PyTorch-YOLOv3

PyTorch-YOLOv3 copied to clipboard

## What I'm trying to do I want to be able to have thersholds per class in the evaluate stage. Most significantly the confidence threshold, the nms threshold could also...

avivples

how to understand this:loss_conf = self.obj_scale * loss_conf_obj + self.noobj_scale * loss_conf_noobj

3

92ypli

RuntimeError: binary_op(): expected both inputs to be on same device, but input a is on cuda:0 and input b is on cuda:1

2

When i want to train the model on multiple gpus, this error appears.So I can only train it on gpu:0

ZhxJia

Why is the output of the model has no grads?

2

Hi, thank you for building this pytorch version of YOLO v3. I am implementing my codes on the basis of this project. I used the darknet model in this project...

wzjin2017

Training is slow on Google Cloud Nvidia Tesla P100

7

I am running my model on Google Cloud with 8vCPUs and 52GB RAM and 1GPU - Nvidia Tesla P100 with a batch size of 16. But, It is taking around...

Muthu2093

Error about custom data training

7

I met error as below when I try to train the custom data(512*512 jpg images). What is the problem? Please review my error message. ------------------------------------------------------------------------------------------- >python3 train.py --model_def config/yolov3-custom.cfg --data_config...

ansebby

Does this implementation can create a model from YOLOv3-spp.cfg

1

asebaq

Using the model as an argument to a class method

1

I am passing in **model = Darknet(config_file, img_size=opt_img_size).to(device)** into my custom class method as an object. Previously the model did not output any detections. I found out that the problem...

b-sia

low mAP on coco val2014

10

Has anyone been able to reproduce the mAP of ~55 by running: python3 test.py --weights_path weights/yolov3.weights I was only able to get mAP 50? I got mAP 53 by setting...

janetwise

Multiple GPU

4

I am trying to run training on two GPUs ``` StreamExecutor device (0): GeForce RTX 2080 Ti, Compute Capability 7.5 StreamExecutor device (1): GeForce RTX 2080 Ti, Compute Capability 7.5...

MinaGabriel