Update app.py

app.py

@@ -1,134 +1,134 @@
-import gradio as gr
-import cv2
-import time
-from ultralytics import YOLO
-import numpy as np
-
-# Model yükle
-model = YOLO("runs/detect/train2/weights/best.pt")
-device = 0  # GPU cihazı (0 = ilk GPU)
-
-def infer_fire(image, conf_thresh):
-    # RGB→BGR çevir, infer
-    img_bgr = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)
-    start = time.time()
-    res = model.predict(
-        source=img_bgr,
-        device=device,
-        imgsz=640,
-        conf=conf_thresh
-    )[0]
-    elapsed = (time.time() - start) * 1000  # ms cinsinden
-
-    annotated = img_bgr.copy()
-    h, w = annotated.shape[:2]
-    boxes_np = (
-        res.boxes.data.cpu().numpy()
-        if hasattr(res.boxes, "data")
-        else np.empty((0, 6))
-    )
-
-    for x1, y1, x2, y2, conf, cls in boxes_np:
-        x1, y1, x2, y2 = map(int, (x1, y1, x2, y2))
-        # 1) Box'u resim sınırları içinde tut
-        x1 = max(0, x1);   y1 = max(0, y1)
-        x2 = min(w-1, x2); y2 = min(h-1, y2)
-
-        label = f"{model.names[int(cls)]} {conf:.2f}"
-
-        # 2) Kalın box çiz
-        cv2.rectangle(annotated, (x1, y1), (x2, y2), (0, 255, 0), 4)
-
-        # 3) Yazı boyutunu al
-        (text_w, text_h), baseline = cv2.getTextSize(
-            label, cv2.FONT_HERSHEY_SIMPLEX, 0.7, 3
-        )
-
-        # 4) Etiket pozisyonunu ayarla
-        text_x = x1
-        text_y = y1 - 5
-        # Üst sınırı aşmasın
-        if text_y - text_h - baseline < 0:
-            text_y = y1 + text_h + 5
-        # Sağ sınırı aşmasın
-        if text_x + text_w > w:
-            text_x = w - text_w - 5
-
-        # 5) Dolgu arkaplanlı dikdörtgen
-        cv2.rectangle(
-            annotated,
-            (text_x, text_y - text_h - baseline),
-            (text_x + text_w, text_y + baseline),
-            (0, 255, 0),
-            cv2.FILLED
-        )
-
-        # 6) Siyah renkli yazı
-        cv2.putText(
-            annotated,
-            label,
-            (text_x, text_y),
-            cv2.FONT_HERSHEY_SIMPLEX,
-            0.7,
-            (0, 0, 0),
-            3
-        )
-
-    out_img = cv2.cvtColor(annotated, cv2.COLOR_BGR2RGB)
-    return out_img, f"Inference time: {elapsed:.1f} ms"
-
-# Gradio arayüzü
-examples = [
-    ["datasets/test/images/WEB10432.jpg", 0.25],
-    ["datasets/test/images/WEB11791.jpg", 0.25],
-    ["datasets/test/images/WEB11706.jpg", 0.25],
-]
-
-with gr.Blocks() as demo:
-    gr.Markdown("## 🔥 Wildfire Smoke & Fire Detector")
-    gr.Markdown(
-        "Upload an image below, adjust the confidence threshold, "
-        "and the model will highlight any smoke or fire regions."
-    )
-
-    with gr.Row():
-        with gr.Column(scale=1, min_width=200):
-            input_img   = gr.Image(type="numpy", label="Input Image")
-            conf_slider = gr.Slider(0.0, 1.0, 0.25, 0.01, label="Confidence Threshold")
-            run_btn     = gr.Button("Detect 🔍", variant="primary")
-        with gr.Column(scale=5, min_width=1000):
-            output_img = gr.Image(
-                type="numpy",
-                label="Annotated Output",
-                height=800,
-                width=1000
-            )
-            time_txt   = gr.Textbox(label="Performance", interactive=False)
-
-    run_btn.click(
-        fn=infer_fire,
-        inputs=[input_img, conf_slider],
-        outputs=[output_img, time_txt]
-    )
-
-    gr.Examples(
-        examples=examples,
-        inputs=[input_img, conf_slider],
-        outputs=[output_img, time_txt],
-        fn=infer_fire,
-        cache_examples=False
-    )
-
-    gr.Markdown(
-        "---\n"
-        "Model trained on a custom wildfire dataset using YOLOv8. "
-        "Adjust the threshold to trade off between false positives and false negatives."
-    )
-
-if __name__ == "__main__":
-    demo.launch()
-       # server_name="0.0.0.0",
-       # server_port=7861,
-       # share=False,
-       # inbrowser=True
-   
+import gradio as gr
+import cv2
+import time
+from ultralytics import YOLO
+import numpy as np
+
+# Model yükle
+model = YOLO("best.pt")
+device = 0  # GPU cihazı (0 = ilk GPU)
+
+def infer_fire(image, conf_thresh):
+    # RGB→BGR çevir, infer
+    img_bgr = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)
+    start = time.time()
+    res = model.predict(
+        source=img_bgr,
+        device=device,
+        imgsz=640,
+        conf=conf_thresh
+    )[0]
+    elapsed = (time.time() - start) * 1000  # ms cinsinden
+
+    annotated = img_bgr.copy()
+    h, w = annotated.shape[:2]
+    boxes_np = (
+        res.boxes.data.cpu().numpy()
+        if hasattr(res.boxes, "data")
+        else np.empty((0, 6))
+    )
+
+    for x1, y1, x2, y2, conf, cls in boxes_np:
+        x1, y1, x2, y2 = map(int, (x1, y1, x2, y2))
+        # 1) Box'u resim sınırları içinde tut
+        x1 = max(0, x1);   y1 = max(0, y1)
+        x2 = min(w-1, x2); y2 = min(h-1, y2)
+
+        label = f"{model.names[int(cls)]} {conf:.2f}"
+
+        # 2) Kalın box çiz
+        cv2.rectangle(annotated, (x1, y1), (x2, y2), (0, 255, 0), 4)
+
+        # 3) Yazı boyutunu al
+        (text_w, text_h), baseline = cv2.getTextSize(
+            label, cv2.FONT_HERSHEY_SIMPLEX, 0.7, 3
+        )
+
+        # 4) Etiket pozisyonunu ayarla
+        text_x = x1
+        text_y = y1 - 5
+        # Üst sınırı aşmasın
+        if text_y - text_h - baseline < 0:
+            text_y = y1 + text_h + 5
+        # Sağ sınırı aşmasın
+        if text_x + text_w > w:
+            text_x = w - text_w - 5
+
+        # 5) Dolgu arkaplanlı dikdörtgen
+        cv2.rectangle(
+            annotated,
+            (text_x, text_y - text_h - baseline),
+            (text_x + text_w, text_y + baseline),
+            (0, 255, 0),
+            cv2.FILLED
+        )
+
+        # 6) Siyah renkli yazı
+        cv2.putText(
+            annotated,
+            label,
+            (text_x, text_y),
+            cv2.FONT_HERSHEY_SIMPLEX,
+            0.7,
+            (0, 0, 0),
+            3
+        )
+
+    out_img = cv2.cvtColor(annotated, cv2.COLOR_BGR2RGB)
+    return out_img, f"Inference time: {elapsed:.1f} ms"
+
+# Gradio arayüzü
+examples = [
+    ["datasets/test/images/WEB10432.jpg", 0.25],
+    ["datasets/test/images/WEB11791.jpg", 0.25],
+    ["datasets/test/images/WEB11706.jpg", 0.25],
+]
+
+with gr.Blocks() as demo:
+    gr.Markdown("## 🔥 Wildfire Smoke & Fire Detector")
+    gr.Markdown(
+        "Upload an image below, adjust the confidence threshold, "
+        "and the model will highlight any smoke or fire regions."
+    )
+
+    with gr.Row():
+        with gr.Column(scale=1, min_width=200):
+            input_img   = gr.Image(type="numpy", label="Input Image")
+            conf_slider = gr.Slider(0.0, 1.0, 0.25, 0.01, label="Confidence Threshold")
+            run_btn     = gr.Button("Detect 🔍", variant="primary")
+        with gr.Column(scale=5, min_width=1000):
+            output_img = gr.Image(
+                type="numpy",
+                label="Annotated Output",
+                height=800,
+                width=1000
+            )
+            time_txt   = gr.Textbox(label="Performance", interactive=False)
+
+    run_btn.click(
+        fn=infer_fire,
+        inputs=[input_img, conf_slider],
+        outputs=[output_img, time_txt]
+    )
+
+    gr.Examples(
+        examples=examples,
+        inputs=[input_img, conf_slider],
+        outputs=[output_img, time_txt],
+        fn=infer_fire,
+        cache_examples=False
+    )
+
+    gr.Markdown(
+        "---\n"
+        "Model trained on a custom wildfire dataset using YOLOv8. "
+        "Adjust the threshold to trade off between false positives and false negatives."
+    )
+
+if __name__ == "__main__":
+    demo.launch()
+       # server_name="0.0.0.0",
+       # server_port=7861,
+       # share=False,
+       # inbrowser=True
+