_complexity: number;
-    reading_gimmick: number;
-}
-
-export namespace Factor {
-    export function getAllNotes(course: Course): any[] {
-        const notes: any[] = [];
-        for (const group of course.noteGroups) {
-            if (group instanceof Bar) {
-                notes.push(...group.getNotes());
-            }
-        }
-        return notes;
-    }
-
-    export function getPhysicalDensity(course: Course): number {
-        const notes = getAllNotes(course);
-        if (notes.length === 0) return 0;
-        const bars = course.noteGroups.length;
-        return notes.length / (bars || 1);
-    }
-
-    export function getStaminaRequirement(course: Course): number {
-        const notes = getAllNotes(course);
-        let maxStream = 0;
-        let currentStream = 0;
-        
-        for (const note of notes) {
-            if (note.type !== '0') {
-                currentStream++;
-            } else {
-                maxStream = Math.max(maxStream, currentStream);
-                currentStream = 0;
-            }
-        }
-        return Math.max(maxStream, currentStream) / 100;
-    }
-
-    export function getPatternComplexity(course: Course): number {
-        const notes = getAllNotes(course).filter(n => n.type === '1' || n.type === '2');
-        let transitions = 0;
-        
-        for (let i = 1; i < notes.length; i++) {
-            if (notes[i].type !== notes[i-1].type) {
-                transitions++;
-            }
-        }
-        return notes.length > 0 ? transitions / notes.length : 0;
-    }
-
-    export function getRhythmicComplexity(course: Course): number {
-        let complexNotes = 0;
-        const allNotes = getAllNotes(course);
-        for (const group of course.noteGroups) {
-            if (group instanceof Bar) {
-                const division = group.getNotes().length;
-                if (division % 4 !== 0 || division > 16) {
-                    complexNotes += division;
-                }
-            }
-        }
-        return allNotes.length > 0 ? complexNotes / allNotes.length : 0;
-    }
-
-    export function getReadingGimmick(course: Course): number {
-        // BPM 변화나 Scroll 변화 빈도 측정
-        // tja-parser의 구조에 따라 구현 (여기서는 placeholder)
-        return 0;
-    }
-}
-
-export function factorize(course: Course): Factors {
-    return {
-        physical_density: Factor.getPhysicalDensity(course),
-        stamina_requirement: Factor.getStaminaRequirement(course),
-        pattern_complexity: Factor.getPatternComplexity(course),
-        rhythmic_complexity: Factor.getRhythmicComplexity(course),
-        reading_gimmick: Factor.getReadingGimmick(course)
-    };
-}

script/parse.ts

@@ -1,51 +0,0 @@
-import tjaParser, { Bar, Branch, Course } from 'tja-parser'
-
-export function parseTja(tja: string): Partial<Record<'oni' | 'edit', Course>> | null {
-    try {
-        const song = tjaParser.Song.parse(tja);
-
-        let oni: Course | undefined = undefined;
-        let edit: Course | undefined = undefined;
-
-        if (song.course?.oni) {
-            const noteGroups = song.course.oni.noteGroups;
-            oni = song.course.oni;
-            oni.noteGroups = []
-
-            for (const noteGroup of noteGroups) {
-                if (noteGroup instanceof Bar) {
-                    oni.pushNoteGroups(noteGroup)
-                }
-                else if (noteGroup instanceof Branch) {
-                    const bar = noteGroup.master || noteGroup.advanced || noteGroup.normal;
-                    if (bar) {
-                        oni.pushNoteGroups(...bar)
-                    }
-                }
-            }
-        }
-        if (song.course?.edit) {
-            const noteGroups = song.course.edit.noteGroups;
-            edit = song.course.edit;
-            edit.noteGroups = []
-
-            for (const noteGroup of noteGroups) {
-                if (noteGroup instanceof Bar) {
-                    edit.pushNoteGroups(noteGroup)
-                }
-                else if (noteGroup instanceof Branch) {
-                    const bar = noteGroup.master || noteGroup.advanced || noteGroup.normal;
-                    if (bar) {
-                        edit.pushNoteGroups(...bar)
-                    }
-                }
-            }
-        }
-
-        return { oni, edit }
-    }
-    catch (err){
-        console.error(err)
-        return null;
-    }
-}

script/predict.ts

@@ -1,38 +0,0 @@
-import { parseTja } from './parse';
-import { factorize, Factors } from './factorize';
-import { join } from 'path';
-import { readFileSync } from 'fs';
-
-async function predict() {
-    const args = process.argv.slice(2);
-    const workingDir = args[0];
-    const scriptDir = args[1];
-    const tjaPath = args[2];
-    const difficulty = (args[3] || 'oni').toLowerCase() as 'oni' | 'edit';
-
-    if (!tjaPath) {
-        console.error('Usage: bun script/predict.ts <working_dir> <script_dir> <tja_path> <difficulty>');
-        process.exit(1);
-    }
-
-    const factorJsonPath = join(scriptDir, 'factor.json');
-    const weights: Factors = JSON.parse(readFileSync(factorJsonPath, 'utf-8'));
-
-    const tjaContent = readFileSync(tjaPath, 'utf-8');
-    const parsed = parseTja(tjaContent);
-
-    if (!parsed || !parsed[difficulty]) {
-        console.error(`Error: Could not find difficulty '${difficulty}' in TJA.`);
-        process.exit(1);
-    }
-
-    const factors = factorize(parsed[difficulty]!);
-    const prediction = Object.keys(factors).reduce((sum, key) => 
-        sum + (factors[key as keyof Factors] * weights[key as keyof Factors]), 0);
-
-    console.log(`Prediction for ${tjaPath} (${difficulty}):`);
-    console.log(`- Predicted Value: ${prediction.toFixed(4)}`);
-    console.log('- Factors:', factors);
-}
-
-predict();

script/predict_tja.ts

@@ -0,0 +1,94 @@
+import Bun from 'bun';
+import path from 'node:path';
+import { parseArgs } from 'node:util';
+import fs from 'node:fs';
+import { featurize } from '../preprocess/featurize';
+import { parseTja } from '../preprocess/parse';
+
+const { values } = parseArgs({
+    args: Bun.argv,
+    options: {
+        tja: {
+            type: "string"
+        },
+        workingDir: {
+            type: "string"
+        },
+        script: {
+            type: "string"
+        }
+    },
+    allowPositionals: true,
+});
+
+if (!values.tja || !values.workingDir || !values.script) {
+    console.error("Usage: bun script/predict_tja.ts --tja <path_to_tja> [--workingDir <dir>] [--script <python_script>]");
+    process.exit(1);
+}
+
+const tjaPath = values.tja;
+if (!fs.existsSync(tjaPath)) {
+    console.error(`File not found: ${tjaPath}`);
+    process.exit(1);
+}
+
+const workingDir = values.workingDir!;
+const pythonScript = values.script!;
+
+if (!fs.existsSync(pythonScript)) {
+    console.error(`Python script not found: ${pythonScript}`);
+    process.exit(1);
+}
+
+// 1. TJA 파싱 및 피처 추출
+const tjaContent = fs.readFileSync(tjaPath, 'utf-8');
+const songno = path.basename(tjaPath, '.tja');
+const parsed = parseTja(tjaContent);
+
+const features: any[] = [];
+if (parsed.oni) {
+    features.push({
+        songno,
+        difficulty: 'oni',
+        ...featurize(parsed.oni)
+    });
+}
+if (parsed.edit) {
+    features.push({
+        songno,
+        difficulty: 'ura',
+        ...featurize(parsed.edit)
+    });
+}
+
+if (features.length === 0) {
+    console.error("No Oni or Ura difficulty found in the TJA file.");
+    process.exit(1);
+}
+
+// 2. 임시 피처 파일 저장
+const tempFeaturePath = path.join(workingDir, `temp_predict_${Date.now()}.json`);
+fs.writeFileSync(tempFeaturePath, JSON.stringify(features, null, 2), 'utf-8');
+
+try {
+    // 3. Python 예측 스크립트 실행
+    const proc = Bun.spawnSync([
+        "python3",
+        pythonScript,
+        "--workingDir", workingDir,
+        "--feature", tempFeaturePath,
+        "--songno", songno
+    ]);
+
+    if (proc.success) {
+        console.log(proc.stdout.toString());
+    } else {
+        console.error("Prediction failed:");
+        console.error(proc.stderr.toString());
+    }
+} finally {
+    // 4. 임시 파일 삭제
+    if (fs.existsSync(tempFeaturePath)) {
+        fs.unlinkSync(tempFeaturePath);
+    }
+}

script/preprocess.ts

@@ -0,0 +1,66 @@
+import Bun from 'bun';
+import path from 'node:path';
+import { parseArgs } from 'node:util';
+import fs, { mkdirSync } from 'node:fs';
+import { featurize } from '../preprocess/featurize';
+import { parseTja } from '../preprocess/parse'
+
+const { values } = parseArgs({
+    args: Bun.argv,
+    options: {
+        workingDir: {
+            type: "string"
+        },
+        dataDir: {
+            type: "string"
+        },
+        fileName: {
+            type: "string"
+        }
+    },
+    allowPositionals: true,
+})
+
+if (!values.dataDir || !values.workingDir) {
+    console.error("--workingDir --dataDir");
+    process.exit(1);
+}
+
+const workingDir = values.workingDir ?? '';
+if (!fs.existsSync(workingDir)) mkdirSync(workingDir)
+const dataDir = values.dataDir ?? '';
+
+const tjaDir = path.join(dataDir, 'tja');
+const files = fs.readdirSync(tjaDir);
+
+const features: ({ songno: string, difficulty: 'oni' | 'ura' } & {})[] = [];
+for (const file of files) {
+    const tja = fs.readFileSync(path.join(tjaDir, file), 'utf-8');
+    const songno = path.basename(file, '.tja');
+    try {
+        const parsed = parseTja(tja);
+        const oni = parsed?.oni;
+        const edit = parsed?.edit;
+        if (oni) {
+            features.push({
+                songno,
+                difficulty: 'oni',
+                ...featurize(oni)
+            })
+        }
+        if (edit) {
+            features.push({
+                songno,
+                difficulty: 'ura',
+                ...featurize(edit)
+            })
+        }
+    }
+    catch (err) {
+        console.error(err);
+        console.error(file);
+    }
+}
+
+const featurePath = path.join(workingDir, values.fileName ?? 'features.json');
+fs.writeFileSync(featurePath, JSON.stringify(features, null, 2), 'utf-8');

script/train.py

@@ -1,67 +0,0 @@
-import pandas as pd
-import json, os, sys, joblib
-import numpy as np
-from sklearn.ensemble import GradientBoostingRegressor
-from sklearn.model_selection import train_test_split
-
-def train(script_dir, working_dir, data_dir, train_count, val_count, margin):
-    features = pd.read_json(os.path.join(working_dir, 'features.json'))
-    measures = pd.read_csv(os.path.join(data_dir, 'measure.csv'))
-    
-    features['songno'] = features['songno'].astype(int)
-    measures['songno'] = measures['songno'].astype(int)
-    measures['diff'] = measures['diff'].replace('ura', 'edit')
-    
-    df = pd.merge(features, measures, on=['songno', 'diff'])
-    
-    with open(os.path.join(script_dir, 'factor.json'), 'r') as f:
-        weights = json.load(f)
-    for col in ['physical_density', 'stamina_requirement', 'pattern_complexity', 'rhythmic_complexity', 'reading_gimmick']:
-        df[col] = df[col] * weights.get(col, 1.0)
-
-    X_cols = ['physical_density', 'stamina_requirement', 'pattern_complexity', 'rhythmic_complexity', 'reading_gimmick']
-    model_path = os.path.join(working_dir, 'model.pkl')
-    model = joblib.load(model_path) if os.path.exists(model_path) else GradientBoostingRegressor(n_estimators=200, learning_rate=0.05, max_depth=3)
-
-    iteration = 1
-    while True:
-        # 3. 데이터 샘플링
-        df_sample = df.sample(n=int(train_count) + int(val_count))
-        train_df, val_df = train_test_split(df_sample, test_size=int(val_count))
-
-        X_train, y_train = train_df[X_cols], train_df['상수']
-        X_val, y_val = val_df[X_cols], val_df['상수']
-
-        # 4. 학습 (최대 10회)
-        for attempt in range(1, 11):
-            model.fit(X_train, y_train)
-            train_err = np.max(np.abs(np.clip(model.predict(X_train), 1.0, 12.0) - y_train))
-            print(f"Iteration {iteration} - Attempt {attempt} - Train Error: {train_err:.4f}")
-            if train_err <= float(margin): break
-            model.set_params(n_estimators=model.n_estimators + 50)
-
-        # 5. 검증
-        pred_val = np.clip(model.predict(X_val), 1.0, 12.0)
-        val_errors = np.abs(pred_val - y_val)
-        
-        # 6. 검증 실패 시 재시도
-        if np.any(val_errors > float(margin)):
-            print(f"Validation failed (max error: {np.max(val_errors):.4f}). Retrying...")
-            iteration += 1
-            continue
-            
-        val_result = pd.DataFrame({
-            'songno': val_df['songno'],
-            'difficulty': val_df['diff'],
-            'measure': y_val,
-            'predicted_measure': pred_val,
-            'error': val_errors
-        })
-        val_result.to_csv(os.path.join(working_dir, f'validate_result_{iteration}.csv'), index=False)
-        val_result.to_csv(os.path.join(working_dir, 'validate_result.csv'), index=False)
-        break
-
-    joblib.dump(model, model_path)
-
-if __name__ == "__main__":
-    train(sys.argv[1], sys.argv[2], sys.argv[3], sys.argv[4], sys.argv[5], sys.argv[6])

script/train.ts

@@ -1,108 +1,91 @@
-import { parseTja } from './parse';
-import { factorize, Factors } from './factorize';
-import { join } from 'path';
-import { readFileSync, writeFileSync, readdirSync } from 'fs';
-import { parse } from 'csv-parse/sync';
+import Bun from 'bun';
+import { execSync, spawn, spawnSync } from 'node:child_process';
+import { parseArgs } from 'node:util';
+import fs from 'fs';
+import path from 'path';
+import { featurize } from '../preprocess/featurize';
+import { parseTja } from '../preprocess/parse'
 
-async function train() {
-    const args = process.argv.slice(2);
-    const workingDir = args[0];
-    const scriptDir = args[1];
-    const dataDir = args[2];
-    const trainCount = parseInt(args[3]) || 100;
-    const validateCount = parseInt(args[4]) || 20;
-    const margin = parseFloat(args[5]) || 0.1;
-
-    const factorJsonPath = join(scriptDir, 'factor.json');
- // 항상 script/ 위치 참조
-    let weights: Factors = JSON.parse(readFileSync(factorJsonPath, 'utf-8'));
-    
-    // 결과물(로그 등)을 저장할 경로 (필요 시 활용)
-    const logPath = join(workingDir, 'training_log.txt'); 
-
-    const measureCsv = readFileSync(join(dataDir, 'measure.csv'), 'utf-8');
-    const records = parse(measureCsv, { columns: true, skip_empty_lines: true });
-
-    const tjaFiles = readdirSync(join(dataDir, 'tja')).filter(f => f.endsWith('.tja'));
-    const shuffled = tjaFiles.sort(() => 0.5 - Math.random());
-    const trainFiles = shuffled.slice(0, trainCount);
-    const validateFiles = shuffled.slice(trainCount, trainCount + validateCount);
-
-    console.log(`Training with ${trainFiles.length} files...`);
-
-    // 학습 로직 (단순화된 경사 하강법 또는 반복 최적화)
-    let error = Infinity;
-    let iterations = 0;
-    while (error > margin && iterations < 100) {
-        let totalError = 0;
-        let count = 0;
-
-        for (const file of trainFiles) {
-            const songno = file.replace(/\D/g, '');
-            const tjaContent = readFileSync(join(dataDir, 'tja', file), 'utf-8');
-            const parsed = parseTja(tjaContent);
-            if (!parsed) continue;
-
-            for (const diff of ['oni', 'edit'] as const) {
-                const course = parsed[diff];
-                if (!course) continue;
-
-                const target = records.find((r: any) => r.songno === songno && (r.diff === (diff === 'oni' ? 'oni' : 'ura')));
-                if (!target) {
-                    // console.log(`[!] No target for ${songno} diff=${diff}`);
-                    continue;
-                }
-
-                const factors = factorize(course);
-                const prediction = Object.keys(factors).reduce((sum, key) => 
-                    sum + (factors[key as keyof Factors] * weights[key as keyof Factors]), 0);
-                
-                const targetValue = parseFloat(target.상수);
-                const diff_val = targetValue - prediction;
-                totalError += Math.abs(diff_val);
-                count++;
-
-                // 가중치 업데이트
-                for (const key in weights) {
-                    const k = key as keyof Factors;
-                    weights[k] += diff_val * factors[k] * 0.05; // 학습률 조정
-                }
-            }
+const { values } = parseArgs({
+    args: Bun.argv,
+    options: {
+        workingDir: {
+            type: "string"
+        },
+        dataDir: {
+            type: "string"
+        },
+        trainScript: {
+            type: "string"
+        },
+        trainSize: {
+            type: 'string'
+        },
+        validSize: {
+            type: 'string'
         }
-        error = totalError / (count || 1);
-        console.log(`Iteration ${iterations}: Mean Error = ${error.toFixed(4)}`);
-        iterations++;
-    }
+    },
+    allowPositionals: true,
+});
 
-    writeFileSync(factorJsonPath, JSON.stringify(weights, null, 2));
-    writeFileSync(join(workingDir, 'training_result.json'), JSON.stringify({ finalError: error, weights }, null, 2));
-    console.log(`Training complete. Weights saved to ${factorJsonPath}, result saved to ${workingDir}`);
-
-    // 검증 로직
-    console.log('\nValidation Results:');
-    for (const file of validateFiles) {
-        const songno = file.replace(/\D/g, '');
-        const tjaContent = readFileSync(join(dataDir, 'tja', file), 'utf-8');
-        const parsed = parseTja(tjaContent);
-        if (!parsed) continue;
-
-        for (const diff of ['oni', 'edit'] as const) {
-            const course = parsed[diff];
-            if (!course) continue;
-
-            const target = records.find((r: any) => r.songno === songno && (r.diff === (diff === 'oni' ? 'oni' : 'ura')));
-            if (!target) {
-                console.log(`[!] No match for ${songno} diff=${diff === 'oni' ? 'oni' : 'ura'}`);
-                continue;
-            }
-
-            const factors = factorize(course);
-            const prediction = Object.keys(factors).reduce((sum, key) => 
-                sum + (factors[key as keyof Factors] * weights[key as keyof Factors]), 0);
-            
-            console.log(`[${songno}] Target: ${target.상수}, Predicted: ${prediction.toFixed(2)}, Diff: ${Math.abs(parseFloat(target.상수) - prediction).toFixed(2)}`);
-        }
-    }
+if (!values.dataDir || !values.workingDir || !values.trainScript) {
+    console.error("--workingDir --dataDir --trainDir");
+    process.exit(1);
 }
 
-train();
+generateFeatures();
+const child = spawn("python3", [values.trainScript,
+    "--workingDir", values.workingDir,
+    "--dataDir", values.dataDir,
+    "--trainSize", (Number(values.trainSize) || 500).toString(),
+    "--validSize", (Number(values.validSize) || 100).toString(),
+]);
+child.stdout.pipe(process.stdout);
+child.stderr.pipe(process.stderr);
+child.on("close", () => {
+    process.exit()
+})
+
+// funcs
+function generateFeatures() {
+    const workingDir = values.workingDir ?? '';
+    if (!fs.existsSync(workingDir)) fs.mkdirSync(workingDir);
+    const featurePath = path.join(workingDir, 'features.json');
+    if (fs.existsSync(featurePath)) return;
+    const dataDir = values.dataDir ?? '';
+
+    const tjaDir = path.join(dataDir, 'tja');
+    const files = fs.readdirSync(tjaDir);
+
+    const features: ({ songno: string, difficulty: 'oni' | 'ura' } & {})[] = [];
+    for (const file of files) {
+        const tja = fs.readFileSync(path.join(tjaDir, file), 'utf-8');
+        const songno = path.basename(file, '.tja');
+        try {
+            const parsed = parseTja(tja);
+            const oni = parsed?.oni;
+            const edit = parsed?.edit;
+            if (oni) {
+                features.push({
+                    songno,
+                    difficulty: 'oni',
+                    ...featurize(oni)
+                })
+            }
+            if (edit) {
+                features.push({
+                    songno,
+                    difficulty: 'ura',
+                    ...featurize(edit)
+                })
+            }
+        }
+        catch (err) {
+            console.error(err);
+            console.error(file);
+        }
+    }
+
+    fs.writeFileSync(featurePath, JSON.stringify(features, null, 2), 'utf-8');
+    console.log('features.json generated')
+}

spec.md

@@ -1,75 +0,0 @@
-# 학습 
-## CLI
-- cli로 학습할 수 있는 쉘 스크립트를 작성한다.
-- argument
-    - script_dir
-    - working_dir
-    - data_dir
-    - train_data_count = 100
-    - validate_data_count = 20
-    - margin = 0.1
-
-## 학습 모델
-
-## 학습 과정
-1. `{data_dir}/tja`의 tja를 파싱한다.
-    - `{script_dir}/parse.ts`를 사용한다.
-2. `oni`와 `edit` course를 factorize한다.
-    - `{script_dir}/factorize.ts`를 사용하며, `{working_dir}/features.json`에 저장한다. 이 때, songno와 난이도(oni | edit)을 같이 저장하여야한다.
-    - `{working_dir}/features.json`이 이미 존재하면 이 과정은 생략한다.
-3. factorized된 데이터를 `train_data_count` 만큼 랜덤으로 가져온다.
-4. train data로 학습을 진행한다.
-    - 예측 값이 [1, 12]에 없으면 완전히 잘못 트레이닝 된 것이다.
-    - `{data_dir}/measure.csv`로 오차 확인
-    - course가 `edit`이면 measure.json에서는 `ura`이므로 적절히 매칭한다.
-    - Train error를 계속 로깅한다.
-    - Train error가 `margin`보다 클 경우 이 과정(4번)을 다시 진행한다.
-        - 최대 10번만 진행한다.
-5. validate_data_count 만큼 랜덤으로 validate data를 가져온 후 validate data로 각각의 measure, 예측한 measure, 오차를 `validate_result_{n}.csv`로 저장한다.
-    - `validate_result_{n}.csv`에는 `songno`, `difficulty`, `measure`, `predicted_measure`, `error`을 작성한다. (중요)
-    - `n`에는 몇번째 iteration인지 적는다.
-6. 만약 validate한 데이터 중 오차가 `margin`보다 큰 것이 있다면 3~5과정을 다시 반복한다.
-7. 모델 파일을 저장한다.
-
-### 유의 사항
-- 모든 출력되는 데이터는 working_dir에 저장한다.
-- 이미 working_dir에 모델이 존재하면 모델을 삭제하지 않고 업데이트하는 식으로 학습한다.
-- 애플 실리콘 cpu를 사용할 경우 적극적으로 gpu를 사용할 수 있도록 한다.
-
-# 예측
-## CLi
-- cli로 예측 데이터를 뽑을 수 있는 쉘 스크립트를 작성한다.
-- argument
-    - working_dir
-    - tja
-    - difficulty: oni | edit
-
-## 예측 과정
-- `working_dir`에 있는 model로 예측을 진행한다.
-- `tja` argument에 tja파일이 존재한다.
-- `difficulty`로 예측을 진행할 난이도를 선택한다.
-
-# Factor
-## 종류
-- physical_density: 곡의 전반적인 및 순간적인 타격 밀도 (Global NPS, Peak NPS)
-- stamina_requirement: 긴 연타 구간과 휴식 구간의 비율을 통한 체력 소모량
-- pattern_complexity: 색상 전환 및 손 배치 전환(Hand-switching)의 복잡도
-- rhythmic_complexity: 엇박, 다양한 음표 단위 사용 등으로 인한 리듬의 난해함
-- reading_gimmick: BPM 변화, 스크롤 속도(SV) 변동 등 시각적 요소 및 기믹
-
-## 가중치 (하이퍼 파라미터)
-`script/factor.json`에 `Record<factor 이름, 가중치>`를 기록한다.
-
-# Scripts
-## `script/parse.ts`
-- `parseTja` 함수
-    - tja를 string으로 받는다.
-    - `Partial<Record<'oni' | 'edit', Course>> | null`을 반환한다.
-    - null이 반환되면 에러가 나온 것이므로 무시한다.
-## `script/factorize.ts`
-- `factorize` 함수
-    - course를 input으로 받는다.
-    - factor들을 반환한다.
-- `Factor` namespace
-    - 각 factor을 반환하는 함수의 namespace이다.
-    - `factorize` 함수는 `Factor` namespace에서 함수들을 호출하여 factor들을 반환한다.

train.sh

@@ -1,23 +0,0 @@
-#!/bin/bash
-# Arguments: script_dir working_dir data_dir train_data_count validate_data_count margin
-SCRIPT_DIR=${1:-"script"}
-WORKING_DIR=${2:-"."}
-DATA_DIR=${3:-"datas"}
-TRAIN_COUNT=${4:-100}
-VAL_COUNT=${5:-20}
-MARGIN=${6:-0.1}
-
-# 작업 디렉토리가 없으면 생성
-if [ ! -d "$WORKING_DIR" ]; then
-    echo "Creating directory: $WORKING_DIR"
-    mkdir -p "$WORKING_DIR"
-fi
-
-# 피처 추출 (존재 시 생략)
-if [ ! -f "$WORKING_DIR/features.json" ]; then
-    echo "Extracting features..."
-    bun "$SCRIPT_DIR/extract.ts" "$WORKING_DIR" "$DATA_DIR" "$TRAIN_COUNT" "$VAL_COUNT"
-fi
-
-echo "Training model..."
-python3 "$SCRIPT_DIR/train.py" "$SCRIPT_DIR" "$WORKING_DIR" "$DATA_DIR" "$TRAIN_COUNT" "$VAL_COUNT" "$MARGIN"

train/train_lightgbm.py

@@ -0,0 +1,152 @@
+import argparse
+import csv
+import json
+import math
+import os
+import random
+import joblib
+import numpy as np
+import lightgbm as lgb
+from sklearn.preprocessing import StandardScaler
+from sklearn.metrics import mean_absolute_error
+
+# =========================================================
+# Hyper Parameters
+# =========================================================
+
+TRAIN_SIZE = 0
+VALID_SIZE = 0
+
+RANDOM_STATE = 42
+
+# LightGBM 특정 하이퍼파라미터
+PARAMS = {
+    'objective': 'regression',
+    'metric': 'mae',
+    'verbosity': -1,
+    'boosting_type': 'gbdt',
+    'random_state': RANDOM_STATE,
+    'learning_rate': 0.02, # 더 정밀한 학습을 위해 하향
+    'num_leaves': 63,      # 더 복잡한 패턴 학습을 위해 상향
+    'feature_fraction': 0.9,
+    'bagging_fraction': 0.8,
+    'bagging_freq': 5,
+    'n_estimators': 3000   # 학습량 대폭 상향
+}
+
+CONTINUE_TRAINING = True
+ERROR_TOLERANCE = 0.1
+
+# =========================================================
+# 파일명
+# =========================================================
+
+FEATURES_FILENAME = "features.json"
+MEASURE_FILENAME = "measure.csv"
+MODEL_FILENAME = "model_lgbm.pkl"
+SCALER_FILENAME = "scaler_lgbm.pkl"
+FEATURE_NAMES_FILENAME = "features_lgbm.txt"
+
+IGNORE_KEYS = {"songno", "difficulty"}
+
+def safe_float(value):
+    if value is None: return 0.0
+    x = float(value)
+    return x if math.isfinite(x) else 0.0
+
+def train_model(working_dir: str, data_dir: str):
+    random.seed(RANDOM_STATE)
+
+    features_path = os.path.join(working_dir, FEATURES_FILENAME)
+    measure_path = os.path.join(data_dir, MEASURE_FILENAME)
+    model_path = os.path.join(working_dir, MODEL_FILENAME)
+    scaler_path = os.path.join(working_dir, SCALER_FILENAME)
+    feature_names_path = os.path.join(working_dir, FEATURE_NAMES_FILENAME)
+
+    with open(features_path, "r", encoding="utf-8") as f:
+        feature_data = json.load(f)
+
+    if len(feature_data) == 0:
+        raise ValueError("features.json is empty")
+
+    feature_map = {(str(item["songno"]), str(item["difficulty"])): item for item in feature_data}
+    feature_names = sorted([k for k in feature_data[0].keys() if k not in IGNORE_KEYS])
+
+    dataset = []
+    with open(measure_path, "r", encoding="utf-8") as f:
+        reader = csv.reader(f)
+        next(reader, None)
+        for row in reader:
+            if len(row) < 3: continue
+            measure, songno, diff = safe_float(row[0]), str(row[1]), str(row[2])
+            key = (songno, diff)
+            if key in feature_map:
+                features = [safe_float(feature_map[key].get(k, 0)) for k in feature_names]
+                dataset.append((features, measure))
+
+    random.shuffle(dataset)
+    if len(dataset) < (TRAIN_SIZE + VALID_SIZE):
+        raise ValueError(f"Not enough dataset ({len(dataset)} < {TRAIN_SIZE + VALID_SIZE})")
+
+    train_dataset = dataset[:TRAIN_SIZE]
+    valid_dataset = dataset[TRAIN_SIZE:TRAIN_SIZE + VALID_SIZE]
+
+    X_train = np.array([x for x, _ in train_dataset], dtype=np.float32)
+    y_train = np.array([y for _, y in train_dataset], dtype=np.float32)
+    X_valid = np.array([x for x, _ in valid_dataset], dtype=np.float32)
+    y_valid = np.array([y for _, y in valid_dataset], dtype=np.float32)
+
+    print(f"Train Size: {len(X_train)} | Valid Size: {len(X_valid)} | Features: {len(feature_names)}")
+
+    if CONTINUE_TRAINING and os.path.exists(scaler_path):
+        print("Loading existing scaler...")
+        scaler = joblib.load(scaler_path)
+    else:
+        print("Creating new scaler...")
+        scaler = StandardScaler()
+        scaler.fit(X_train)
+
+    X_train = scaler.transform(X_train)
+    X_valid = scaler.transform(X_valid)
+
+    if CONTINUE_TRAINING and os.path.exists(model_path):
+        print("Loading existing model for incremental training...")
+        model = joblib.load(model_path)
+        model.fit(
+            X_train, y_train,
+            eval_set=[(X_valid, y_valid)],
+            init_model=model,
+            callbacks=[lgb.early_stopping(stopping_rounds=100)]
+        )
+    else:
+        print("Creating new LightGBM model...")
+        model = lgb.LGBMRegressor(**PARAMS)
+        model.fit(
+            X_train, y_train,
+            eval_set=[(X_valid, y_valid)],
+            callbacks=[lgb.early_stopping(stopping_rounds=100)]
+        )
+
+    pred = model.predict(X_valid)
+    mae = mean_absolute_error(y_valid, pred)
+    accuracy = np.sum(np.abs(pred - y_valid) <= ERROR_TOLERANCE) / len(y_valid)
+
+    print(f"\nMAE: {mae:.4f} | Accuracy (±{ERROR_TOLERANCE}): {accuracy:.4f}")
+
+    joblib.dump(model, model_path)
+    joblib.dump(scaler, scaler_path)
+    with open(feature_names_path, "w", encoding="utf-8") as f:
+        for name in feature_names: f.write(name + "\n")
+
+    print(f"\nSaved to {working_dir}: {MODEL_FILENAME}, {SCALER_FILENAME}")
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--workingDir", required=True)
+    parser.add_argument("--dataDir", required=True)
+    parser.add_argument("--trainSize", required=True, type=int)
+    parser.add_argument("--validSize", required=True, type=int)
+    args = parser.parse_args()
+
+    TRAIN_SIZE, VALID_SIZE = args.trainSize, args.validSize
+    train_model(args.workingDir, args.dataDir)

train/train_xgboost.py

@@ -0,0 +1,369 @@
+import argparse
+import csv
+import json
+import math
+import os
+import random
+import joblib
+import numpy as np
+
+from xgboost import XGBRegressor
+from sklearn.preprocessing import StandardScaler
+from sklearn.metrics import mean_absolute_error
+
+# =========================================================
+# Hyper Parameters
+# =========================================================
+
+TRAIN_SIZE = 0
+VALID_SIZE = 0
+
+RANDOM_STATE = 42
+
+N_ESTIMATORS = 500
+MAX_DEPTH = 6
+LEARNING_RATE = 0.05
+SUBSAMPLE = 0.8
+COLSAMPLE_BYTREE = 0.8
+
+CONTINUE_TRAINING = True
+
+# 예측 성공으로 간주할 허용 오차
+ERROR_TOLERANCE = 0.1
+
+# =========================================================
+# 파일명
+# =========================================================
+
+FEATURES_FILENAME = "features.json"
+MEASURE_FILENAME = "measure.csv"
+
+MODEL_FILENAME = "model.pkl"
+SCALER_FILENAME = "scaler.pkl"
+FEATURE_NAMES_FILENAME = "features.txt"
+
+# =========================================================
+# 무시할 key
+# =========================================================
+
+IGNORE_KEYS = {
+    "songno",
+    "difficulty"
+}
+
+
+# =========================================================
+# safe float
+# =========================================================
+
+def safe_float(value):
+    if value is None:
+        return 0.0
+
+    x = float(value)
+
+    if not math.isfinite(x):
+        return 0.0
+
+    return x
+
+
+# =========================================================
+# train
+# =========================================================
+
+def train_model(
+    working_dir: str,
+    data_dir: str
+):
+    random.seed(RANDOM_STATE)
+
+    # =====================================================
+    # path
+    # =====================================================
+
+    features_path = os.path.join(
+        working_dir,
+        FEATURES_FILENAME
+    )
+
+    measure_path = os.path.join(
+        data_dir,
+        MEASURE_FILENAME
+    )
+
+    model_path = os.path.join(
+        working_dir,
+        MODEL_FILENAME
+    )
+
+    scaler_path = os.path.join(
+        working_dir,
+        SCALER_FILENAME
+    )
+
+    feature_names_path = os.path.join(
+        working_dir,
+        FEATURE_NAMES_FILENAME
+    )
+
+    # =====================================================
+    # features.json
+    # =====================================================
+
+    with open(features_path, "r", encoding="utf-8") as f:
+        feature_data = json.load(f)
+
+    if len(feature_data) == 0:
+        raise ValueError("features.json is empty")
+
+    # =====================================================
+    # feature map
+    # =====================================================
+
+    feature_map = {}
+
+    for item in feature_data:
+        key = (
+            str(item["songno"]),
+            str(item["difficulty"])
+        )
+
+        feature_map[key] = item
+
+    # =====================================================
+    # feature names
+    # =====================================================
+
+    feature_names = sorted([
+        k for k in feature_data[0].keys()
+        if k not in IGNORE_KEYS
+    ])
+
+    # =====================================================
+    # dataset build
+    # =====================================================
+
+    dataset = []
+
+    with open(measure_path, "r", encoding="utf-8") as f:
+        reader = csv.reader(f)
+
+        next(reader, None)
+
+        for row in reader:
+            if len(row) < 3:
+                continue
+
+            measure = safe_float(row[0])
+            songno = str(row[1])
+            diff = str(row[2])
+
+            key = (songno, diff)
+
+            if key not in feature_map:
+                print(
+                    f"[WARN] feature not found: "
+                    f"{songno} {diff}"
+                )
+                continue
+
+            feature_item = feature_map[key]
+
+            features = [
+                safe_float(feature_item.get(k, 0))
+                for k in feature_names
+            ]
+
+            dataset.append((
+                features,
+                measure
+            ))
+
+    # =====================================================
+    # shuffle
+    # =====================================================
+
+    random.shuffle(dataset)
+
+    required_size = TRAIN_SIZE + VALID_SIZE
+
+    if len(dataset) < required_size:
+        raise ValueError(
+            f"Not enough dataset "
+            f"({len(dataset)} < {required_size})"
+        )
+
+    # =====================================================
+    # split
+    # =====================================================
+
+    train_dataset = dataset[:TRAIN_SIZE]
+    valid_dataset = dataset[
+        TRAIN_SIZE:
+        TRAIN_SIZE + VALID_SIZE
+    ]
+
+    X_train = np.array(
+        [x for x, _ in train_dataset],
+        dtype=np.float32
+    )
+
+    y_train = np.array(
+        [y for _, y in train_dataset],
+        dtype=np.float32
+    )
+
+    X_valid = np.array(
+        [x for x, _ in valid_dataset],
+        dtype=np.float32
+    )
+
+    y_valid = np.array(
+        [y for _, y in valid_dataset],
+        dtype=np.float32
+    )
+
+    print(f"Train Size: {len(X_train)}")
+    print(f"Valid Size: {len(X_valid)}")
+    print(f"Feature Count: {len(feature_names)}")
+
+    # =====================================================
+    # scaler
+    # =====================================================
+
+    if CONTINUE_TRAINING and os.path.exists(scaler_path):
+        print("Loading existing scaler...")
+
+        scaler = joblib.load(scaler_path)
+
+    else:
+        print("Creating new scaler...")
+
+        scaler = StandardScaler()
+        scaler.fit(X_train)
+
+    X_train = scaler.transform(X_train)
+    X_valid = scaler.transform(X_valid)
+
+    # =====================================================
+    # model
+    # =====================================================
+
+    if CONTINUE_TRAINING and os.path.exists(model_path):
+        print("Loading existing model...")
+
+        model = joblib.load(model_path)
+
+        previous_booster = model.get_booster()
+
+        model.fit(
+            X_train,
+            y_train,
+            xgb_model=previous_booster
+        )
+
+    else:
+        print("Creating new model...")
+
+        model = XGBRegressor(
+            n_estimators=N_ESTIMATORS,
+            max_depth=MAX_DEPTH,
+            learning_rate=LEARNING_RATE,
+            subsample=SUBSAMPLE,
+            colsample_bytree=COLSAMPLE_BYTREE,
+            objective="reg:squarederror",
+            random_state=RANDOM_STATE
+        )
+
+        model.fit(X_train, y_train)
+
+    # =====================================================
+    # evaluate
+    # =====================================================
+
+    pred = model.predict(X_valid)
+
+    mae = mean_absolute_error(y_valid, pred)
+
+    correct = np.sum(
+        np.abs(pred - y_valid) <= ERROR_TOLERANCE
+    )
+
+    accuracy = correct / len(y_valid)
+
+    print(f"\nMAE: {mae:.4f}")
+
+    print(
+        f"Accuracy "
+        f"(±{ERROR_TOLERANCE}): "
+        f"{accuracy:.4f}"
+    )
+
+    # =====================================================
+    # feature importance
+    # =====================================================
+
+    print("\nFeature Importance:")
+
+    importance = model.feature_importances_
+
+    pairs = list(zip(feature_names, importance))
+    pairs.sort(key=lambda x: x[1], reverse=True)
+
+    for name, score in pairs:
+        print(f"{name:25} {score:.6f}")
+
+    # =====================================================
+    # save
+    # =====================================================
+
+    joblib.dump(model, model_path)
+    joblib.dump(scaler, scaler_path)
+
+    with open(feature_names_path, "w", encoding="utf-8") as f:
+        for name in feature_names:
+            f.write(name + "\n")
+
+    print("\nSaved:")
+    print(model_path)
+    print(scaler_path)
+    print(feature_names_path)
+
+
+# =========================================================
+# main
+# =========================================================
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+
+    parser.add_argument(
+        "--workingDir",
+        required=True
+    )
+
+    parser.add_argument(
+        "--dataDir",
+        required=True
+    )
+    
+    parser.add_argument(
+        "--trainSize",
+        required=True
+    )
+    
+    parser.add_argument(
+        "--validSize",
+        required=True
+    )
+
+    args = parser.parse_args()
+    
+    TRAIN_SIZE = int(args.trainSize)
+    VALID_SIZE = int(args.validSize)
+
+    train_model(
+        args.workingDir,
+        args.dataDir
+    )