Update app.py

app.py

@@ -6,18 +6,19 @@ import json
 import gradio as gr
 
 # -------------------
-# Lightweight config
+# Config
 # -------------------
 REPO_ID = "mshamrai/yolov8s-visdrone"
 FILENAME = "weights/best.pt"
 
 SAMPLES_DIR = "samples"
-SAMPLE_IMAGE = os.path.join(SAMPLES_DIR, "drone_sample.jpg")
-SAMPLE_VIDEO = os.path.join(SAMPLES_DIR, "airspace_sample.mp4")
+# Embedded samples (auto-downloaded on start)
+TEST_IMAGE = os.path.join(SAMPLES_DIR, "test_image.jpg")
+TEST_VIDEO = os.path.join(SAMPLES_DIR, "test_video.mp4")
 
 SAMPLE_URLS = {
-    SAMPLE_IMAGE: "https://huggingface.co/datasets/hf-internal-testing/example-documents/resolve/main/airplane.jpg",
-    SAMPLE_VIDEO: "https://huggingface.co/datasets/hf-internal-testing/example-documents/resolve/main/short_harvard_bridge.mp4",
+    TEST_IMAGE: "https://github.com/ultralytics/yolov5/releases/download/v1.0/zidane.jpg",  # small image
+    TEST_VIDEO: "https://github.com/ultralytics/assets/releases/download/v0.0.0/drone.mp4",   # short drone clip
 }
 
 def ensure_samples():
@@ -30,7 +31,7 @@ def ensure_samples():
         if os.path.exists(local_path):
             continue
         try:
-            r = requests.get(url, timeout=20)
+            r = requests.get(url, timeout=30)
             r.raise_for_status()
             with open(local_path, "wb") as f:
                 f.write(r.content)
@@ -242,7 +243,7 @@ def export_pdf_vid(summary: str, counts: dict):
     return _save_pdf("Airspace Drone Detector — Video Report", summary or "No summary.", counts or {}, None)
 
 # -------------------
-# UI
+# UI (preloaded samples)
 # -------------------
 EXAMPLE_NOTE = (
     "Tip: This model is trained on VisDrone-style aerial objects (small targets). "
@@ -253,24 +254,26 @@ with gr.Blocks(title="Airspace Drone Detector (YOLOv8 VisDrone)") as demo:
     gr.Markdown(
         """
 # Airspace Drone Detector (Pretrained YOLOv8 - VisDrone)
-Upload an **image or video** and detect small aerial objects (e.g., drones) using a **pre-trained VisDrone model**.
-No dataset or training required — just run it.
+The sample **image** and **video** are already loaded below — just click **Run**.
 
-**Controls:** Adjust Confidence / NMS to tune detections.  
-**Exports:** Download detections as CSV or a polished PDF report.  
-**Note:** On CPU Spaces, long videos are truncated via **Max frames** for responsiveness.
+**Exports:** CSV + PDF reports.  
+**Note:** On CPU Spaces, long videos are truncated via **Max frames**.
         """
     )
 
     with gr.Tabs():
-        # IMAGE
+        # IMAGE (preloaded)
         with gr.TabItem("Image"):
             with gr.Row():
-                image_in = gr.Image(label="Upload Image", type="numpy")
+                image_in = gr.Image(
+                    value=TEST_IMAGE if os.path.exists(TEST_IMAGE) else None,
+                    type="numpy",
+                    label="Input Image (preloaded)"
+                )
                 with gr.Column():
-                    conf_img = gr.Slider(0.05, 0.8, 0.25, step=0.05, label="Confidence")
+                    conf_img = gr.Slider(0.05, 0.8, 0.35, step=0.05, label="Confidence")
                     iou_img = gr.Slider(0.1, 0.9, 0.45, step=0.05, label="NMS IoU")
-                    run_img = gr.Button("Detect")
+                    run_img = gr.Button("Run Detection")
                     gr.Markdown(EXAMPLE_NOTE)
 
             image_out = gr.Image(label="Annotated Image")
@@ -299,18 +302,18 @@ No dataset or training required — just run it.
                 outputs=[pdf_img_path],
             )
 
-            if os.path.exists(SAMPLE_IMAGE):
-                gr.Examples(examples=[[SAMPLE_IMAGE]], inputs=[image_in], label="Try with a sample image (preloaded)")
-
-        # VIDEO
+        # VIDEO (preloaded)
         with gr.TabItem("Video"):
             with gr.Row():
-                video_in = gr.Video(label="Upload Video (mp4/mov)")
+                video_in = gr.Video(
+                    value=TEST_VIDEO if os.path.exists(TEST_VIDEO) else None,
+                    label="Input Video (preloaded)"
+                )
                 with gr.Column():
-                    conf_vid = gr.Slider(0.05, 0.8, 0.25, step=0.05, label="Confidence")
+                    conf_vid = gr.Slider(0.05, 0.8, 0.35, step=0.05, label="Confidence")
                     iou_vid = gr.Slider(0.1, 0.9, 0.45, step=0.05, label="NMS IoU")
                     max_frames = gr.Slider(60, 2000, 300, step=10, label="Max frames to process")
-                    run_vid = gr.Button("Detect")
+                    run_vid = gr.Button("Run Detection")
                     gr.Markdown(EXAMPLE_NOTE)
 
             video_out = gr.Video(label="Annotated Video")
@@ -349,10 +352,7 @@ No dataset or training required — just run it.
     gr.Markdown(
         f"""
 **Weights:** `{REPO_ID}/{FILENAME}` (downloaded lazily)  
-**Diagnostics**  
-- FFmpeg available: {'Yes' if _ffmpeg_ok() else 'No'}  
-- Python: 3.10  
-- Torch & Ultralytics load on first run.  
+**Diagnostics** — FFmpeg: {'Yes' if _ffmpeg_ok() else 'No'} • Python: 3.10
         """
     )