Jixuan Fan comments

Results 5 comments of


                                            Jixuan Fan

Could you explain the q_sample ?

@kada0720 @sandeepjangir07 You can find it in [DDPM](https://arxiv.org/abs/2006.11239) page 4 Here `continuous_sqrt_alpha_cumprod` is $\sqrt{\bar{\alpha}_t}$, thus ` continuous_sqrt_alpha_cumprod**2` equals to $\bar{\alpha}_t$ Besides, you could find similar formula in SR3 paper on...

How to mimic `--loading_half_params --use_tile_vae --load_8bit_llava` for `test.py`?

I have encounterd `RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:1 and cuda:0! (when checking argument for argument weight in method...

Question of pretrained model

Hi, @yxldd `_opt` is like .`state` file which contains training parameters like epoch / iter / scheduler / optimizer etc. while `_gen` is like `.pth` file which contains model parameters....

Setting up DCNv2 properly on RTX 4000 series GPUs

> I am attempting to fine-tune train a reference-based super resolution model with a personal, custom dataset. However, I am unable to properly install and run _any_ of the codebases...

Setting up DCNv2 properly on RTX 4000 series GPUs

@heng0607 Yes and I found an easier way to solve this issue. You just need this: ```bash pip install mmcv-full ``` Because in this line, https://github.com/caojiezhang/DATSR/blob/76faa616774cbef983acaa855eaaec75d5dc9d8c/datsr/models/archs/dcn_v2.py#L14 mmcv==0.4.4 (which is mmcv-lite...