YashRunwal

Using a different backbone (CSPNet) and retraining on custom images

1

Hello, I have been working on a project which involved detecting objects such as Person, Car, Bicycle, etc. In total there are 6 classes. I read the paper of EfficientDet...

Real time inference on Nvidia Jetson?

5

Hello, Firstly, I would like to thank you for making this repo. I am working on an Object Detection project (multiclass). And after training the EfficientDet architecture on my custom...

Lr-finder with multiple inputs, outputs and losses

21

Hello, Firstly, thank you for this wonderful library. I have a model which expects 2 inputs. I am working with 2 kinds of images, one of size (512, 1536) and...

Pre-trained ResNet Models

1

Hello, Is it possible for you to provide Pre-trained ResNet-18/34/50/101 model with GroupNormalization+WeightStandardization? Thank you.

Error when using heroku repo:gc -a appname

11

When I try to run following command `heroku repo:gc -a indian-notes-classifier` It gives me following error: The requested URL returned error: 404 Not Found Why is that and how can...

Hello, Is this repo maintained? I had a few questions regarding the DLA-X46C backbone? Thanks.

DLA-X-46-C backbone

1

Hello, Has anyone tried using the "DLA-X46-C" backbone from the "Deep Layer Aggregation" paper? This is a compact model with fewer params and high FPS. Thank You.

Class imbalance

3

@yjh0410, My dataset is highly imbalanced. The class count in the training dataset (2700 samples) is as follows: `{'1': 306, '2': 60, '3': 133, '4': 14197, '6': 3886, '7': 69}`...

txty_loss not decreasing and mAP returns the value 0.0

15

@yjh0410 , As you know, I am training the model with grayscale images (512, 1536) with a few augmentation techniques. I have trained the model (pre-trained Resnet 18) for about...

Modifying gt_tensor

2

@yjh0410 Hi, I am using this CenterNet model for my dataset. I have modified a few things to make it work for me and I can train for Object Detection....