Detect source as .txt error

[pngmafia]

When running detect.py with source as a txt it is loading the images but getting this error:
Traceback (most recent call last): File "detect.py", line 181, in <module> detect() File "detect.py", line 49, in detect dataset = LoadStreams(source, img_size=imgsz) File "/mnt1/yolo3/utils/datasets.py", line 286, in __init__ s = np.stack([letterbox(x, new_shape=self.img_size)[0].shape for x in self.imgs], 0) # inference shapes File "/mnt1/yolo3/utils/datasets.py", line 286, in <listcomp> s = np.stack([letterbox(x, new_shape=self.img_size)[0].shape for x in self.imgs], 0) # inference shapes File "/mnt1/yolo3/utils/datasets.py", line 696, in letterbox shape = img.shape[:2] # current shape [height, width] AttributeError: 'NoneType' object has no attribute 'shape'

please help!!

[pngmafia]

command used
nohup python -u detect.py --weights ./runs/train/exp72/weights/best.pt --source ./../run.txt --conf-thres 0.4 --iou-thres 0.45 --save-txt --save-conf --device 0,1,2,3 > ll.txt

run.txt has paths to the images, just like we use in training

[glenn-jocher]

@pngmafia at the moment a text file passed to detect.py may only contain RTSP/HTTP/RTMP streaming addresses, not local media.

[sbmalik]

@glenn-jocher are you guys working on providing support for .txt file, which contains path of images.

[glenn-jocher]

@m3sibti there's no current plans for this. Could you use the glob command instead?

python detect.py --source path/*.jpg

Also note that YOLOv5 autoshape models from PyTorch Hub have the capability for inputting lists of local or internet image paths. See https://github.com/ultralytics/yolov5#pytorch-hub

[glenn-jocher]

For YOLOv5 authoshape model inference from image paths or URIs:

    def forward(self, imgs, size=640, augment=False, profile=False):
        # Inference from various sources. For height=720, width=1280, RGB images example inputs are:
        #   filename:   imgs = 'data/samples/zidane.jpg'
        #   URI:             = 'https://github.com/ultralytics/yolov5/releases/download/v1.0/zidane.jpg'
        #   OpenCV:          = cv2.imread('image.jpg')[:,:,::-1]  # HWC BGR to RGB x(720,1280,3)
        #   PIL:             = Image.open('image.jpg')  # HWC x(720,1280,3)
        #   numpy:           = np.zeros((720,1280,3))  # HWC
        #   torch:           = torch.zeros(16,3,720,1280)  # BCHW
        #   multiple:        = [Image.open('image1.jpg'), Image.open('image2.jpg'), ...]  # list of images

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.