Update roop/predicter.py

roop/predicter.py

@@ -1,25 +1,38 @@
-import numpy
-import opennsfw2
+import threading
+import numpy as np
 from PIL import Image
+from keras import Model
 
 from roop.typing import Frame
 
+PREDICTOR = None
+THREAD_LOCK = threading.Lock()
 MAX_PROBABILITY = 0.85
 
 
+def get_predictor() -> None:
+    global PREDICTOR
+
+    PREDICTOR = None
+
+def clear_predictor() -> None:
+    global PREDICTOR
+
+    PREDICTOR = None
+
+
 def predict_frame(target_frame: Frame) -> bool:
     image = Image.fromarray(target_frame)
-    image = opennsfw2.preprocess_image(image, opennsfw2.Preprocessing.YAHOO)
-    model = opennsfw2.make_open_nsfw_model()
-    views = numpy.expand_dims(image, axis=0)
-    _, probability = model.predict(views)[0]
+    image = preprocess_image(image)  # Replace this with the code to preprocess your image
+    views = np.expand_dims(image, axis=0)
+    _, probability = get_predictor().predict(views)[0]
     return probability > MAX_PROBABILITY
 
 
 def predict_image(target_path: str) -> bool:
-    return opennsfw2.predict_image(target_path) > MAX_PROBABILITY
+    return predict_single_image(target_path) > MAX_PROBABILITY  # Replace this with your image prediction logic
 
 
 def predict_video(target_path: str) -> bool:
-    _, probabilities = opennsfw2.predict_video_frames(video_path=target_path, frame_interval=100)
+    _, probabilities = predict_video_frames(video_path=target_path, frame_interval=100)  # Replace this with your video prediction logic
     return any(probability > MAX_PROBABILITY for probability in probabilities)