james-imi

Opening this up.this code ``` bbox_params=A.BboxParams(format="pascal_voc", min_visibility=0.3), ``` Can produce mask without a box  and even produces mismatched outputs the bounding boxes has 1 output while the mask has...

I'm pretty sure YOLOv5 can handle instance segmentation as you have trained models for it and I have used it before. I just forgot the annotation format. On Wed, Jan...

Hmm. I think i used it without the boxes and only the segmentation points. Isnt it? On Wed, Jan 31, 2024, 14:54 Glenn Jocher ***@***.***> wrote: > Apologies for the...

For #1 ``` Name: comet-ml Version: 3.37.0 Summary: Supercharging Machine Learning Home-page: https://www.comet.com Author: Comet ML Inc. Author-email: [email protected] License: Proprietary Location: /home/james/anaconda3/envs/yolov5/lib/python3.10/site-packages Requires: dulwich, everett, jsonschema, psutil, python-box, requests,...

If you open it via `Image.open`, the mode is `I` instead of `RGB`

alerady mentioned it before but in case its not clear 

So you dont have to quantize it every time you load it?

@BloodAxe Hi thanks for the info. so using super gradients way of `predict()` is not a go-to for local CPU server inference, is that it? For quantization-aware training, what would...

@BloodAxe any function in the repo for post processing ONNX results?

just my own notebook. also, by looking at your code. the `Losses` does not have `DiceLoss` in there