Chunyang Fu

GPU memory leak when runing **sp1*sp2** ```python import MinkowskiEngine as ME import numpy as np import os import psutil import pynvml pynvml.nvmlInit() gpu_id=0 handler = pynvml.nvmlDeviceGetHandleByIndex(gpu_id) device = 'cuda:0' coords...

What's the tensorflow version

1

I get different bit rates result when I test 'model/128/my-model-770' with different versions of TensorFlow, e.g. TF 1.15 15bpp，TF 1.12 0.8 bpp PSNR 32dB as for kodim01.png. What's the tensorflow...

I met with erratic results

1

I encounter unstable results when I run Python code. Is there any randomness to this method? This is my code: ``` python import numpy as np from pysdf import SDF...

Miss point in MinkowskiConvolution

1

```python dimension = 2 kernel_size = 3 stride= 3 out_channels = kernel_size**dimension conv0_1 = ME.MinkowskiConvolution( in_channels=1, out_channels=out_channels, kernel_size= kernel_size, stride=stride, bias=False, dimension=dimension ) conv0_1.kernel.data = torch.eye(kernel_size**dimension,out_channels).unsqueeze(1) coords =np.array([[0,0],[1,1],[5,5],[0,3]]) feat =np.array([[1],[10],[100],[1000]])...

### Is there an existing issue for this? - [x] I have searched the existing issues ### Current Behavior What precision used in the convolution? How can I improve the...

## Bug I plan to use pytorch 1.31+python 3.10 + cuda 11.7 in conda env. However, when I used pip to install compressai, it automatically downloaded and installed cuda12 and...