lama

lama copied to clipboard

请问从哪里获得big-lama.pt呀

2

wu6-s

- Update albumentations import to fix compatibility - Remove imgaug as its no longer maintained and causing compatibility issues - Change torch.load to have weights_only=False for proper loading of model...

dineshnagumothusr

Current behavior: Regardless of the device configuration in the configuration file(`configs/prediction/default.yaml`), the code will only use the CPU for prediction unless refine is enabled. Expected behavior: Use the corresponding device...

wolf-joe

I have fixed the error of the cv2 as well other libraries by getting a compatible version of them also I have uploaded the new notebook and made changes in...

siddharajvk

As the title suggests, when I don't use `refine = False`, the generated results are quite good. However, when I enable `refine = True`, it seems as if nothing was...

bsbacon0966

Library does not even install

1

Error: numpy not found Steps to reproduce: 1. Make venv and activate 2. Clone repository and install requirements

thebest132

finetune get worse result

1

Hi, thanks for your great work! I am trying to fine-tune the big-lama model on my custom dataset, but the output after fine-tuning seems worse than the original version. Here...

rex-29

Issue with np.sctypes

2

>> indir="C:\Users\adop\Pictures\castle_hyperlapse\frames\frame_0001.png" ` >> outdir="C:\Users\adop\Pictures\castle_hyperlapse\inpainted_frames" >> Traceback (most recent call last): File "C:\Users\adop\Pictures\castle_hyperlapse\lama\bin\predict.py", line 31, in from saicinpainting.training.data.datasets import make_default_val_dataset File "C:\Users\adop\Pictures\castle_hyperlapse\lama\saicinpainting\training\data\datasets.py", line 6, in import albumentations as A File...

AshutoshDhaka23

Amazing repo! Just wanted to put together a short guide to finetune this in 2024 with RunPod or similar, since getting this working too piecing together info from a few...

baptistecumin

Training failed due to num_samples should be a positive integer value, but got num_samples=0:

3

Hello. I am encountering error "Training failed due to num_samples should be a positive integer value, but got num_samples=0:" when running the train.py script on my custom images. The folder...

andrea-aleardi