-
### Describe the bug
When we try to train on distributed environment, model should be wrapped with `DDP`. Then, original model can be accessed as `model.module`
In 2-stage detection pipeline, there …
-
### Describe the bug
Import error occurred when I try to execute code with python 3.10.12
```
ImportError: cannot import name 'Sequence' from 'collections'
```
It's because `Sequence` has be…
-
### Describe the bug
When I executed example code in example_train.sh, I couldn't get fx and onnx files.
```
netspresso-train --data config/data/beans.yaml --augmentation config/augmentation/clas…
-
## Description
- Run multi-gpu options without starting with `torch.distributed.run` or `torchrun`.
- Equalize with `netspresso-trainer` or `netspressor-trainer-fx` and add `--num-gpus` options to…
-
## Description
- Apply ResNet50 with NP Launcher by PyNetsPresso
- If applicable, apply NP-compressor-compressed model with NP Launcher by PyNetsPresso
## Checklists
- [ ] I would create a c…
-
## Description
- SegFormer training
- PyNetsPresso compressor
- SegFormer finetuning with compressed model
## Checklists
- [ ] I would create a corresponding branch for this issue from the …
-
- [x] I have searched to see if a similar issue already exists.
**Is there any feature that you would like to add?**
There is `model_name` check before return tranform function.
https://githu…
-
### Describe the bug
Error occurred when initialize `BasePipeline`,
`self.save_optimizer_state = self.conf.logging.save_optimizer_state`
I tried to find `save_optimizer_state` over the project, b…
-
## Description
- Present NetsPresso Trainer detection model can't pass `train -> compression -> retrain -> benchmark` process because PyNetsPresso can't compress present detection head.
- In this …
-
### Describe the bug
The recent libraries with FastAPI require an up-to-date typing_extensions (e.g. `4.8.0`)
The version requirements are too specific, so it is hard to install with other libra…