a-PyTorch-Tutorial-to-Object-Detection
a-PyTorch-Tutorial-to-Object-Detection copied to clipboard
sgrvinod
SSD: Single Shot MultiBox Detector | a PyTorch Tutorial to Object Detection
I traced back where the nan is coming from the loss function. It leads to `cxcy_to_gcxgcy` function. Division is causing nan value it seems. please provide any solution for this.
Hi, I downloaded your pre-trained model from google drive. But it gives an error when loading the checkpoint. ----> 1 ckp = torch.load('./checkpoint_ssd300.pth.tar') ~/anaconda3/envs/PyTorch/lib/python3.8/site-packages/torch/serialization.py in load(f, map_location, pickle_module, **pickle_load_args) 591...
I noticed in the paper each Prediction Convolution is formulated with output channels = n_boxes * (n_classes + 4), but in the [code](https://github.com/sgrvinod/a-PyTorch-Tutorial-to-Object-Detection/blob/43fd8be9e82b351619a467373d211ee5bf73cef8/model.py#L218) you have separated each level into separate...
I am trying to use my own dataset to train SSD300 but I can't find the way to use it and the correct format. Could you please guide me on...
Good morning! I am using SSD architecture to detect faces on images (2 classes - background and face). I have a labeled WIDERFACES dataset and Im trying to train a...
Could you please update your pth.tar file with pretrained model - it is broken and i cannot unzip it. Thank you!
Hi, In the multi-box loss calculation, in the parameter **predicted_locs** what does N signify? Is it the batch size or the number of objects predicted by the network in the...
I downloaded the `checkpoint_ssd300.pth.tar` and ran the `detect.py` using and I got the detection as Any possible solutions? Thanks
wanted to know if we can detect or classify full person gender classification in the same model can you help with it or can you suggest better training model for...
need help for calculating these in eval.py can you guide how to calculate it for this model