basic inference now works

app.py

@@ -1,7 +1,66 @@
 import gradio as gr
+from ultralytics import YOLO
+from sahi.prediction import ObjectPrediction
+from sahi.utils.cv import visualize_object_predictions, read_image
 
-def greet(name):
-    return "Hello " + name + "!!"
+from gradio.components import Slider, Image, Dropdown
 
-iface = gr.Interface(fn=greet, inputs="text", outputs="text")
-iface.launch()
+
+def yolov8_inference(
+    image: Image = None,
+    model_path: Dropdown = None,
+    image_size: Slider = 640,
+    confidence_threshold: Slider = 0.25,
+    iou_threshold: Slider = 0.45,
+):
+    model = YOLO(model_path)
+    model.conf = confidence_threshold
+    model.iou = iou_threshold
+    results = model.predict(image, imgsz=image_size)
+    object_prediction_list = []
+    for _, image_results in enumerate(results):
+        if len(image_results) != 0:
+            image_predictions_in_xyxy_format = image_results.boxes.data
+            for pred in image_predictions_in_xyxy_format:
+                x1, y1, x2, y2 = (
+                    int(pred[0]),
+                    int(pred[1]),
+                    int(pred[2]),
+                    int(pred[3]),
+                )
+                bbox = [x1, y1, x2, y2]
+                score = pred[4]
+                category_name = model.model.names[int(pred[5])]
+                category_id = pred[5]
+                object_prediction = ObjectPrediction(
+                    bbox=bbox,
+                    category_id=int(category_id),
+                    score=score,
+                    category_name=category_name,
+                )
+                object_prediction_list.append(object_prediction)
+    
+    output_image = visualize_object_predictions(image=image, object_prediction_list=object_prediction_list)
+    return output_image['image']
+
+inputs = [
+    "image",
+    Dropdown(choices=["yolo/runs/detect/train10/weights/best.pt"], label="Model"),
+    Slider(minimum=320, maximum=1280, step=32, label="Image Size"),
+    Slider(minimum=0.0, maximum=1.0, step=0.05, label="Confidence Threshold"),
+    Slider(minimum=0.0, maximum=1.0, step=0.05, label="IOU Threshold"),
+]
+
+title = "Smartathon Pothole Challenge"
+
+examples = [['examples/0011.png', 'yolo/runs/detect/train10/weights/best.pt', 640, 0.25, 0.45], ['examples/0014.png', 'yolo/runs/detect/train10/weights/best.pt', 640, 0.25, 0.45], ['examples/0021.png', 'yolo/runs/detect/train10/weights/best.pt', 640, 0.25, 0.45]]
+
+iface = gr.Interface(
+    fn=yolov8_inference,
+    inputs=inputs,
+    outputs="image",
+    title=title,
+    examples=examples,
+    theme="default",
+)
+iface.launch(debug=True)

requirements.txt

@@ -0,0 +1,12 @@
+pytorch-lightning==1.7.3
+matplotlib
+opencv-python
+tqdm
+imageio
+path
+scipy
+configargparse
+kornia
+gradio
+ultrlytics
+sahi