Spaces:
Sleeping
Sleeping
Update main.py
Browse files
main.py
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
# ======================================================================
|
| 2 |
-
# --- main.py (Modular, dengan builder eksternal) ---
|
| 3 |
# ======================================================================
|
| 4 |
|
| 5 |
import os
|
|
@@ -11,9 +11,8 @@ from fastapi import FastAPI, HTTPException
|
|
| 11 |
from pydantic import BaseModel
|
| 12 |
from typing import List, Dict, Any
|
| 13 |
|
| 14 |
-
|
| 15 |
from .explanation_builder import build_explanation_from_rules
|
| 16 |
-
# ---------------------
|
| 17 |
|
| 18 |
|
| 19 |
# ======================================================================
|
|
@@ -22,14 +21,13 @@ from .explanation_builder import build_explanation_from_rules
|
|
| 22 |
app = FastAPI(
|
| 23 |
title="GCOMPRO",
|
| 24 |
description="API Prediksi Risiko Akademik dan Rekomendasi Mata Kuliah.",
|
| 25 |
-
version="1.
|
| 26 |
)
|
| 27 |
|
| 28 |
# ======================================================================
|
| 29 |
# 2. Struktur Data Input/Output (Pydantic)
|
| 30 |
# ======================================================================
|
| 31 |
|
| 32 |
-
|
| 33 |
# --- Model untuk API Prediksi Risiko (App 1) ---
|
| 34 |
class StudentFeatures(BaseModel):
|
| 35 |
IPK_Terakhir: float
|
|
@@ -45,6 +43,19 @@ class StudentFeatures(BaseModel):
|
|
| 45 |
IPK_Ternormalisasi_SKS: float
|
| 46 |
Profil_Tren: str
|
| 47 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 48 |
# --- Model untuk API Rekomendasi MK (App 2) ---
|
| 49 |
class RecommendationRequest(BaseModel):
|
| 50 |
current_semester: int
|
|
@@ -84,7 +95,7 @@ course_details_map = {}
|
|
| 84 |
prereq_map = {}
|
| 85 |
out_degree_map = {}
|
| 86 |
|
| 87 |
-
|
| 88 |
|
| 89 |
def load_ml_model():
|
| 90 |
"""Memuat model ML dari file .skops"""
|
|
@@ -148,7 +159,6 @@ def on_startup():
|
|
| 148 |
# 4. Helper Function (Untuk API Rekomendasi)
|
| 149 |
# ======================================================================
|
| 150 |
|
| 151 |
-
|
| 152 |
def get_recommendations_logic(current_semester: int, courses_passed_list: List[str]) -> List[Dict[str, Any]]:
|
| 153 |
"""Inti dari logika rekomendasi, HANYA dengan validasi prereq."""
|
| 154 |
passed_set = set(courses_passed_list)
|
|
@@ -186,17 +196,15 @@ def get_recommendations_logic(current_semester: int, courses_passed_list: List[s
|
|
| 186 |
# 5. Endpoints API
|
| 187 |
# ======================================================================
|
| 188 |
|
| 189 |
-
|
| 190 |
-
|
| 191 |
@app.get("/")
|
| 192 |
def read_root():
|
| 193 |
return {
|
| 194 |
-
"message": "Selamat Datang di API Layanan Akademik Mahasiswa
|
| 195 |
"status": "ready",
|
| 196 |
"endpoints": ["/predict/", "/recommend/"]
|
| 197 |
}
|
| 198 |
|
| 199 |
-
@app.post("/predict/")
|
| 200 |
def predict_risk(student_data: StudentFeatures):
|
| 201 |
if ml_model is None:
|
| 202 |
raise HTTPException(status_code=503, detail="Model ML belum siap. Silakan coba lagi nanti.")
|
|
@@ -213,7 +221,7 @@ def predict_risk(student_data: StudentFeatures):
|
|
| 213 |
classes = ml_model.classes_
|
| 214 |
probabilities = dict(zip(classes, prediction_proba[0]))
|
| 215 |
|
| 216 |
-
|
| 217 |
|
| 218 |
if hasattr(ml_model, 'tree_'):
|
| 219 |
try:
|
|
@@ -223,22 +231,16 @@ def predict_risk(student_data: StudentFeatures):
|
|
| 223 |
path = ml_model.decision_path(input_encoded)
|
| 224 |
node_indices = path.indices[path.indptr[0]:path.indptr[1]]
|
| 225 |
|
| 226 |
-
structured_rules = []
|
| 227 |
|
| 228 |
for node_id in node_indices[:-1]:
|
| 229 |
feature_index = tree.feature[node_id]
|
| 230 |
feature_name = feature_names[feature_index]
|
| 231 |
threshold = tree.threshold[node_id]
|
| 232 |
-
|
| 233 |
sample_value = input_encoded.iloc[0, feature_index]
|
| 234 |
|
| 235 |
-
condition_str = ""
|
| 236 |
-
if sample_value <= threshold:
|
| 237 |
-
condition_str = "rendah" # (<=)
|
| 238 |
-
else:
|
| 239 |
-
condition_str = "tinggi" # (>)
|
| 240 |
|
| 241 |
-
# Simpan aturan dalam format terstruktur
|
| 242 |
structured_rules.append({
|
| 243 |
"feature": feature_name,
|
| 244 |
"condition": condition_str,
|
|
@@ -246,20 +248,27 @@ def predict_risk(student_data: StudentFeatures):
|
|
| 246 |
"value": sample_value
|
| 247 |
})
|
| 248 |
|
| 249 |
-
# Panggil
|
| 250 |
-
|
| 251 |
-
# ---------------------
|
| 252 |
|
| 253 |
except Exception as e:
|
| 254 |
-
|
|
|
|
|
|
|
|
|
|
| 255 |
else:
|
| 256 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 257 |
|
| 258 |
-
return {
|
| 259 |
-
"prediction": prediction_val,
|
| 260 |
-
"probabilities": probabilities,
|
| 261 |
-
"explanation": explanation_str
|
| 262 |
-
}
|
| 263 |
except Exception as e:
|
| 264 |
raise HTTPException(status_code=500, detail=f"Terjadi kesalahan saat prediksi: {e}")
|
| 265 |
|
|
@@ -311,3 +320,4 @@ async def recommend_courses(request: RecommendationRequest):
|
|
| 311 |
)
|
| 312 |
|
| 313 |
return response_output
|
|
|
|
|
|
| 1 |
# ======================================================================
|
| 2 |
+
# --- main.py (Modular, dengan builder eksternal & format respons baru) ---
|
| 3 |
# ======================================================================
|
| 4 |
|
| 5 |
import os
|
|
|
|
| 11 |
from pydantic import BaseModel
|
| 12 |
from typing import List, Dict, Any
|
| 13 |
|
| 14 |
+
# Impor fungsi builder dari file baru
|
| 15 |
from .explanation_builder import build_explanation_from_rules
|
|
|
|
| 16 |
|
| 17 |
|
| 18 |
# ======================================================================
|
|
|
|
| 21 |
app = FastAPI(
|
| 22 |
title="GCOMPRO",
|
| 23 |
description="API Prediksi Risiko Akademik dan Rekomendasi Mata Kuliah.",
|
| 24 |
+
version="1.5.0"
|
| 25 |
)
|
| 26 |
|
| 27 |
# ======================================================================
|
| 28 |
# 2. Struktur Data Input/Output (Pydantic)
|
| 29 |
# ======================================================================
|
| 30 |
|
|
|
|
| 31 |
# --- Model untuk API Prediksi Risiko (App 1) ---
|
| 32 |
class StudentFeatures(BaseModel):
|
| 33 |
IPK_Terakhir: float
|
|
|
|
| 43 |
IPK_Ternormalisasi_SKS: float
|
| 44 |
Profil_Tren: str
|
| 45 |
|
| 46 |
+
class PredictionExplanation(BaseModel):
|
| 47 |
+
""" Model Pydantic untuk struktur penjelasan baru """
|
| 48 |
+
opening_line: str
|
| 49 |
+
factors: List[str]
|
| 50 |
+
|
| 51 |
+
class PredictionResponse(BaseModel):
|
| 52 |
+
""" Model Pydantic untuk seluruh respons /predict/ """
|
| 53 |
+
prediction: str
|
| 54 |
+
probabilities: Dict[str, float]
|
| 55 |
+
explanation: PredictionExplanation
|
| 56 |
+
# --------------------------------------------------------
|
| 57 |
+
|
| 58 |
+
|
| 59 |
# --- Model untuk API Rekomendasi MK (App 2) ---
|
| 60 |
class RecommendationRequest(BaseModel):
|
| 61 |
current_semester: int
|
|
|
|
| 95 |
prereq_map = {}
|
| 96 |
out_degree_map = {}
|
| 97 |
|
| 98 |
+
# --- Fungsi Pemuatan (dipanggil saat startup) ---
|
| 99 |
|
| 100 |
def load_ml_model():
|
| 101 |
"""Memuat model ML dari file .skops"""
|
|
|
|
| 159 |
# 4. Helper Function (Untuk API Rekomendasi)
|
| 160 |
# ======================================================================
|
| 161 |
|
|
|
|
| 162 |
def get_recommendations_logic(current_semester: int, courses_passed_list: List[str]) -> List[Dict[str, Any]]:
|
| 163 |
"""Inti dari logika rekomendasi, HANYA dengan validasi prereq."""
|
| 164 |
passed_set = set(courses_passed_list)
|
|
|
|
| 196 |
# 5. Endpoints API
|
| 197 |
# ======================================================================
|
| 198 |
|
|
|
|
|
|
|
| 199 |
@app.get("/")
|
| 200 |
def read_root():
|
| 201 |
return {
|
| 202 |
+
"message": "Selamat Datang di API Layanan Akademik Mahasiswa",
|
| 203 |
"status": "ready",
|
| 204 |
"endpoints": ["/predict/", "/recommend/"]
|
| 205 |
}
|
| 206 |
|
| 207 |
+
@app.post("/predict/", response_model=PredictionResponse)
|
| 208 |
def predict_risk(student_data: StudentFeatures):
|
| 209 |
if ml_model is None:
|
| 210 |
raise HTTPException(status_code=503, detail="Model ML belum siap. Silakan coba lagi nanti.")
|
|
|
|
| 221 |
classes = ml_model.classes_
|
| 222 |
probabilities = dict(zip(classes, prediction_proba[0]))
|
| 223 |
|
| 224 |
+
explanation_obj = {} # Ganti nama dari explanation_str
|
| 225 |
|
| 226 |
if hasattr(ml_model, 'tree_'):
|
| 227 |
try:
|
|
|
|
| 231 |
path = ml_model.decision_path(input_encoded)
|
| 232 |
node_indices = path.indices[path.indptr[0]:path.indptr[1]]
|
| 233 |
|
| 234 |
+
structured_rules = []
|
| 235 |
|
| 236 |
for node_id in node_indices[:-1]:
|
| 237 |
feature_index = tree.feature[node_id]
|
| 238 |
feature_name = feature_names[feature_index]
|
| 239 |
threshold = tree.threshold[node_id]
|
|
|
|
| 240 |
sample_value = input_encoded.iloc[0, feature_index]
|
| 241 |
|
| 242 |
+
condition_str = "rendah" if sample_value <= threshold else "tinggi"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 243 |
|
|
|
|
| 244 |
structured_rules.append({
|
| 245 |
"feature": feature_name,
|
| 246 |
"condition": condition_str,
|
|
|
|
| 248 |
"value": sample_value
|
| 249 |
})
|
| 250 |
|
| 251 |
+
# Panggil builder, yang sekarang mengembalikan DICT
|
| 252 |
+
explanation_obj = build_explanation_from_rules(structured_rules, prediction_val)
|
|
|
|
| 253 |
|
| 254 |
except Exception as e:
|
| 255 |
+
explanation_obj = {
|
| 256 |
+
"opening_line": f"Gagal membuat penjelasan: {str(e)}",
|
| 257 |
+
"factors": []
|
| 258 |
+
}
|
| 259 |
else:
|
| 260 |
+
explanation_obj = {
|
| 261 |
+
"opening_line": "Penjelasan (decision path) tidak tersedia untuk tipe model ini.",
|
| 262 |
+
"factors": []
|
| 263 |
+
}
|
| 264 |
+
|
| 265 |
+
# Kembalikan objek Pydantic yang sudah divalidasi
|
| 266 |
+
return PredictionResponse(
|
| 267 |
+
prediction=prediction_val,
|
| 268 |
+
probabilities=probabilities,
|
| 269 |
+
explanation=explanation_obj
|
| 270 |
+
)
|
| 271 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 272 |
except Exception as e:
|
| 273 |
raise HTTPException(status_code=500, detail=f"Terjadi kesalahan saat prediksi: {e}")
|
| 274 |
|
|
|
|
| 320 |
)
|
| 321 |
|
| 322 |
return response_output
|
| 323 |
+
|