Delete app.py

app.py

@@ -1,204 +0,0 @@
-import gradio as gr
-import argparse
-import glob
-import os
-import random
-import time
-from pathlib import Path
-
-import cv2
-import numpy as np
-import torch
-import onnxruntime as ort
-from PIL import Image
-from torchvision import transforms
-from tqdm import tqdm
-
-from utils.general import non_max_suppression, scale_coords, letterbox_for_img
-from utils.plots import show_seg_result, plot_one_box, driving_area_mask, lane_line_mask
-from utils.torch_utils import select_device, AverageMeter, time_synchronized
-
-
-def detect(path, model):
-    # with torch.no_grad()：
-    parser = argparse.ArgumentParser()
-    parser.add_argument('--weights', type=str, default=model + '.onnx', help='model.pth path(s)')
-    parser.add_argument('--source', type=str, default=path, help='file/folder  ex:inference/images')
-    parser.add_argument('--img-size', type=int, default=640, help='inference size (pixels)')
-    parser.add_argument('--conf-thres', type=float, default=0.25, help='object confidence threshold')
-    parser.add_argument('--iou-thres', type=float, default=0.45, help='IOU threshold for NMS')
-    parser.add_argument('--device', default='cpu', help='cuda device, i.e. 0 or 0,1,2,3 or cpu')
-    parser.add_argument('--save_dir', type=str, default='runs/detect', help='directory to save results')
-    parser.add_argument('--original_shape', default=True, help='maintain original shape')
-    opt = parser.parse_args()
-
-    device = select_device(device=opt.device)
-    half = device.type != 'cpu'  # half precision only supported on CUDA
-
-    # Load model
-    inf_time = AverageMeter()
-    nms_time = AverageMeter()
-
-    ort.set_default_logger_severity(4)
-    ort_session = ort.InferenceSession('weights/' + opt.weights)
-
-    # Set Dataloader
-    dataset = LoadImages(opt.source, img_size=opt.img_size)
-
-    transform = transforms.Compose([
-        transforms.ToTensor(),
-        transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
-    ])
-
-    t0 = time.time()
-    for i, (path, img, img_det, vid_cap, shapes) in tqdm(enumerate(dataset), total=len(dataset)):
-        img = transform(img).to(device)
-        img = img.half() if half else img.float()  # uint8 to fp16/32
-        if img.ndimension() == 3:
-            img = img.unsqueeze(0)
-
-        img = img.numpy()
-
-        # Inference
-        t1 = time_synchronized()
-        if opt.weights == 'mtpnet.onnx':
-            det_out, da_seg_out, ll_seg_out = ort_session.run(
-                ['detect_output', 'driving_area_segmentation', 'lane_line_segmentation'], input_feed={"images": img}
-            )
-        else:
-            det_out, da_seg_out, ll_seg_out = ort_session.run(
-                ['det_out', 'drive_area_seg', 'lane_line_seg'], input_feed={"images": img}
-            )
-        t2 = time_synchronized()
-
-        det_out = torch.from_numpy(det_out).float()
-        da_seg_out = torch.from_numpy(da_seg_out).float()
-        ll_seg_out = torch.from_numpy(ll_seg_out).float()
-
-        # Apply NMS
-        t3 = time_synchronized()
-        det_pred = non_max_suppression(det_out, opt.conf_thres, opt.iou_thres)
-        t4 = time_synchronized()
-
-        height, width = img.shape[2], img.shape[3]
-        pad_w, pad_h = shapes[1][1]
-        pad_w, pad_h = int(pad_w), int(pad_h)
-
-        inf_time.update(t2 - t1, img.shape[0])
-        nms_time.update(t4 - t3, img.shape[0])
-
-        da_seg_mask = driving_area_mask(da_seg_out, width, height, pad_w, pad_h, 0.5)
-        ll_seg_mask = lane_line_mask(ll_seg_out, width, height, pad_w, pad_h, 0.5)
-        img_det = show_seg_result(img_det, (da_seg_mask, ll_seg_mask), batch=0, is_demo=True)
-
-        det = det_pred[0]
-        if len(det):
-            det[:, :4] = scale_coords(img.shape[2:], det[:, :4], img_det.shape).round()
-            for *xyxy, conf, cls in reversed(det):
-                label_det_pred = 'Car ' + f'{conf * 100:.2f}' + '%'
-                plot_one_box(xyxy, img_det, label=label_det_pred, color=[0, 255, 255], line_thickness=1)
-
-        if opt.original_shape:
-            ori_height = int(640 / shapes[1][0][0])
-            ori_width = int(640 / shapes[1][0][1])
-            img_det = cv2.resize(img_det, (ori_width, ori_height), interpolation=cv2.INTER_LINEAR)
-
-        print('Done. (%.3fs)' % (time.time() - t0))
-        print('inf : (%.4fs/frame)   nms : (%.4fs/frame)' % (inf_time.avg, nms_time.avg))
-
-    return Image.fromarray(img_det[:, :, ::-1])
-
-
-class LoadImages:  # for inference
-    def __init__(self, path, img_size=640, stride=32):
-        p = str(Path(path))  # os-agnostic
-        p = os.path.abspath(p)  # absolute path
-        if '*' in p:
-            files = sorted(glob.glob(p, recursive=True))  # glob
-        elif os.path.isdir(p):
-            files = sorted(glob.glob(os.path.join(p, '*.*')))  # dir
-        elif os.path.isfile(p):
-            files = [p]  # files
-        else:
-            raise Exception('ERROR: %s does not exist' % p)
-
-        img_formats = ['.bmp', '.jpg', '.jpeg', '.png', '.tif', '.tiff', '.dng']
-        vid_formats = ['.mov', '.avi', '.mp4', '.mpg', '.mpeg', '.m4v', '.wmv', '.mkv']
-        images = [x for x in files if os.path.splitext(x)[-1].lower() in img_formats]
-        videos = [x for x in files if os.path.splitext(x)[-1].lower() in vid_formats]
-        ni, nv = len(images), len(videos)
-
-        self.img_size = img_size
-        self.files = images + videos
-        self.nf = ni + nv  # number of files
-        self.video_flag = [False] * ni + [True] * nv
-        self.mode = 'images'
-        self.frame = 0
-        self.nframes = 0
-        if any(videos):
-            self.new_video(videos[0])  # new video
-        else:
-            self.cap = None
-        assert self.nf > 0, f'No images or videos found in {p}. ' \
-                            f'Supported formats are:\nimages: {img_formats}\nvideos: {vid_formats}'
-
-    def __iter__(self):
-        self.count = 0
-        return self
-
-    def __next__(self):
-        if self.count == self.nf:
-            raise StopIteration
-        path = self.files[self.count]
-
-        if self.video_flag[self.count]:
-            # Read video
-            self.mode = 'video'
-            ret_val, img0 = self.cap.read()
-            if not ret_val:
-                self.count += 1
-                self.cap.release()
-                if self.count == self.nf:  # last video
-                    raise StopIteration
-                else:
-                    path = self.files[self.count]
-                    self.new_video(path)
-                    ret_val, img0 = self.cap.read()
-            h0, w0 = img0.shape[:2]
-
-            self.frame += 1
-            print('\n video %g/%g (%g/%g) %s: ' % (self.count + 1, self.nf, self.frame, self.nframes, path), end='')
-
-        else:
-            # Read image
-            self.count += 1
-            img0 = cv2.imread(path, cv2.IMREAD_COLOR | cv2.IMREAD_IGNORE_ORIENTATION)  # BGR
-            assert img0 is not None, 'Image Not Found ' + path
-            print('image %g/%g %s: \n' % (self.count, self.nf, path), end='')
-            h0, w0 = img0.shape[:2]
-
-        # Padded resize
-        img0 = cv2.resize(img0, (1280, 720), interpolation=cv2.INTER_LINEAR)
-        img, ratio, pad = letterbox_for_img(img0, new_shape=self.img_size, auto=False)
-        h, w = img.shape[:2]
-        shapes = (h0, w0), ((h / h0, w / w0), pad)
-
-        # Convert
-        # img = img[:, :, ::-1].transpose(2, 0, 1)  # BGR to RGB
-        img = np.ascontiguousarray(img)
-
-        return path, img, img0, self.cap, shapes
-
-    def new_video(self, path):
-        self.frame = 0
-        self.cap = cv2.VideoCapture(path)
-        self.nframes = int(self.cap.get(cv2.CAP_PROP_FRAME_COUNT))
-
-    def __len__(self):
-        return self.nf  # number of files
-
-
-gr.Interface(detect, [gr.Image(type='filepath'), gr.Dropdown(choices=["mtpnet", "yolop"])], gr.Image(type="pil"),
-                      title="Mtpnet", examples=[["example.jpg", "mtpnet"], ["example2.jpg", "mtpnet"]],
-                      description="Mtpnet 🚀: demo for multitask panoptic driving perception network").launch()
-