Zunayed Mahmud

Thank you @MatthiasHinz for mentioning the issue and suggesting a workaround. I followed your suggestion by commenting out [line 180](https://github.com/open-mmlab/mmdetection3d/blame/5c0613be29bd2e51771ec5e046d89ba3089887c7/mmdet3d/visualization/local_visualizer.py#L180). However, I was having this error: ``` TypeError: paint_uniform_color(): incompatible...

> In convert.py > > ``` > ### Image undistortion > ## We need to undistort our images into ideal pinhole intrinsics. > img_undist_cmd = (colmap_command + " image_undistorter \...

Hi @CuongTranXuan, Thank you for sharing the shell script. I ran into the same issue and used your script to download the 7B weight. However, seems like this is also...

@IoanaEmma Hi, sorry about my late response. No, I couldn't move further with this issue. I'll share if I could find a workaround.

Hi, I'm not sure what the issue is. Could you please check which version of python you are using? The particular file you mentioned needs to be installed with python...

I'm not sure exactly what happened but I'm guessing the wheel file somehow got corrupted while downloading. Can you redownload the wheel file and verify the SHA-256 checksum? You can...

@ptsmap, the author addresses the issue [here](https://github.com/VITA-Group/LightGaussian/issues/39#issuecomment-2408713890) and suggested a fix.

As the error suggests, the problem is within your dataset. Your dataset needs to contain valid input tensors of same shape and dtype.

Are you using this codebase for Mask R-CNN? Because this repo is updated for TensorFlow 2.x versions and `tf.log` doesn't even exist in the [`utils.py`](https://github.com/z-mahmud22/Mask-RCNN_TF2.14.0/blob/407a6fdebd3202698ea98b6bd075a65c24536c2a/mrcnn/utils.py#L202) script as it was depricated...

Are your training images grayscale? If yes, can you try converting them to RGB via: `graytorgb = cv2.cvtColor(gray, cv2. COLOR_GRAY2RGB)` and then train?