wg

Maybe you can add the following line before `BB = BB[sorted_ind, :]` in voc_eval.py `if sorted_ind.size > 0:`. this "if" statement should encompass from line 149 to 195. Then add...

I've got similar error. By changing `fg_rois_per_this_image = min(fg_rois_per_image, fg_inds.size)` to `fg_rois_per_this_image = np.int64(min(fg_rois_per_image, fg_inds.size))` in line 182 in proposal_target_layer.py works for me. Hope this would help.

> Please read the issues. I hope you find answers to your questions in #3 and #11. > > I intend to add more documentation and clean up the code...

The following is the code of my GTUtility class. I tried to modify the code based on your reply for another issue, https://github.com/mvoelk/ssd_detectors/issues/12#issuecomment-485686377. ``` class GTUtility(BaseGTUtility): def __init__(self, data_path, polygon=True):...

The inputs.shape is (1, 512, 512, 3). The data is [array([[0.2275 , 0.44125, 0.2275 , 0.55875, 0.77125, 0.55875, 0.77125, 0.44125, 1. ]])]. data.shape raises AttributeError: 'list' object has no attribute...

I changed batch size to 2 with 100 samples, but still got similar invalid argument error: `InvalidArgumentError: Reshape cannot infer the missing input size for an empty tensor unless all...

Thanks. I used `print(model.input_shape)` below `# SegLink + DenseNet` `model = DSODSL512()` The shape is (None, 512, 512, 3). It seems as it should be.

I changed the SegLinkLoss to `loss = SegLinkFocalLoss(lambda_segments=1.0, lambda_offsets=1.0, lambda_links=1.0)` and it works as magic. Thanks so much. But why I didn't get negative samples in the local ground truth?...

How do I use LabelImg tool to create custom dataset? For example, I have an image below needs to be labeled, then how to do the labeling? Should I give...