Lei-Lei Ma
Lei-Lei Ma
## Describe the bug jittor.jittor_core.Var 转换成 numpy 失败 def to_image(pic): pic = pic.multiply(255).uint8() print(type(pic)) npimg = np.transpose(pic.numpy(), (1, 2, 0)) return Image.fromarray(npimg) ## Full Log ``` npimg = np.transpose(pic.numpy(), (1,...
## Describe the bug Import error and python test error export LDFLAGS="-L/home/linuxbrew/.linuxbrew/opt/libomp/lib" export CPPFLAGS="-I/home/linuxbrew/.linuxbrew/opt/libomp/include" export nvcc_path="/opt/cuda/bin/nvcc" OS: Arch Linux x86_64 Kernel: 5.13.10-arch1-1 Uptime: 47 mins Packages: 850 (pacman) Shell: zsh...
 This result is difficult to reproduce. Not only am I having a hard time solving this problem, but others are also having this problem too. Would you share the...
one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [36, 3, 128, 128]], which is output 0 of torch::autograd::CopySlices, is at version 9;...
`$ python train.py Namespace(arch='BAT', aug=1, bt_size=8, cross=0, dataset='isic2016', exp_name='', fold='0', gpu='0', lr_seg=0.0001, n_epochs=200, net_layer=50, patience=500, point_pred=1, ppl=6, pre=0, seg_loss=0, trans=1) Loaded 900 frames Loaded 379 frames pretrained resnet, 50 ERROR:...
TypeError: spmm() missing 1 required positional argument: 'matrix' pytorch == 1.4.0 pytorch_sparse Refer to the official website to install `--------------Parameters-------------- Namespace(batch=1, check_point=1, cuda=True, dataset='THAD6h', dropout=0.2, embedding='thailand', epochs=1000, f_dim=100, fastmode=False, lr=0.001,...
RuntimeError: The detected CUDA version (12.1) mismatches the version that was used to compile PyTorch (11.6). Please make sure to use the same CUDA versions.
fix transformer
PartialLabelingCSL is an useful work. Why not provide the dataset file to loading openImageV6?