We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
GPU memory leak when runing sp1*sp2
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 = np.random.randint(0, 100, size=(10000, 3)) colors = np.random.randint(0, 255, size=(10000, 1)) feats = np.ones((colors.shape[0],1)) coords_batch, feats_batch = ME.utils.sparse_collate([coords], [feats]) spw = ME.SparseTensor(features=feats_batch.float(), coordinates=coords_batch,device=device,requires_grad=False) feats = colors coords_batch, feats_batch = ME.utils.sparse_collate([coords], [feats]) spc = ME.SparseTensor(features=feats_batch.float(), coordinates=coords_batch, coordinate_manager=spw.coordinate_manager,device=device) for i in range(100000000): s = spc*spw if i%1000==0: meminfo = pynvml.nvmlDeviceGetMemoryInfo(handler) used = round(meminfo.used / 1024 / 1024, 2) print(i,'GPU Mem used', used, 'MB')
The text was updated successfully, but these errors were encountered:
No branches or pull requests
GPU memory leak when runing sp1*sp2
The text was updated successfully, but these errors were encountered: