AlexTS1980

> Check out the last two annotations listed here: https://github.com/ieee8023/covid-chestxray-dataset#annotations I believe they have more than just lung annotations. Thanks. I'm building a model that can generalize to arbitrary datasets....

> Here are masks for the RSNA pneumonia dataset: https://github.com/mlmed/torchxrayvision/blob/master/scripts/xray_masks.ipynb Awsm thanks!

> Here are masks for the RSNA pneumonia dataset: https://github.com/mlmed/torchxrayvision/blob/master/scripts/xray_masks.ipynb RSNA are more like bounding boxes than masks, but it's probably the best there is. The only x-ray lesion masks...

> If you find anymore mask datasets please share! I found a very good chest ct scan dataset that I used for my models (https://github.com/AlexTS1980/COVID-Single-Shot-Model) here: http://ncov-ai.big.ac.cn/download. It's 750 slices...

> Hi, I've fixed the problem. I noticed that you used > `stylegan2_pytorch --name sg2_0 --data '/data/oxford-102-flowers/jpg' --aug-prob 0.25 --results_dir /results --models_dir /to/models` > to train your network, but when...

I don't think it is very large, maybe 950x500. Any other possible reason you think?

OK, so I use the first image 1502x640 in the script and in the online demo. The result from the demo is far better. When I do it with a...

What changes do I need to make?

I'm sorry, I don't understand why label values have to be between 0 and 255? If a label in my case is (x,y,h,w) bouding box parameters, for example?

But this means most numbers will be fractional. This is a problem.