ariel415el

Sampling algorithm differ from paper.

5

Hi, I want to elaborate on #2: The sampling algorithm in your paper is a bit different that what shown in the paper. The paper suggests this sample step ![Screenshot...

Fix: Read Lua weights with Pytorch > 1.0

2

from torch.utils.serialization import load_lua doesnt work in current pytorch versions here is a possible fix with **torchfile** ``` class pytorch_lua_wrapper: def __init__(self, lua_path): self.lua_model = torchfile.load(lua_path) def get(self, idx): return...

Hi, I didn't see a License_plate label on the COCO dataset. Can you explain exactly on what data did you train yolo? thanks

Hey @ashertrockman, @tmp-iclr ! wave This pull request makes it possible to run your model inside a Docker environment, which makes it easier for other people to run it. We're...

Typo in diffusion_02_model.ipynb

1

Hi, I think you have an error in:  while it should be q(x_t|x_0).

Add Replicate demo and API

1

Hey @GaParmar and @junyanz ! 👋 Good work on this project I was very impressed by the reconstruction quality of the suggested inversion process. The images actually look better that...

Wrong weights filename for e4e cats

1

Hi, I think the filename for "cats" is wrong here: https://github.com/adobe-research/sam_inversion/blob/4852a2a033ac5af981f91b9eb2baa3df6e2229fa/src/sam_inv_optimization.py#L117 Since there is only "e4e_cats.pt" in https://www.cs.cmu.edu/~SAMInversion/ckpt/ But I think this one is bad too as I'm not able...

Size of input

2

I'm running `python src/sam_inv_optimization.py --image_category "cars" --image_path car1.jpg --output_path "output/cars/" --sweep_thresholds --generate_edits` where car1.jpg is a 1200x800 rgb image. I get the following errors: ```Traceback (most recent call last): File...

Hey @lhoyer ! 👋 This pull request makes it possible to run your model inside a Docker environment, which makes it easier for other people to run it. We're using...

Hi Your implementation as all others I saw clips the model output: https://github.com/hmdolatabadi/denoising_diffusion/blob/392d59a95a85b0ac19918c9b3174d1a71be1c5d1/lib/diffusion.py#L172 This is does not show in the paper but seem crucial for image quality.. Can you explain...