import gradio as gr
import torch
from sahi.classification import ImageClassification
from sahi.utils.cv import visualize_object_predictions, read_image
from ultralyticsplus import YOLO

def yolov8_inference(
    image: gr.Image = None,
    model_path: gr.Dropdown = None,
    image_size: gr.Slider = 640,
    conf_threshold: gr.Slider = 0.25,
    iou_threshold: gr.Slider = 0.45,
):
    """
    YOLOv8 inference function
    Args:
        image: Input image
        model_path: Path to the model
        image_size: Image size
        conf_threshold: Confidence threshold
        iou_threshold: IOU threshold
    
    """
    model = YOLO(model_path)
    model.overrides['conf'] = conf_threshold
    model.overrides['iou']= iou_threshold
    model.overrides['agnostic_nms'] = False  # NMS class-agnostic
    model.overrides['max_det'] = 1000 
    
    # observe results
    top_class_index = torch.argmax(results[0].probs).item()
    Class = model.names[top_class_index]
    print(Class)
        

inputs = [
    gr.Image(type="filepath", label="Input Image"),
    gr.Dropdown(["foduucom/Tyre-Quality-Classification-AI"], 
                default="foduucom/Tyre-Quality-Classification-AI", label="Model"),
    gr.Slider(minimum=320, maximum=1280, default=640, step=32, label="Image Size"),
    gr.Slider(minimum=0.0, maximum=1.0, default=0.25, step=0.05, label="Confidence Threshold"),
    gr.Slider(minimum=0.0, maximum=1.0, default=0.45, step=0.05, label="IOU Threshold"),
]

outputs = gr.Image(type="filepath", label="Output Image")
title = "AI-Powered Tire Quality Inspection: YOLOv8s Enhanced Classification"




Space Description:"""
Welcome to our 🤖 AI-Powered Tire Quality Inspection Space – a cutting-edge solution harnessing the capabilities of YOLOv8s to revolutionize 🚗 tire quality control processes.
"""
About This Space: """
This interactive platform empowers you to classify tires with unparalleled precision, utilizing a fine-tuned YOLOv8s model 🎯 specifically developed for identifying defects in tire manufacturing. By submitting an image of a tire, you can instantly determine whether it meets the rigorous quality standards required in the industry, helping to ensure safety and reliability in automotive products.
"""
examples = [['samples/1.jpeg', 'foduucom/thermal-image-object-detection', 640, 0.25, 0.45], ['samples/2.jpg', 'foduucom/thermal-image-object-detection', 640, 0.25, 0.45]]
demo_app = gr.Interface(
    fn=yolov8_inference,
    inputs=inputs,
    outputs=outputs,
    title=title,
    description=description,
    examples=examples,
    cache_examples=True,
    theme='huggingface',
)
demo_app.queue().launch(debug=True)