pytorch-retinanet

pytorch-retinanet copied to clipboard

Thanks for this great project! This pull request allows CSV datasets to use a relative path in the file. This is accomplished by specifying the path to the directory containing...

biomiker

UnpicklingError: invalid load key, '\x0a'

1

I'm trying to validate this model by using colab, and the Unpickling Error occurred like below: Traceback (most recent call last): File "coco_validation.py", line 51, in main() File "coco_validation.py", line...

miramira227

question about loss

2

in advance, thanks for your code. 1. cls loss. I think a value is wrong now a value is assigned 0 (back ground) for 0.75, and 1 for 0.25, i...

beomgonyu

In function `evaluate_coco(dataset, model, threshold=0.05)` I noticed that when I put the threshold which is different from default value 0.05, I receive the low MAP. Then I found that the...

EonianCoda

I have noticed that sometimes when I put some threshold, I receive low score labels. After debugging I find some issue in this part of code: https://github.com/yhenon/pytorch-retinanet/blob/39938d285fc3c602ae6a697ff690bbca29a502be/visualize_single_image.py#L103_L119 I use torch...

EzYoloFrodo

The Training loss is 0

6

 As this image, the classification loss is 0 and 2.30212 repeatedly. The object I want to find is very small defect, while my image size is 4608 x 3288....

tingweii

The performance of the retinanet

1

@yhenon-nextdroid @yhenon @xu1718191411 @mimoralea @rvandeghen Thanks to your great works! But i got some training issues in this repo. The original retinanet using SGD and finetune 12epochs on the ImageNet...

WangJian981002

only one side keep 640，and another side is changing😂 this is my output shape: torch.Size([1, 3, 928, 640]) #the first number is batchsize Epoch: 0 | Iteration: 67 | Cl:...

Cleehao

Question about precision and recall metrics

1

Hi I would like to know how to calcule precision, recall and specificity metrics. Using the csv_eval.py to calcule the metrics I got the following results on my custom dataset:...

alancarlosml

Sporadic message : input tensors must be on the same device. Received cpu and cuda:0

8

During training i have a sporadic message : input tensors must be on the same device. Received cpu and cuda:0 At the end it does not have any influence on...

wvalcke