Shuai Zhao comments

Repositories
Issues
Comments

Results 2 comments of


                                            Shuai Zhao

Why the result of FCN-8s was better than deeplabv3?

Very nice work. Thank you for your endeavor. I am also interested in this phenomenon. I see your provided [DeepLabv3](https://github.com/kazuto1011/deeplab-pytorch/blob/master/libs/models/deeplabv3plus.py), it used `output_stride=16` with `ResNet-101`. On the other hand, FCN-8s...

Why add the following code snippet since it would bring some errors?

> Hello, the correct way to save the intermediate checkpoints for PEFT when using Trainer would be to use Callbacks. An example is shown here: https://github.com/huggingface/peft/blob/main/examples/int8_training/peft_bnb_whisper_large_v2_training.ipynb > > ``` >...