PyTorch-YOLOv3
PyTorch-YOLOv3 copied to clipboard
eriklindernoren
Minimal PyTorch implementation of YOLOv3
Hello, I am training Yolo with Draknet backend on my costum dataset with only 1 class. I see that loss is decreasing, so training sounds to be working, but cls_acc...
Hi @eriklindernoren , I'm training with a custom dataset (1 class, 2900 images for train, 900 for evaluation). With another repo of Yolov3 ([this one](https://github.com/AlexeyAB/darknet)) I get an mAP of...
when i train my datasets,after i execute “python train.py --model_def config\yolov3.cfg --data_config config\custom.data --pretrained_weights weights\darknet53.conv.74“ in cmd,it gives me the following "Runtime error": **Traceback (most recent call last): File "train.py",...
I tried this command python3 train.py --data_config config/coco.data --pretrained_weights weights/darknet53.conv.74 but get some worry. why the recall,precision,cls_acc values is always 0. 
I am confused about what the model should be outputting by the model here: https://github.com/eriklindernoren/PyTorch-YOLOv3/blob/47b7c912877ca69db35b8af3a38d6522681b3bb3/detect.py#L77 I would expect coordinates [0,1] representing the center of the object, but the values are...
If I'm using only custom data set to train yolo model, how can I save the trained weights in to weights folder and use it for testing my custom data?
In the code, the optimizer is Adam. I trained it for a long time but the loss is around 8 and MAP is around 0.06. This is frustrateing. It is...
Hi Why you use `model.apply(weights_init_normal)` after created the model? Is this line necessary?
Should we add some restriction to the output of Yolo layer?
```python for epoch in range(opt.epochs): model.train() start_time = time.time() for batch_i, (_, imgs, targets) in enumerate(dataloader): batches_done = len(dataloader) * epoch + batch_i imgs = Variable(imgs.to(device)) targets = Variable(targets.to(device), requires_grad=False)...
Metadata
Owner
Metadata
Minimal PyTorch implementation of YOLOv3