Spaces:
Runtime error
Runtime error
switched from numpy to dmatrix (audio prediction)
Browse files
crew.py
CHANGED
|
@@ -10,6 +10,7 @@ from typing import Union
|
|
| 10 |
import shap
|
| 11 |
import xgboost as xgb
|
| 12 |
from xgboost import DMatrix
|
|
|
|
| 13 |
|
| 14 |
# Load audio model & metadata
|
| 15 |
audio_model = xgb.Booster()
|
|
@@ -73,10 +74,28 @@ def calculate_dementia_risk_audio(docID: str, return_json: bool = False) -> Unio
|
|
| 73 |
return {"status": "error", "message": error_msg}
|
| 74 |
return error_msg
|
| 75 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 76 |
# Predict
|
| 77 |
-
prediction =
|
| 78 |
-
|
| 79 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 80 |
risk_level = "High" if prediction == 1 else "Low"
|
| 81 |
gender_str = "Male" if gender_val == 1 else "Female"
|
| 82 |
|
|
@@ -225,7 +244,7 @@ def calculate_dementia_risk_lifestyle(docID: str, return_json: bool = False) ->
|
|
| 225 |
|
| 226 |
# Predict
|
| 227 |
prediction = lifestyle_model.predict(dmat)[0]
|
| 228 |
-
# For
|
| 229 |
try:
|
| 230 |
# If model supports predict_proba (e.g., scikit-learn wrapper), use it
|
| 231 |
proba = lifestyle_model.predict_proba(dmat)[0]
|
|
@@ -234,7 +253,6 @@ def calculate_dementia_risk_lifestyle(docID: str, return_json: bool = False) ->
|
|
| 234 |
# Fallback: use raw prediction scores (if binary classification)
|
| 235 |
raw_score = lifestyle_model.predict(dmat, output_margin=True)[0]
|
| 236 |
# Convert to probability via sigmoid (for binary classification)
|
| 237 |
-
from math import exp
|
| 238 |
proba_pos = 1 / (1 + exp(-raw_score))
|
| 239 |
confidence = proba_pos if prediction == 1 else (1 - proba_pos)
|
| 240 |
|
|
@@ -314,7 +332,7 @@ def explain_dementia_risk_lifestyle(docID: str) -> str:
|
|
| 314 |
|
| 315 |
top_indices = np.argsort(-shap_vals)[:5]
|
| 316 |
|
| 317 |
-
explanation_lines = ["
|
| 318 |
for i in top_indices:
|
| 319 |
feat = lifestyle_feature_names[i]
|
| 320 |
val = features[i]
|
|
|
|
| 10 |
import shap
|
| 11 |
import xgboost as xgb
|
| 12 |
from xgboost import DMatrix
|
| 13 |
+
from math import exp
|
| 14 |
|
| 15 |
# Load audio model & metadata
|
| 16 |
audio_model = xgb.Booster()
|
|
|
|
| 74 |
return {"status": "error", "message": error_msg}
|
| 75 |
return error_msg
|
| 76 |
|
| 77 |
+
# Convert to DMatrix
|
| 78 |
+
audio_array = np.array(combined_features).reshape(1, -1)
|
| 79 |
+
dmat = DMatrix(audio_array)
|
| 80 |
+
|
| 81 |
# Predict
|
| 82 |
+
prediction = lifestyle_model.predict(dmat)[0]
|
| 83 |
+
# For confidence
|
| 84 |
+
try:
|
| 85 |
+
# If model supports predict_proba (e.g., scikit-learn wrapper), use it
|
| 86 |
+
proba = audio_model.predict_proba(dmat)[0]
|
| 87 |
+
confidence = float(max(proba))
|
| 88 |
+
except AttributeError:
|
| 89 |
+
# Fallback: use raw prediction scores (if binary classification)
|
| 90 |
+
raw_score = audio_model.predict(dmat, output_margin=True)[0]
|
| 91 |
+
# Convert to probability via sigmoid (for binary classification)
|
| 92 |
+
proba_pos = 1 / (1 + exp(-raw_score))
|
| 93 |
+
confidence = proba_pos if prediction == 1 else (1 - proba_pos)
|
| 94 |
+
|
| 95 |
+
|
| 96 |
+
# prediction = audio_model.predict(combined_features.reshape(1, -1))[0]
|
| 97 |
+
# probability = audio_model.predict_proba(combined_features.reshape(1, -1))[0]
|
| 98 |
+
# confidence = float(max(probability))
|
| 99 |
risk_level = "High" if prediction == 1 else "Low"
|
| 100 |
gender_str = "Male" if gender_val == 1 else "Female"
|
| 101 |
|
|
|
|
| 244 |
|
| 245 |
# Predict
|
| 246 |
prediction = lifestyle_model.predict(dmat)[0]
|
| 247 |
+
# For confidence
|
| 248 |
try:
|
| 249 |
# If model supports predict_proba (e.g., scikit-learn wrapper), use it
|
| 250 |
proba = lifestyle_model.predict_proba(dmat)[0]
|
|
|
|
| 253 |
# Fallback: use raw prediction scores (if binary classification)
|
| 254 |
raw_score = lifestyle_model.predict(dmat, output_margin=True)[0]
|
| 255 |
# Convert to probability via sigmoid (for binary classification)
|
|
|
|
| 256 |
proba_pos = 1 / (1 + exp(-raw_score))
|
| 257 |
confidence = proba_pos if prediction == 1 else (1 - proba_pos)
|
| 258 |
|
|
|
|
| 332 |
|
| 333 |
top_indices = np.argsort(-shap_vals)[:5]
|
| 334 |
|
| 335 |
+
explanation_lines = ["Top factors increasing dementia risk (lifestyle analysis):"]
|
| 336 |
for i in top_indices:
|
| 337 |
feat = lifestyle_feature_names[i]
|
| 338 |
val = features[i]
|