change dataset config

README.md

@@ -9,3 +9,23 @@ app_file: start.py
 pinned: false
 license: apache-2.0
 ---
+
+# MuscleCare Dataset API
+
+이 Space는 수집된 운동 피로도 데이터를 Hugging Face Dataset 저장소에 업로드하는 FastAPI 서비스를 제공합니다.
+
+## 엔드포인트
+- `GET /health` : 서버 상태 확인
+- `POST /upload_dataset` : `fatigue_dataset` 스키마에 맞춘 배치 데이터를 Hugging Face Hub로 저장
+
+## 실행 방법
+```bash
+./start_server.sh
+```
+
+## 테스트
+```bash
+./test_api.sh
+```
+
+테스트 스크립트는 샘플 배치 데이터를 `/upload_dataset` 엔드포인트로 전송합니다. 실행 전 `HF_DATA_REPO_ID`와 `HF_DATA_TOKEN` 환경변수가 설정되어 있어야 합니다.

app.py

@@ -18,7 +18,8 @@ app = FastAPI(title="MuscleCare FastAPI Server")
 
 # ----- 모델 -----
 class DatasetItem(BaseModel):
-    user_id: int
+    id: Optional[int] = None
+    user_id: str
     session_id: Optional[str] = None
     window_id: int
     window_start_ms: int
@@ -112,6 +113,7 @@ async def upload_dataset(payload: DatasetBatchPayload):
 
         # 새 스키마 정의
         target_cols = [
+            "id",
             "user_id",
             "session_id",
             "window_id",
@@ -199,7 +201,7 @@ async def upload_dataset(payload: DatasetBatchPayload):
             rec = item.model_dump()
             if not rec.get("timestamp_utc"):
                 rec["timestamp_utc"] = datetime.now().isoformat()
-            user_groups.setdefault(str(item.user_id), []).append(rec)
+            user_groups.setdefault(item.user_id, []).append(rec)
 
         results = {}
 

backup/2025-10-23/new_user.json

@@ -1 +0,0 @@
-[{"user_id": "new_user", "rms": 0.95, "freq": 60.1, "fatigue": 0.88, "timestamp": "2025-10-23T15:02:00.000000"}]

backup/2025-10-23/test_user_1.json

@@ -1 +0,0 @@
-[{"user_id": "test_user_1", "rms": 0.85, "freq": 52.3, "fatigue": 0.72, "timestamp": "2025-10-23T14:55:00.000000"}]

backup/2025-10-23/test_user_2.json

@@ -1 +0,0 @@
-[{"user_id": "test_user_2", "rms": 0.91, "freq": 58.7, "fatigue": 0.81, "timestamp": "2025-10-23T14:55:30.000000"}]

test_api.sh

@@ -1,4 +1,4 @@
-ㅔㅛ#!/bin/bash
+#!/bin/bash
 
 # MuscleCare API 테스트 스크립트
 
@@ -8,7 +8,7 @@
 BASE_URL="${1:-http://localhost:7860}"
 
 echo "======================================"
-echo "🧪 MuscleCare API 테스트"
+echo "🧪 MuscleCare Dataset API 테스트"
 echo "======================================"
 echo "서버: $BASE_URL"
 echo ""
@@ -21,64 +21,47 @@ curl -X GET "$BASE_URL/health" \
 echo ""
 echo ""
 
-# 2. 사용자 상태 업로드
-echo "2️⃣ 사용자 상태 업로드"
+# 2. 배치 데이터 업로드
+echo "2️⃣ 배치 데이터 업로드"
 echo "--------------------------------------"
-curl -X POST "$BASE_URL/upload_state" \
+curl -X POST "$BASE_URL/upload_dataset" \
   -H "Content-Type: application/json" \
   -H "accept: application/json" \
   -d '{
-    "user_id": "local_user",
-    "rms_base": 0.11,
-    "freq_base": 50.1,
-    "user_emb": [0.135, 0.012, 49.3, 0.8, 0.84, 0.02, 0.11, 0.0021, 0.35, 0.71, 0.92, 0.88],
-    "model_version": "v1.0.0"
-  }' | jq .
-echo ""
-echo ""
-
-# 3. 로그 데이터 업로드 (새로운 개별 API)
-echo "3️⃣ 로그 데이터 업로드"
-echo "--------------------------------------"
-curl -X POST "$BASE_URL/upload_logs" \
-  -H "Content-Type: application/json" \
-  -H "accept: application/json" \
-  -d '{
-    "user_id": "test_user_001",
-    "session_id": "session_123",
-    "measure_date": "2025-10-24",
-    "rms": 0.42,
-    "freq": 45.3,
-    "fatigue": 0.25,
-    "rms_base": 0.40,
-    "freq_base": 44.8,
-    "user_emb": [0.2, 0.05, 45.1, 0.7, 0.8, 0.03, 0.12, 0.0015, 0.4, 0.75, 0.9, 0.85],
-    "mode": "training",
-    "window_count": 5,
-    "measurement_count": 1
-  }' | jq .
-echo ""
-echo ""
-
-# 4. 추가 로그 데이터 업로드
-echo "4️⃣ 추가 로그 데이터 업로드"
-echo "--------------------------------------"
-curl -X POST "$BASE_URL/upload_logs" \
-  -H "Content-Type: application/json" \
-  -H "accept: application/json" \
-  -d '{
-    "user_id": "test_user_001",
-    "session_id": "session_124",
-    "measure_date": "2025-10-24",
-    "rms": 0.38,
-    "freq": 46.1,
-    "fatigue": 0.32,
-    "rms_base": 0.40,
-    "freq_base": 44.8,
-    "user_emb": [0.22, 0.06, 45.5, 0.72, 0.82, 0.035, 0.13, 0.0018, 0.42, 0.77, 0.92, 0.87],
-    "mode": "testing",
-    "window_count": 7,
-    "measurement_count": 1
+    "batch_data": [
+      {
+        "user_id": "user_001",
+        "session_id": "session_001",
+        "window_id": 1001,
+        "window_start_ms": 0,
+        "window_end_ms": 2000,
+        "timestamp_utc": "2025-10-24T10:00:00Z",
+        "acc_x_mean": 0.01,
+        "acc_y_mean": -0.02,
+        "acc_z_mean": 0.98,
+        "gyro_x_mean": 0.001,
+        "gyro_y_mean": -0.003,
+        "gyro_z_mean": 0.002,
+        "rms_acc": 0.45,
+        "rms_gyro": 0.12,
+        "mean_freq_acc": 45.2,
+        "mean_freq_gyro": 12.5,
+        "entropy_acc": 0.62,
+        "entropy_gyro": 0.31,
+        "jerk_mean": 0.08,
+        "jerk_std": 0.01,
+        "stability_index": 0.9,
+        "rms_base": 0.4,
+        "freq_base": 40.0,
+        "user_emb": [0.1, 0.05, 0.02, 0.07, 0.11, 0.03, 0.04, 0.02, 0.09, 0.01, 0.08, 0.05],
+        "fatigue_prev": 0.3,
+        "fatigue": 0.32,
+        "fatigue_level": 1,
+        "quality_flag": 1,
+        "window_size_ms": 2000,
+        "overlap_rate": 0.5
+      }
+    ]
   }' | jq .
 echo ""
 echo ""