ssd.pytorch

ssd.pytorch copied to clipboard

Training when images may not have any ground truth bboxes

6

Hi, I've a dataset which has images which may or may not have ground truth boxes. I've tweaked the dataloader and MultiboxLoss accordingly. In cases where the image doesn't have...

pyaf

DeprecationWarning: elementwise comparison failed; this will raise an error in the future.

1

BerryLJ

IndexError: too many indices for array

15

File "/home/kun/Software/ssd-pytorch-master/data/voc0712.py", line 145, in pull_item img, boxes, labels = self.transform(img, target[:, :4], target[:, 4]) IndexError: too many indices for array I want to train voc dataset, but after i...

mama110

TypeError: forward() missing 1 required positional argument: 'x'

5

when i use eval.py to evaluate my model. it reports this problem. How can i solve this problem?

BattleOfDance

I'm wondering if it is possible to extract the class confidence and location box loss for each image in the batch, so not having a unique loss metric at the...

zuliani99

Traceback (most recent call last): File "D:/PyCharm/ssd.pytorch-master/train2.py", line 247, in train() File "D:/PyCharm/ssd.pytorch-master/train2.py", line 170, in train loss_l, loss_c = criterion(out, targets) File "D:\PyCharm\hellopytorch\venv\lib\site-packages\torch\nn\modules\module.py", line 727, in _call_impl result =...

0Freeebaby

FALL0404

timer: 5.7830 sec. iter 0 || Loss: 26.5509 || F:\yolov8_change\ssd.pytorch-master\utils\augmentations.py:238: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes)...

woshidamow5534

Running eval.py on custom dataset and mAP = 0

2

Dear all. I am running SSD300 on a custom chest X-ray dataset preprocessed as VOC format, so far train.py, test.py and demo.py works well while for the eval.py, I am...

cattyhubby

sheerin06