Paul Gavrikov

Inside Services post / postRemote only post local

8

Code works in Activities, but inside a service both `post` or `postRemote` only post to the local event listeners. ``` public class WearService extends Service { public WearService() { }...

Request: visualkeras

1

Hi Ashish, would you mind adding https://github.com/paulgavrikov/visualkeras/? It's a pypi package that allows network architecture generation directly from python and uses your keras model.

What dataset are these models pretrained on?

1

Could you please tell me what dataset the following models are pretrained on? - SSD-MobileNet (dataset combinedNMS?) https://github.com/IntelAI/models/tree/master/benchmarks/object_detection/tensorflow/ssd-mobilenet - SSD-ResNet34 https://github.com/IntelAI/models/tree/master/benchmarks/object_detection/tensorflow/ssd-resnet34 - UNet https://github.com/IntelAI/models/tree/master/benchmarks/image_segmentation/tensorflow/unet

Hi, Could you please provide information about the datasets used to train the models provided under https://github.com/ai-med/quickNAT_pytorch/tree/master/saved_models?

Error: can't scan: Unknown HCI Command

6

I am trying to follow the tutorial on mcuboot@Zephyr but can't establish a connection from mcuboot (fails at scanning for the slave). I am on Ubuntu 18.04 and tried with...

ResNet/CNN backbone experiments?

1

Do the authors have any experiments with training or distilling other backbones, e.g. ResNets?

[feat] differentiable solarization

1

#### Changes Based on #2490 I implemented a `torch.where` approximation based on a sigmoid. This allows `kornia.enhance.solarize` to be differentiable with respect to the parameters (and could be used in...

Make kornia.enhance.solarize differentiable

5

## 🚀 Feature Make `kornia.enhance.solarize` differentiable. ## Motivation Many if not all "enhance" operators are (naturally) differentiable, but `solarize` is not. ## Pitch 1) It would be helpful to clarify...

Pretrained Models do not load with pip release

6

I installed the latest version of medclip **via pip** and am trying to load the pretrained models. However, when loading ResNet or ViT I get the following error ``` RuntimeError:...

Could you also release the R50 checkpoints fine-tuned on ImageNet-1k from FractalDB1k/10k?