oceanogeology
oceanogeology
请问为啥子不放ctw2015数据集和指标呢？可以把ctw2015的数据放上来吗？
在运行python prepare.py --action sfm --database_name custom/mouse --colmap 的时候，前面的几步骤（feature_extrator exhaustive matcher mapper image undistorter）都可以正常跑完，但是跑到patch_match_stereo的时候就出错了。感觉是调用gpu的时候报错了，但不知道怎么解决。 
测试数据如何与指定motion对齐
执行单卡训练的时候报错 booster.backward(loss=loss, optimizer=optimizer) File "/root/anaconda3/envs/opensora/lib/python3.10/site-packages/colossalai/booster/booster.py", line 167, in backward optimizer.backward(loss) File "/root/anaconda3/envs/opensora/lib/python3.10/site-packages/colossalai/zero/low_level/low_level_optim.py", line 487, in backward loss.backward(retain_graph=retain_graph) File "/root/anaconda3/envs/opensora/lib/python3.10/site-packages/torch/_tensor.py", line 522, in backward torch.autograd.backward( File "/root/anaconda3/envs/opensora/lib/python3.10/site-packages/torch/autograd/__init__.py", line 266, in backward...
trouble occur when reading some website address as follow, but same address is ok by skimage.io import imread: File "/usr/local/lib/python3.8/site-packages/imageio/v2.py", line 359, in imread with imopen(uri, "ri", **imopen_args) as file:...
 loss曲线如上图，基于预训练模型，在512x512的2w左右2s的视频切分上，训练130k，loss曲线没怎么下降，后面的视频生成效果和原始训练视频差距很大，请教有哪些优化手段。
https://github.com/hpcaitech/Open-Sora/blob/ea41df3d6cc5f389b6824572854d97fa9f7779c3/opensora/datasets/sampler.py#L56 this "parallel_apply" get same sample order in each epoch, but use "apply" get random sampler order. use parallel_apply: [1, 3, 10, 11, 14, 16, 17, 30, 31, 35, 37,...
Here's the MultiHeadCrossAttention code: ```python q = self.q_linear(x).view(1, -1, self.num_heads, self.head_dim) kv = self.kv_linear(cond).view(1, -1, 2, self.num_heads, self.head_dim) k, v = kv.unbind(2) attn_bias = None if mask is not None:...