import gradio as gr
import os

os.system('git clone https://github.com/WongKinYiu/yolov7.git')


def detect(inp):
  os.system('python ./yolov7/detect.py --weights best.pt --conf 0.25 --img-size 640 --source f{inp} "--project","./yolov7/runs/detect ')
  otp=inp.split('/')[2]
  return f"./yolov7/runs/detect/exp/*"
   
  #f"./yolov7/runs/detect/exp/{otp}"
  
  
opt  = {
    
    "weights": "best.pt", # Path to weights file default weights are for nano model
    "yaml"   : "custom.yaml",
    "img-size": 640, # default image size
    "conf-thres": 0.25, # confidence threshold for inference.
    "iou-thres" : 0.45, # NMS IoU threshold for inference.
    "device" : '0',  # device to run our model i.e. 0 or 0,1,2,3 or cpu
    "classes" : classes_to_filter  # list of classes to filter or None

}  
  
def detect2(inp):
  with torch.no_grad():
    weights, imgsz = opt['weights'], opt['img-size']
    set_logging()
    device = select_device(opt['device'])
    half = device.type != 'cpu'
    model = attempt_load(weights, map_location=device)  # load FP32 model
    stride = int(model.stride.max())  # model stride
    imgsz = check_img_size(imgsz, s=stride)  # check img_size
    if half:
      model.half()
  
    names = model.module.names if hasattr(model, 'module') else model.names
    colors = [[random.randint(0, 255) for _ in range(3)] for _ in names]
    if device.type != 'cpu':
      model(torch.zeros(1, 3, imgsz, imgsz).to(device).type_as(next(model.parameters())))
  
    img0 = cv2.imread(inp)
    img = letterbox(img0, imgsz, stride=stride)[0]
    img = img[:, :, ::-1].transpose(2, 0, 1)  # BGR to RGB, to 3x416x416
    img = np.ascontiguousarray(img)
    img = torch.from_numpy(img).to(device)
    img = img.half() if half else img.float()  # uint8 to fp16/32
    img /= 255.0  # 0 - 255 to 0.0 - 1.0
    if img.ndimension() == 3:
      img = img.unsqueeze(0)
  
    # Inference
    t1 = time_synchronized()
    pred = model(img, augment= False)[0]
  
    # Apply NMS
    classes = None
    if opt['classes']:
      classes = []
      for class_name in opt['classes']:
  
        classes.append(names.index(class_name))
  
    if classes:
      
      classes = [i for i in range(len(names)) if i not in classes]
    
    
    pred = non_max_suppression(pred, opt['conf-thres'], opt['iou-thres'], classes= [17], agnostic= False)
    t2 = time_synchronized()
    for i, det in enumerate(pred):
      s = ''
      s += '%gx%g ' % img.shape[2:]  # print string
      gn = torch.tensor(img0.shape)[[1, 0, 1, 0]]
      if len(det):
        det[:, :4] = scale_coords(img.shape[2:], det[:, :4], img0.shape).round()
  
        for c in det[:, -1].unique():
          n = (det[:, -1] == c).sum()  # detections per class
          s += f"{n} {names[int(c)]}{'s' * (n > 1)}, "  # add to string
      
        for *xyxy, conf, cls in reversed(det):
  
          label = f'{names[int(cls)]} {conf:.2f}'
          plot_one_box(xyxy, img0, label=label, color=colors[int(cls)], line_thickness=3)
      return img0
      


   
  
  
  
  
  

inp = gr.inputs.Image(type="filepath", label="Input")
outputs=gr.outputs.Image(type="pil", label="Output Image")
#output = gr.outputs.Image(type="filepath", label="Output")
#.outputs.Textbox()

io=gr.Interface(fn=detect2, inputs=inp, outputs=output, title='Pot Hole Detection With Custom YOLOv7 ',examples=[["Examples/img-300_jpg.rf.6b7b035dff1cda092ce3dc22be8d0135.jpg"]])
#,examples=["Examples/img-300_jpg.rf.6b7b035dff1cda092ce3dc22be8d0135.jpg"]
io.launch(debug=True,share=False)